Biocontrol Science and Technology, Vol. 19, S1, 2009, 3Á22

REVIEW ARTICLE Improving the cost-effectiveness, trade and safety of biological control for agricultural pests using nuclear techniques Jorge Hendrichsa*, Kenneth Bloemb, Gernot Hochc, James E. Carpenterd, Patrick Greanye, and Alan S. Robinsona

aJoint FAO/IAEA Programme of Nuclear Techniques in Food and Agriculture, International Atomic Energy Agency, Wagramerstrasse 5, A-1400 Vienna, Austria; bCenter for Plant Health Science & Technology (CPHIST), USDA-APHIS-PPQ, 1730 Varsity Drive, Suite 400, Raleigh, NC 27606, USA; cDepartment of Forest and Soil Sciences, BOKU Á University of Natural Resources and Applied Life Sciences, Vienna Hasenauerstrasse 38, A-1190 Vienna, Austria; dUSDA-ARS Crop Protection and Management Research Unit, Tifton, GA 31793, USA; e2770 Pine Ridge Road, Tallahassee, FL 32308, USA

If appropriately applied, biological control offers one of the most promising, environmentally sound, and sustainable control tactics for pests and weeds for application as part of an integrated pest management (IPM) approach. Public support for biological control as one of the preferred methods of managing non-indigenous and indigenous pests is increasing in many countries. An FAO/ IAEA Coordinated Research Project (CRP) addressed constraints related to costly production systems for biological control agents, and the presence of accompanying pest organisms during their shipment. These constraints can be alleviated using nuclear techniques such as ionizing radiation or X-rays to reduce production and handling costs (e.g., by expanding the period of host suitability, increasing shelf life, avoiding unnecessary sorting steps before shipment, etc.), and to eliminate the risk of shipping fertile host or prey pest individuals or other hitchhiking pests. These nuclear techniques can also help to reduce the risks associated with the introduction of exotic biological control agents, which can become pests of non-target organisms if not carefully screened under semi-natural or natural conditions. Radiation is also a very useful tool to study host- physiological interactions, such as host immune responses, by suppressing

Downloaded By: [Hendrichs, Jorge] At: 16:18 4 November 2009 defensive reactions of natural or factitious hosts. Applied at a very low-dose, radiation may be used to stimulate reproduction of some entomophagous . Additionally, radiation can be applied to semi- or completely sterilize hosts or prey for deployment in the field to increase the initial survival and build-up of natural or released biological control agents in advance of seasonal pest population build-up. Finally, the work carried out under this CRP has demonstrated the feasibility of integrating augmentative and sterile insect releases in area-wide IPM programmes, and to utilise by-products from insect mass- rearing facilities in augmentative biological control programmes. This special issue provides an overview of the research results of the CRP. Keywords: biological control; radiation; biocontrol agents; weeds; irradiated host; prey; insects; nematodes; ; sterile insects; sterile insect technique; inherited sterility; F1 sterility

*Corresponding author. Email: [email protected]

First Published Online 18 June 2009 ISSN 0958-3157 print/ISSN 1360-0478 online # 2009 Taylor & Francis DOI: 10.1080/09583150902985620 http://www.informaworld.com 4 J. Hendrichs et al.

Introduction Many pests, including and weeds, adversely affect agricultural produc- tion, and pre- and post-harvest losses of the order of 30Á40% are common (Yudelman, Ratta, and Nygaard 1998). Management of these pests still relies heavily on the use of pesticides with their associated limitations. For these reasons there is pressure to develop improved methods of pest control, with an emphasis on biologically and ecologically based tactics that can be applied as part of an area-wide integrated pest management (AW-IPM) approach (Vreysen, Robinson, and Hen- drichs 2007). If appropriately applied, biological control offers one of the most promising, environmentally sound, and sustainable tools for control of arthropod pests and weeds (van Lenteren, Bale, Bigler, Hokkanen, and Loomans 2006; van Driesche, Hoddle, and Center 2008). Public support for biological control as one of the preferred methods of managing non-indigenous and indigenous pests is increasing in many countries. There appear to be significant opportunities for increasing the use and cost-effectiveness of the application of classical and augmentative biological control through nuclear techniques for the production, shipping and release of biological control agents. Nuclear techniques are already applied in certain areas of entomology (Bakri, Heather, Hendrichs, and Ferris 2005a) and include the use of radiation sources for (1) studying sperm precedence, parasitoidÁhost interaction studies, etc., (2) post-harvest disinfestation for quarantine or phytosanitary security in support of agricultural international trade (IDIDAS 2004), and (3) insect sterilization as part of the application of the Sterile Insect Technique (SIT) (Dyck, Hendrichs, and Robinson 2005), where exposure to carefully selected irradiation doses of gamma or X-rays maximizes the induction of dominant lethal mutations in germ cells of pest insects, while minimizing other physiological changes (Bakri, Mehta, and Lance 2005b). Nuclear techniques in a wider sense also include the use of stable isotopes to study insect biology, behaviour, and physiology (IAEA 2009), although their use in biological control will not be considered here. In classical biological control, non-indigenous biological control agents, usually selected from the suite of parasitoids, predators and diseases that co-evolved with the pest, are introduced into the target area. One of the key concerns in this approach is Downloaded By: [Hendrichs, Jorge] At: 16:18 4 November 2009 the host specificity and host range of the introduced biological control agents. There are numerous examples of situations where introduced biological control agents have ‘jumped hosts’ (Follet and Duan 2000; Lockwood, Purcell, and Howarth 2001; Wajnberg, Scott and Quimby 2001), prompting some serious criticisms of this control method (Hamilton 2000; Louda, Pemberton, Johnson, and Follett 2003). In view of the growing awareness and concern, countries and their respective national plant protection organizations are increasingly implementing stringent environmen- tal risk assessment methods in order to screen potential biological control agents before release (van Lenteren et al. 2006). In cases where doubts remain about very promising natural enemies of weeds or insect pests, the release of such biological control agents that have been radiation-sterilized would enable a more definite and safe assessment of host specificity under natural conditions without any risk of permanent establishment. Inundative biological control involves the use of indigenous or non-indigenous biological control agents against indigenous or non-indigenous pests. These control Biocontrol Science and Technology 5

agents generally do not establish permanently, often because of adverse seasonal eco- climatic conditions in the area of introduction, and are mass-reared and released in very large numbers, often several times each season. Though the commercial biological control agent industry is growing, it still only represents a niche market with less than 3% of pest control sales (Cornell 2007). Regulatory, technical and other constraints on biological control have kept the market share relatively small. Challenges facing augmentative biological control include the high cost of production of biological control agents, adequate quality control and assurance, trade barriers and regulations that complicate shipping. Lack of enabling regulations probably has been amongst the most important barriers to the wider implementation of biological control globally. Appropriate regulations are needed that facilitate the importation and use of natural enemies. In many countries, ‘gatekeeper’ regulations place barriers in the way of efficient introduction of agents. However, the Secretariat of the International Plant Protection Convention, at the Food and Agriculture Organization of the United Nations (FAO) has developed and recently revised the International Standard for Phytosanitary Measures on ‘Guidelines for the export, shipment, import, and release of biological control agents and other beneficial organisms’ (ISPM No. 3) (FAO 2005), which should help to solve some of these problems and therefore increase cross-boundary trade in biological control agents. There are several ways in which nuclear techniques can improve the efficiency of augmentative biological control and the Joint FAO/IAEA Programme of Nuclear Techniques in Food and Agriculture initiated a 6-year Coordinated Research Project (CRP) entitled ‘Evaluating the Use of Nuclear Techniques for the Colonization and Production of Natural Enemies of Agricultural Insect Pests’ (Greany and Carpenter 1999) to address some of these aspects. For example, the cost of production may be decreased by simplifying the rearing process, increasing host suitability and shelf life, improving diets and dealing with disease and contamination. Possible trade barriers related to shipment of biological control agents include the accidental inclusion of fertile pest individuals or other hitchhikers, or the deliberate inclusion of live fertile food (the prey or pest insect for entomophagous agents) in the shipments, problems which can be overcome by irradiation. Properly timed distribution of the appropriate Downloaded By: [Hendrichs, Jorge] At: 16:18 4 November 2009 stage of a biological control agent is another crucial component for success. Providing irradiated hosts/prey in the field at the time of introduction would provide greater flexibility in timing the release and allow for development of the biological control agent population so that the most effective stage is present for optimum pest suppression. This introductory paper presents an overview of the research carried out under this CRP. Some of the results are described in research papers collected in this special issue, while some others are already published elsewhere.

The FAO/IAEA Coordinated Research Project (CRP) This international research network consisted of 18 research teams from 15 countries. The general objective of the CRP was to assess potential applications of nuclear techniques to increase the cost-effectiveness, trade and safety in the use of biological control agents of agricultural insect pests. It focused on the six major areas 6 J. Hendrichs et al.

listed in Table 1. The main research results in these major areas are described below, and a summary of the main findings is presented in Table 2.

Improving rearing Suppressing host immune reactions Host immune reactions may reduce rearing efficiency of biological control agents or prevent the use of factitious or non-habitual hosts that are easier or more economical to mass-rear. Exposure to radiation has been shown to suppress host immune system responses (Vey and Causse 1979) and it can also make older of irradiated larvae suitable for parasitoid development and thus increase rearing efficiency and parasitoid quality. Irradiated haemolymph of the greater wax Galleria mellonella (L.) showed severely reduced capability to encapsulate artificially introduced Sephadex beads (both in vivo and in vitro) probably due to radiation- damage to haemocytes. G. mellonella larvae irradiated with 65 Gy were found to be suitable for parasitization by Venturia canescens (Gravenhorst), thus facilitating the use of G. mellonella as a potential factitious host for the rearing of this biological control agent (Genchev, Milcheva-Dimitrova, and Kozhuharova 2007). The use of prey or hosts that are easier or more economical to mass-rear was also facilitated for the green lacewing predator Chrysoperla carnea (Stephens), where irradiated Sitotroga cerealella (Olivier) eggs, provided as a prey substitute, increased larval survival, fecundity and the proportion of female predators (Hamed, Nadeem, and Riaz 2009). Rearing of the parasitoid Psyttalia concolor (Sze´pligeti) on a factitious host, i.e., irradiated Mediterranean fruit Ceratitis capitata (Wiede- mann) larvae, allowed mass-releases of this parasitoid against the olive fruit fly Bactrocera oleae (Gmelin) (Hepdurgun, Turanli, and Zu¨mreog˘lu 2009a). Behavioural and physiological interactions between hosts and parasitoids are complex, often difficult to study, and not well understood in terms of improving rearing efficiency. Certain physiological processes in the host (e.g., defence mechanisms, hormone metabolism) can be selectively modified by radiation, thereby facilitating the study of particular hostÁparasitoid interactions. Radiation can likewise be used to modify or terminate certain parasitoid processes that affect Downloaded By: [Hendrichs, Jorge] At: 16:18 4 November 2009 host physiology and behaviour, e.g., by sterilizing the or parasitoid eggs (Bai, Chen, Cheng, Fu, and He 2003). Irradiation of Glyptapanteles liparidis (Bouche´) wasps caused temporary sterilization and a reduction in oviposition and reduced the total number of eggs laid per female, but did not reduce longevity (Tillinger, Hoch, and Schopf 2004). These wasps were used to study the action of the parasitoid’s polydnavirus and venom in gypsy moth (L.) larvae. During oviposition the irradiated wasps injected non-viable eggs, together with polydnavirus and venom in normal amounts and with minimal traumatic impact on the host (Hoch, Marktl, and Schopf 2009a). This ‘pseudoparasitization’ of L. dispar larvae by irradiated female wasps caused delayed larval development, morphological abnorm- alities and high mortality during pupation or in the pupal stage. The immune response (haemocytic encapsulation and haemolymph melanization) of the host was also suppressed by pseudoparasitization (Tillinger et al. 2004; Hoch et al. 2009a), as was juvenile hormone esterase activity (Schafellner, Marktl, and Schopf 2007). Moreover, pseudoparasitization of L. dispar host larvae by irradiated Biocontrol Science and Technology 7

Table 1. Potential applications of nuclear techniques assessed to increase the cost-effective- ness, trade and safety in the use of biological control agents focused on six major areas.

Areas of assessment Description of specific potential nuclear applications

1. Improving rearing (a) Understanding host-parasitoid physiological interactions to be able to modulate defensive reactions of hosts/prey; (b) expanding the time window suitable for host parasitisation; (c) allowing for increased storage and stockpiling of hosts or prey; (d) using very low-dose radiation to affect the sex ratio or to stimulate reproduction by entomophagous insects; (e) improving rearing media (either natural hosts/prey or artificial diets) through irradiation to reduce the microbial load and also to allow terminal sterilization after packaging; and (f) utilising by-products of mass rearing facilities producing sterile insects for simultaneous production of biological control agents. 2. Facilitating handling, shipment, (a) Eliminating the cost and logistics of holding and trade and release separating of parasitoids and non-parasitized pest adults before being able to ship to customers; (b) avoiding the emergence of pest adults from non- parasitized immature stages; (c) ameliorating concerns relating to the incidental presence in commercial shipments of fertile individuals of other hitchhiking pests; and (d) provisioning of sterilized natural prey to be used as food during commercial shipments of predators. 3. Supplementing hosts in the field (a) Provisioning of sub-sterile or sterile hosts or prey in for survival or early build-up the field as supplemental food to increase the initial of biological control agents survival of inoculatively released biological control agents; and

Downloaded By: [Hendrichs, Jorge] At: 16:18 4 November 2009 (b) provisioning of sub-sterile or sterile hosts or prey in the field as supplemental food to increase the early build- up of native biological control agents in advance of pest population build-up.

4. Integrating SIT or F1 sterility (a) Integrating natural enemies with the Sterile Insect and biological control Technique (SIT) or inherited sterility results in synergistic action, particularly with biological control agents repro- ducing/feeding on sterile or substerile offspring; and (b) applying the SIT against biocontrol agents that have become pest insects. 5. Reproductively inactivated hosts (a) Exploring for, and collection of, new biological control as sentinels in the field agents; and (b) monitoring of populations of parasitoids, predators and microorganisms. 8 J. Hendrichs et al.

Table 1. (Continued). Areas of assessment Description of specific potential nuclear applications

6. Screening classical biological Facilitating the importation of exotic for classical control agents under field biological control, through sterilisation (particularly conditions insect herbivores of weeds), where host specificity doubts remain, so that they can be released and assessed in the field without the risk of establishing breeding populations.

G. liparidis increased spore production of several species of entomopathogenic microsporidia co-infecting the hosts probably due to immune suppression as well as modified nutritional physiology (Hoch, Solter, and Schopf 2009b).

Expanding the time window when a host is suitable for parasitization Normal host development limits the time window when a host is suitable for parasitization and it is known that radiation can delay normal insect development and thus may extend the time window for host parasitization or modify the internal host environment to the benefit of the biological control agent. This was assessed for parasitoids of the Mediterranean flour moth Ephestia kuehniella (Zeller), the house fly Musca domestica (L.), the Indian meal moth Plodia interpunctella (Hu¨bner) and S. cerealella (Fatima, Ahmad, Memon, Bux, and Ahmad 2009; Hamed et al. 2009; Zapater, Andiarena, Pe´rez-Camargo and Bartoloni 2009). Also, in sugarcane stemborer Chilo infuscatellus (Snellen) larvae, a dose of 60Á80 Gy allowed the normally unsuitable fourth and fifth larvae to be successfully parasitised by flavipes Cameron (Fatima et al. 2009). Furthermore, irradiation of carambola fruit fly Bactrocera carambolae Drew & Hancock eggs with 30Á50 Gy extended the larval period available for parasitization by Psyttalia incisi (Sylvestri) and Fopius vandenboschi (Fullaway) (Kuswadi, Himawan, Indar- watmi, and Nasution 2003); and irradiation of third instar Anastrepha spp. larvae with a dose of 45 Gy extended the parasitization period and increased the quantity and quality of the parasitoid Diachasmimorpha longicaudata (Ashmead) developing Downloaded By: [Hendrichs, Jorge] At: 16:18 4 November 2009 in these hosts (Cancino, Ru´ız, Lo´pez, and Sivinski 2009b). Exposing irradiated hosts sequentially to larval and pupal parasitoids was another way that was explored to expand the suitability window for parasitization in order to maximize the use of hosts in the mass production of fruit fly parasitoids (Cancino, Ru´ız, Hendrichs, and Bloem 2009a).

Allowing for storage and stockpiling of hosts or prey The limited shelf life of hosts and prey for biological control agents restricts their use during mass-production. For certain insect species, radiation can be used to arrest development and thus allow for storage and stockpiling of hosts or prey. Studies on the Mediterranean flour moth E. kuehniella, M. domestica, S. cerealella, and the cotton leafworm or tobacco cutworm litura (F.) showed that irradiation caused a prolongation in the development of host stages suitable for parasitization, thus facilitating the use of these hosts under mass-rearing conditions (Celmer-Warda Biocontrol Science and Technology 9

Table 2. Listing of some of the studies of nuclear applications conducted in conjunction with the FAO/IAEA Coordinated Research Project to improve the cost-effectiveness, trade and safety of biological control of agricultural insect pests using nuclear techniques.

Constraints Biological control addressed Pest species agent References

Suppressing host immune reactions Galleria mellonella (L.) Venturia canescens Genchev et al. (2007) (Gravenhorst) Lymantria dispar (L.) Glyptapanteles Hoch et al. (2009a) liparidis (Bouche´) Lymantria dispar (L.) Microsporidia Hoch et al. (2009b) Ephestia kuehniella Venturia canescens Celmer-Warda (2004) (Zeller) (Gravenhorst) Chilo infuscatellus Cotesia flavipes Fatima et al. (2009) (Snellen) Cameron Sitotroga cerealella Chrysoperla carnea Hamed et al. (2009) (Olivier) (Stephens) Expanding the period of host suitability Anastrepha spp. Various parasitoids Cancino et al. (2009a,b) Plodia interpunctella Venturia canescens Celmer-Warda (2004) (Hu¨bner) (Gravenhorst) Sitotroga cerealella Trichogramma Fatima et al. (2009) (Olivier) chilonis Ishii Musca domestica (L.) Spalangia endius Zapater et al. (2009) Walker Bactrocera carambolae Fopius vandenboschi Kuswadi et al. (2003) Drew & Hancock (Fullaway) Psyttalia incisi (Sylvestri) sorbillans Nysolynx thymus Hasan et al. (2009) (Wiedemann) (Girault)

Downloaded By: [Hendrichs, Jorge] At: 16:18 4 November 2009 Extending storage and stockpiling time for hosts or prey (F.) Steinernema glaseri Seth et al. (2009) (Steiner) Ephestia kuehniella Trichogramma Tunc¸bilek et al. (2009b) (Zeller) evanescens (Westwood) Sitotroga cerealella Trichogramma Tunc¸bilek et al. (2009b) (Olivier) evanescens (Westwood) Musca domestica (L.) Spalangia endius Zapater et al. (2009) Walker Phthorimaea Trichograma spp. Saour (2009) operculella (Zeller) 10 J. Hendrichs et al.

Table 2. (Continued). Constraints Biological control addressed Pest species agent References

Stimulation effects of low dose radiation Ephestia kuehniella Trichogramma Genchev et al. (2008) (Zeller) evanescens (Westwood) Helicoverpa armigera Trichogramma chilonis Wang et al. (2009) (Hu¨bner) Ishii Utilisation of by-products from insect mass rearing facilities Ceratitis capitata Diachasmimorpha Viscarret et al. (2006) (Wiedemann) longicaudata (Ashmed) Ceratitis capitata Orius laevigatus Steinberg and Cayol (Wiedemann) (Fieber) (2009) Ceratitis capitata Spalangia cameroni Steinberg and Cayol (Wiedemann) Perkins (2009) Anastrepha spp. Various parasitoids Cancino et al. (2009b,c,d) Avoiding unnecessary handling and sorting steps before shipment Spodoptera litura (F.) Steinernema glaseri Seth and Barik (2009) (Steiner) Bactrocera oleae Psyttalia concolor Hepdurgun et al. (Gmelin) (Sze´pligeti) (2009a) Musca domestica (L.) Spalangia endius Zapater et al. (2009) Walker Ephestia kuehniella Trichogramma Tunc¸bilek et al. (2009a) (Zeller) evanescens (Westwood) Sitotroga cerealella Trichogramma Tunc¸bilek et al. (2009a) (Olivier) evanescens (Westwood) Downloaded By: [Hendrichs, Jorge] At: 16:18 4 November 2009 Anastrepha spp. Diachasmimorpha Cancino et al. (2009b) longicaudata (Ashmed) Ceratitis capitata Diachasmimorpha Cancino et al. (2009b) (Wiedemann) tryoni (Cameron) Fopius arisanus Cancino et al. (2009c) (Loew) (Sonan) Anastrepha ludens Various pupal Cancino et al. (2009d) (Loew) parasitoids Avoiding the shipment and release of fertile pest individuals Musca domestica (L.) Spalangia endius Zapater et al. (2009) Walker Tetranychus urticae Phytoseiulus persimilis Baptiste et al. (2003) Koch Athias-Henriot Biocontrol Science and Technology 11

Table 2. (Continued). Constraints Biological control addressed Pest species agent References

Anastrepha spp. Various parasitoids Cancino et al. (2009d) Bactrocera oleae Psyttalia concolor Hepdurgun et al. (Gmelin) (Sze´pligeti) (2009b) Shipping sterilized hosts or prey in the absence of biological control agents Musca domestica (L.) Á Zapater et al. (2009) Ceratitis capitata Á Steinberg and Cayol (Wiedemann) (2009) Synergising biological control agents and F1 sterility Helicoverpa armigera Trichogramma chilonis Wang et al. (2009) (Hu¨bner) Ishii Phthorimaea Trichogramma Saour (2009) operculella (Zeller) principium (Sugonyaev & Sorokina) Lymantria dispar (L.) Various Novotny and Zubrik (2003); Zubrik and Novotny (2009) Spodoptera litura (F.) Steinernema glaseri Seth et al. (2009) (Steiner) Thaumatotibia Trichogrammatoidea Carpenter et al. (2004) (Chryptophlebia) cryptophlebiae leucotreta (Meyrick) Nagaraja Liriomyza bryoniae Diglyphus isaea Steinberg and Cayol (Kaltenbach) (Walker) (2009) Exorista sorbillans Nysolynx thymus Hasan et al. (2009) (Wiedemann) (Girault) Downloaded By: [Hendrichs, Jorge] At: 16:18 4 November 2009 Using SIT against biological control agents that have become a pest Exorista sorbillans Á Hasan et al. (2009) (Wiedemann) Cactoblastis cactorum Á Hight et al. (2005) (Berg) Building up natural enemies in advance of pest populations Lymantria dispar (L.) Various Zubrik and Novotny (2009) Helicoverpa armigera Trichogramma chilonis Wang et al. (2009) (Hu¨bner) Ishii 12 J. Hendrichs et al.

Table 2. (Continued). Constraints Biological control addressed Pest species agent References

Chilo infuscatellus Trichogramma chilonis Fatima et al. (2009) (Snellen) Ishii Monitoring natural enemies in the field Lymantria dispar (L.) Various Novotny and Zubrik (2003); Zubrik and Novotny (2009) Ephestia kuehniella Venturia canescens Celmer-Warda (2004); (Zeller) (Gravenhorst) Celmer (2006) Plodia interpunctella Venturia canescens Celmer-Warda (2004) (Hu¨bner) (Gravenhorst) Ceratitis capitata Diachasmimorpha Jordao-Paranhos (Wiedemann) longicaudata (Ashmed) et al. (2003) Helicoverpa armigera Trichogramma chilonis Wang et al. (2009) (Hu¨bner) Ishii Screening classical biological control agents in the field Á Episimus unguiculus Moeri (2007); Moeri Clarke et al. (2009) Á Cactoblastis cactorum Tate et al. (2007, 2009) (Berg)

2004; Seth, Barik, and Chauhan 2009; Zapater et al. 2009). Host eggs of E. kuehniella irradiated at 200 Gy could be stored at 48C for up to 30 days without any quantitative or qualitative loss in the production of Trichogramma evanescens (Westwood) and for up to 60 days with only a minor decrease in quality (Tunc¸bilek, Canpolat, and Sumer 2009b). Parasitoids in diapause could be stored inside irradiated host eggs for a period of 50 days without adverse effect on emergence, and irradiation of eggs did not affect

Downloaded By: [Hendrichs, Jorge] At: 16:18 4 November 2009 acceptance by parasitoids (Tunc¸bilek et al. 2009b). The parasitoid C. flavipes irradiated at 20 Gy could be stored as pupae for 2 months at 108C without apparent loss of quality (Fatima et al. 2009).

Utilisation of by-products from insect mass-rearing facilities Insect mass-rearing facilities often have excess production of particular insect life stages, and they also generate batches of sub-standard insects that are normally discarded. In addition, they create significant amounts of by-products during the production process (large biomass of discarded adults, larvae and pupae from the various colonies and quality control testing). Instead of having to invest in their disposal, these by-products may be processed (Nakashima, Hirose, and Kinjo 1996) or irradiated, if still alive, to support the production of biological control agents. Examples include the use of excess eggs, as well as the use of remnant larvae and pupae to produce egg, larval and pupal parasitoids, respectively. It was shown that Biocontrol Science and Technology 13

irradiated excess eggs of C. capitata and the Mexican fruit fly Anastrepha ludens (Loew) produced in mass-rearing facilities could be used to produce egg parasitoids, and that the discarded larvae and pupae could be used to produce larval and pupal parasitoids (Cancino, Ru´ız, Pe´rez, and Harris 2009c; Cancino, Ru´ız, Sivinski, Ga´lvez, and Aluja 2009d). Another example is the commercial use of C. capitata eggs and pupae, respectively, for the production of the minute pirate bug Orius laevigatus (Fieber), a highly effective predator of Western flower thrips, Frankliniella occidentalis Pergande (Steinberg and Cayol 2009), and the parasitoid Spalangia cameroni Perkins (: Pteromalidae), a parasitoid of M. domestica and other Diptera (Geden 2007). The effective use of these by-products from insect rearing facilities can greatly increase the efficiency and the economics of the rearing process. Mass-rearing of C. capitata genetic sexing strains, which are now used in a majority of Mediterranean fruit fly production facilities (Ca´ceres et al. 2004), allows separating sexes at the larval or pupal stages with the possibility to employ a majority of males for sterile insect releases, while any excess females not returning to the colony can be used to produce larval and pupal parasitoids (Viscarret, La Rossa, Segura, Ovruski, and Cladera 2006).

Reproductive stimulation by use of very low dose radiation The controversial phenomenon known as ‘radiation hormesis’ (Luckey 1991) refers to the use of very low dose radiation to stimulate biological processes. Two of the studies related to this CRP (Genchev, Balevski, Obretenchev, and Obretencheva 2008; Wang, Lu, Liu and Li 2009) report noting a stimulation of reproduction and parasitization parameters in the parasitoids Habrobracon hebetor (Say), Tricho- gramma chilonis Ishii and V. canescens after exposure to very low doses of radiation, an intriguing discovery warranting further investigation.

Facilitating handling, shipment, trade and release Avoiding unnecessary handling and sorting steps before shipment Downloaded By: [Hendrichs, Jorge] At: 16:18 4 November 2009 The continued development and emergence of non-parasitised fertile hosts, as well as of unused prey (pest) insects during mass-production of biological control agents often requires additional handling steps. These procedures, involving separation of significant proportions of non-parasitised hosts (or unused prey individuals) from the biological control agents, decrease efficiency in large scale mass-rearing. During parasitoid mass-production, radiation was used to reproductively sterilize prey, hosts, and factitious hosts such as E. kuehniella, S. cerealella, P. interpunctella, C. infuscatellus, and C. capitata, thereby inhibiting their further development and preventing the need for costly separation procedures. This also removed the delays inherent to the traditional production processes, which are related to waiting for the emergence from pupae of fertile unused pest individuals (Cancino et al. 2009b; Fatima et al. 2009; Tunc¸bilek, Canpolat, and Ayvaz 2009a). In the case of fruit such as the West Indian fruit fly Anastrepha obliqua (Macquart), the sapote fruit fly Anastrepha serpentina (Wiedemann), and A. ludens, irradiation of larvae is used routinely in the mass-production of tens of millions of parasitoids of these pest fruit 14 J. Hendrichs et al.

flies (Cancino et al. 2009b). Irradiation of A. ludens eggs and the pupae of M. domestica and A. ludens avoided having to wait for adult emergence as well as unnecessary handling during the mass-rearing of egg and pupal parasitoids for these pests (Cancino et al. 2009c,d; Zapater et al. 2009).

Avoiding the shipment of fertile pest individuals There exists a real or perceived risk that shipping biological control agents with hosts/prey material could lead to the introduction of non-native, pesticide resistant or new strains of pest insects into new areas or countries and this may exacerbate the ever-stricter quarantine regulations required to obtain permits for their shipment. Procedures are required to guarantee that customers receive pest-free shipments. Research on the use of eggs of the spider mite Tetranychus urticae Koch to provision shipments of several species of predatory mites has confirmed that radiation at a dose of 280 Gy or less, depending on the age of the host eggs, can be used to eliminate the risk of introducing fertile mites, or other hitchhiking arthropods. At the same time, the provisioning of living eggs allows the inclusion of nutritional supplements in the form of prey material to maintain predator quality during shipment (Baptiste, Bloem, Reitz, and Mizell 2003). Irradiation of house fly pupae was shown to be very beneficial for the commercial shipment of house fly pupal parasitoids, allowing early shipment of recently parasitised pupae while ensuring that clean shipments were not contami- nated with unparasitised pupae that would emerge later with the customers (Zapater et al. 2009). In another example, fruit fly parasitoids have been sent from Mexico to South America after being reared on irradiated A. ludens, which is a quarantine pest in this region (J. Cancino, personal communication). Furthermore, the feasibility of inoculative and augmentative releases of entomopathogenic nematodes within sterilized hosts was proposed to establish a safe mode of transport and dispersion without concern for the inadvertent release of uninfected fertile hosts. Laboratory studies demonstrated that hosts irradiated at 40 Gy were suitable for nematode development (although nematode parasitisation efficacy was better in non-irradiated host larvae) and could thus be used for safe, inoculative releases of these biocontrol agents (Seth and Barik 2009). Downloaded By: [Hendrichs, Jorge] At: 16:18 4 November 2009

Shipping sterilized hosts or prey in the absence of biological control agents Needs and opportunities exist for some commercial biological control companies to ship mass produced sterile hosts/prey in the absence of natural enemies for redistribution, both within and between countries, for use as host/prey at smaller rearing facilities. This alternative can be implemented in order to gain efficiencies in the production of biological control agents or to standardize the use of strains of host/prey material to ensure product quality (Steinberg and Cayol 2009; Zapater et al. 2009).

Supplementing hosts in the field for survival or early build-up of biological control agents Many insect pests show cyclical population outbreaks that temporarily escape natural control. By increasing the number of host insects prior to the pest outbreak, Biocontrol Science and Technology 15

the population density of the biological control agents can be significantly increased. Radiation can be used to produce sterile host insects or host insects generating sterile F1 individuals to be released as hosts for the biological control agents without increasing the risk that the released host insects will become pests themselves. Irradiated eggs, as well as sterile F1 eggs and larvae resulting from irradiated parents of the gypsy moth, L. dispar, were distributed in a natural forest and found to be acceptable and suitable as hosts for a number of parasitoid species. Most importantly, the parasitoids did not differentiate, under these natural conditions, between sterile F1 larvae and untreated larvae (Novotny and Zubrik 2003; Zubrik and Novotny 2009). Similar results were obtained with irradiated cotton bollworm Helicoverpa armigera (Hu¨bner) and diamondback moth Plutella xylostella (L.) adults released in field crops during critical periods where their sterile eggs served as hosts for feral egg parasitoids resulting in parasitoid population increases (Wang et al. 2009). In crops with low damage thresholds, the early season use of trap crops in rows or in the surroundings may be required in some situations to minimize any potential damage by such semi-sterile hosts. In sugarcane fields, the provision of supplemental hosts (irradiated sterile host eggs) to T. chilonis early in the season allowed populations of parasitoids to build-up and enhanced their survival during critical periods thereafter. This approach is currently providing effective management of several species of sugarcane borers in a 40,000-ha area of sugarcane in Pakistan (Fatima et al. 2009).

Integrating SIT or F1 sterility and biological control The release of sterile or semi-sterile insects together with biological control agents has been known to have synergistic effects for population suppression when applied simultaneously (Knipling 1992; Wong, Ramadan, Herr and McInnis 1992; Bloem, Bloem, and Knight 1998). This synergy results from the sterile insects impacting on the adult stage, while the biological control agents target mostly the immature stages, including reproducing on the F1 offspring in inherited sterility releases. Laboratory and field trials with H. armigera, the corn earworm Helicoverpa zea (Boddie), L. dispar, the potato tuber moth Phthorimaea operculella (Zeller), P. xylostella, S. litura, and the beet armyworm Spodoptera exigua (Hu¨bner) indicated that sterile Downloaded By: [Hendrichs, Jorge] At: 16:18 4 November 2009 progeny from semi-sterile were acceptable as hosts for egg and larval parasitoids (Novotny and Zubrik 2003; Saour 2009; Wang et al. 2009; Zubrik and Novotny 2009). Experiments under large laboratory cage conditions showed that F1 sterility and releases of Trichogramma principium (Sugonyaev & Sorokina) are effective in suppressing Ph. operculella. Properly timed releases of T. principium together with moths irradiated at 250 Gy produced the greatest reduction in P. operculella F3 progeny, demonstrating the synergistic effects of combining F1 sterility with egg parasitoids (Saour 2009). Eggs of the false codling moth Thaumatotibia (Chryptophlebia) leucotreta (Meyrick) treated with 150Á200 Gy were acceptable and suitable for development of the egg parasitoid Trichogrammatoidea cryptophlebiae Nagaraja under laboratory conditions. Field-cage evaluations in orchards in South revealed that releases of irradiated (150 and 200 Gy) moths combined with releases of T. cryptophlebiae provided synergistic suppression of false codling moth populations 16 J. Hendrichs et al.

(Carpenter, Bloem, and Hofmeyr 2004). These findings have encouraged the establishment of a private company by the South African citrus export industry in the Western Cape Province for the area-wide suppression of false codling moth, a major polyphagous pest that has developed resistance against many insecticides. The compatibility of the application of entomopathogenic nematodes with F1 sterility for population suppression of S. litura was demonstrated in laboratory experiments. Various feasible modes of integration of these two bio-rational strategies have been proposed (Seth et al. 2009). Another system under development for integrating augmentative parasitoid releases with the SIT is the release of Diglyphus isaea (Walker), the parasitoid of celery miner fly Liriomyza bryoniae (Kaltenbach), a serious pest of vegetables and ornamentals, together with sterile males of L. bryoniae for application in greenhouses (Kaspi and Parella 2008; Steinberg and Cayol 2009). Developing the SIT against biocontrol agents that have become pest insects themselves is another application linking nuclear techniques with biological control agents. One case is the cactus moth Cactoblastis cactorum (Berg), a textbook example of very effective classical biological control of introduced cactus Opuntia spp., which has invaded the south-eastern USA, and where its westward expansion is being contained by the integrated application of SIT (Hight, Carpenter, Bloem, and Bloem 2005; Bloem et al. 2007; Tate, Carpenter, and Bloem 2007) to protect native Opuntia- based ecosystems in the south-western USA and Mexico. The Uzi-fly Exorista sorbillans (Wiedemann) is an endoparasitoid of the silkworm (L.) and as such is a pest due to its negative impact on silk production. Radiation studies have been carried out to assess if the SIT can contribute to an integrated control of this pest. In addition, Nesolynx thymus (Girault), a hyperpar- asitoid of the Uzi-fly, has been identified and rearing studies have been carried out on irradiated Uzi-fly pupae to assess if combined releases of sterile insects and parasitoids are feasible (Hasan, Uddin, Khan, and Reza 2009). Mass releases of the olive fruit fly parasitoid P. concolor, reared on irradiated Mediterranean fruit fly larvae, were integrated with mass trapping for an envir- onmentally friendly suppression method of olive fly populations (Hepdurgun, Turanli, and Zu¨mreog˘lu 2009b). Downloaded By: [Hendrichs, Jorge] At: 16:18 4 November 2009 Reproductively inactivating hosts as sentinels in the field The exploration for, and collection of new exotic biological control agents and the monitoring of field populations of native biological control agents are sometimes complicated by the fact that hosts are rare or difficult to locate. Reproductively sterilized host insects may be placed in the field in strategic locations as sentinels to aid in these efforts (Jordao-Paranhos, Walder, and Papadopoulos 2003). Several approaches were evaluated including the use of (1) irradiated eggs of S. cerealella, a factitious host of Trichogramma, to monitor effects of seasonal environmental conditions on the establishment of released Trichogramma in sugarcane fields (Fatima et al. 2009), (2) sterile F1 larvae from irradiated L. dispar for monitoring the density and type of parasitoids and pathogens in forests (Novotny and Zubrik 2003; Zubrik and Novotny 2009), (3) reproductively inactivated larvae (400 and 600 Gy) of E. kuehniella and P. interpunctella to monitor the density of V. canescens and Habrobracon hebetor (Say) in warehouses and mills (Celmer-Warda Biocontrol Science and Technology 17

2004; Celmer 2006), and (4) sterilized M. domestica pupae in traps to monitor wild populations of pteromalid parasitoids in the field and under conditions of livestock production (Zapater, personal communication).

Screening classical biological control agents under field conditions Classical biological control has resulted in many significant successes, but also many cases of direct and indirect non-target impacts have been documented (Howarth 1991; Thomas and Willis 1998; Henneman and Memott 2001). Also, inundative biological control can result in environmental problems (van Lenteren et al. 2003), which has fostered growing concerns about the need to preserve and natural ecosystems. Therefore, the importation of exotic biological control agents, particularly insect herbivores of invasive plants, is becoming increasingly difficult due to concerns over the possibility that imported species may shift hosts and become pests of crops or protected species. In some cases, despite careful selection (Briese 2006; van Lenteren et al. 2003; van Lenteren et al. 2006) and extensive pre- release studies under quarantine conditions, the release of promising biological control agents is ultimately rejected because of remaining doubts about their host specificity. In such situations, exotic biological control agents may be sterilized using radiation so that they can be released and studied under actual field conditions without the risk of establishing permanent breeding populations in space and time. The use of sterilized individuals allows further assessment and confirmation of oviposition behaviour and host (acceptability) associations. Also, the use of F1 sterile larvae of exotic herbivores being considered for introduction and release against plant pests would allow field-testing of larval feeding preferences and the ability of these larvae to develop and survive on related weeds, crops and other native plants that are of concern (Greany and Carpenter 1999; Carpenter, Bloem, and Bloem 2001). A model system that includes Opuntia spp. and C. cactorum has been developed to study the host range of an exotic herbivore. Radiation biology studies revealed that the optimum dose at which females are sterilized and males remain partially fertile and produce sterile progeny is 200 Gy (Carpenter et al. 2001). Whole plant Downloaded By: [Hendrichs, Jorge] At: 16:18 4 November 2009 and single cladode host preference tests demonstrated that C. cactorum females mated with males irradiated at 200 Gy exhibited normal oviposition preferences and can be used safely under field conditions to predict the host range, as well as to study possible interactions with natural enemies (Hight et al. 2005; Tate, Hight, and Carpenter 2009). Another system under evaluation involves the exotic herbivore Episimus unguiculus Clarke (E. utilis Zimmerman), which is currently in quarantine in Florida, for the eventual biological control of the Brazilian pepper tree Schinus terebinthifolius Raddi (Moeri 2007; Moeri, Cuda, Overholt, Bloem, and Carpenter 2009).

Conclusion The results generated by this CRP have confirmed that nuclear techniques have a significant role to play in facilitating the use and increasing the cost-effectiveness and safety of biological control agents (Table 2). Nevertheless, a major constraint faced 18 J. Hendrichs et al.

by practitioners and producers of biological control agents who would like to adopt some of these technologies is access to radiation sources as these sources are only available to the larger commercial producers of biological control agents, some research institutions, or programmes involved with the production of sterile insects for release. Radiation sources represent a major financial investment for any company and bring with them considerable logistic and regulatory constraints. The final paper in this issue provides a review of the smaller radiation sources available as well as a comparison of some key parameters (Mehta 2009). In the future, in view of the increasing difficulty of transporting radioactive materials (IAEA 2008), non-isotopic sources such as those emitting X-rays may become the equipment of choice.

Acknowledgements We would like express our appreciation to Jean Pierre Cayol and Marc Vreysen for useful suggestions and comments to improve this manuscript, and also to thank all the participants of the CRP for their enthusiastic and productive participation in this coordinated research network.

References Bai, S.F., Chen, X.X., Cheng, J.A., Fu, W.J., and He, J.H. (2003), ‘Characterization of Cotesia plutella Polydnavirus and its Physiological Effects on the Diamondback Moth, Plutella xylostella Larvae’, Acta Entomologica Sinica, 46, 401Á408. Bakri, A., Heather, N., Hendrichs, J., and Ferris, I. (2005a), ‘Fifty Years of Radiation Biology in Entomology: Lessons Learned from IDIDAS’, Annals of the Entomological Society of America, 98, 1Á18. Bakri, A., Mehta, K., and Lance, D.R. (2005b), ‘Sterilizing Insects with Ionizing Radiation’,inSterile Insect Technique, Principles and Practice in Area-wide Integrated Pest Management, eds. V.A. Dyck, J. Hendrichs, and A.S. Robinson, Dordrecht, The Netherlands: Springer, pp. 233Á268. Baptiste, S.J., Bloem, K., Reitz, S., and Mizell III, R. (2003), ‘Use of Radiation to Sterilize Two-spotted Spider Mite (Acari: Tetranychidae) Eggs Used as a Food Source for Predatory Mites’, Florida Entomologist, 86, 389Á394. Bloem, S., Bloem, K., and Knight, A.L. (1998), ‘Oviposition by Sterile Codling Moths, Cydia pomonella (Lepidoptera: Tortricidae) and Control of Wild Populations with Combined Downloaded By: [Hendrichs, Jorge] At: 16:18 4 November 2009 Releases of Sterile Moths and Egg Parasitoids’, Journal Entomological Society of British Columbia, 95, 99Á109. Bloem, K.A., Bloem, S., Carpenter, J.E., Hight, S., Floyd, J., and Zimmermann, H. (2007), ‘Don’t Let Cacto Blast Us: Development of a Bi-national Plan to Stop the Spread of the Cactus Moth Cactoblastis cactorum in North America’,inArea-wide Control of Insect Pests. From Research to Field Implementation, eds. M.J.B. Vreysen, A.S. Robinson, and J. Hendrichs, Dordrecht, The Netherlands: Springer, pp. 337Á344. Briese, D.T. (2006), ‘Can an A Priori Strategy be Developed for Biological Control? The Case of Onopordum spp. Thistles in , Special Issue: Agent Selection in Weed Biocontrol’, Australian Journal of Entomology, 45, 317Á323. Ca´ceres, C., Cayol, J.P., Enkerlin, W., Franz, G., Hendrichs, J., and Robinson, A.S. (2004), ‘Comparison of Mediterranean fruit fly(Ceratitis capitata: Tephritidae) Bisexual and Genetic Sexing Strains: Development, Evaluation and Economics’,inProceedings Sixth Intl. Symposium on Fruit Flies of Economic Importance, ed. B. Barnes, Irene, South Africa: Isteg Scientific Publications, pp. 367Á381. Cancino, J., Ru´ız, L., Hendrichs, J., and Bloem, K. (2009a), ‘Evaluation of Sequential Exposure of Irradiated Hosts to Maximize the Mass Rearing of Fruit Fly Parasitoids’, Biocontrol Science and Technology, this volume. Biocontrol Science and Technology 19

Cancino, J., Ru´ız, L., Lo´pez, P., and Sivinski, J. (2009b), ‘The Suitability of Anastrepha spp. and Ceratitis capitata (Diptera: Tephritidae) Larvae as Hosts of Diachasmimorpha longicaudata and Diachasmimorpha tryoni (Hymenoptera: ): Effects of Host Age and Radiation Dose and Implications for Quality Control in Mass Rearing’, Biocontrol Science and Technology, this volume. Cancino, J., Ru´ız, L., Pe´rez, J., and Harris, E. (2009c), ‘Irradiation of Anastrepha ludens (Diptera: Tephritidae) Eggs for the Rearing of the Fruit Fly Parasitoids Fopius arisanus and Diachasmimorpha longicaudata (Hymenoptera: Braconidae)’, Biocontrol Science and Technology, this volume. Cancino, J., Ru´ız, L., Sivinski, J., Ga´lvez, F.O., and Aluja, M. (2009d), ‘Rearing of Five Hymenopterous Larval-prepupal (Braconidae, Figitidae) and Three Pupal (Diapriidae, Chalcidoidea, Eurytomidae) Native Parasitoids of the Anastrepha (Diptera: Tephritidae) on Irradiated A. ludens Larvae and Pupae’, Biocontrol Science and Technology, this volume. Carpenter, J.E., Bloem, S., and Bloem, K.A. (2001), ‘Inherited Sterility in Cactoblastis cactorum (Lepidoptera: Pyralidae)’, Florida Entomologist, 84, 537Á542. Carpenter, J.E., Bloem, S., and Hofmeyr, S. (2004), ‘Acceptability and Suitability of Eggs of False Codling Moth (Lepidoptera: Tortricidae) from Irradiated Parents to Parasitism by Trichogrammatoidea cryptophlebiae (Hymenoptera: Trichogrammatidae)’, Biological Con- trol, 30, 351Á359. Celmer-Warda, K. (2004), ‘Preliminary Studies on Suitability and Acceptability of Irradiated Host Larvae (Plodia interpunctella) by Larval Parasitoids Venturia canescens (Graven- horst)’, Annals of Warsaw Agricultural University-SGGW, Horticulture, Landscape Archi- tecture, 25, 67Á73. Celmer, K. (2006), ‘Use of Gamma Radiation to Develop a Mass Rearing Method of the Parasitoid Venturia canescens (Gravenhorst)’, PhD. Thesis, University of Warsaw, Poland. Cornell, J.D. (2007), ‘Biocontrol: Limits to Use’,inEncyclopedia of Pest Management Volume II, ed. D. Pimentel, Boca Raton, FL, USA: CRC Press, pp. 33Á36. Dyck, V.A., Hendrichs, J., and Robinson, A.S. (2005), Sterile Insect Technique, Principles and Practice in Area-wide Integrated Pest Management, Dordrecht, The Netherlands: Springer, 787 pp. FAO (Food and Agriculture Organization of the United Nations) (2005), ‘International Standard for Phytosanitary Measures No. 3. Guidelines for the Export, Shipment, Import, and Release of Biological Control Agents and other Beneficial Organisms’, Secretariat of the International Plant Protection Convention, FAO, Rome, Italy. Fatima, B., Ahmad, N., Memon, R.M., Bux, M., and Ahmad, Q. (2009), ‘Enhancing Biological Control of Sugarcane Shoot Borer, Chilo infuscatellus (Lepidoptera: Pyralidae), through Use of Radiation to Improve Laboratory Rearing and Field Augmentation of Egg and Larval Parasitoids’, Biocontrol Science and Technology, this volume. Downloaded By: [Hendrichs, Jorge] At: 16:18 4 November 2009 Follet, P.A., and Duan, J.J. (2000), Non-Target Effects of Biological Control, Dordrecht, The Netherlands: Kluwer Academic Publishers, 316 pp. Geden, C.J. (2007), ‘Development of Spalangia cameroni and Muscidifurax raptor on Live House Fly Pupae and Pupae Killed by Heat Shock, Irradiation and Cold’, Environmental Entomology, 36, 34Á39. Genchev, N.P., Milcheva-Dimitrova, R.Y., and Kozhuharova, M.V. (2007), ‘Use of Gamma Radiation for Suppression of the Hemocytic Immune Response in Larvae of Galleria mellonella (Lepidoptera) against Venturia canescens (Hymenoptera)’, Journal of Balkan Ecology, 10, 411Á419. Genchev, N.P., Balevski, N., Obretenchev, D.A, and Obretencheva, A.D. (2008), ‘Stimulation Effects of Low Doses of Gamma Radiation on Adults Habrobracon hebetor Say (Hymenoptera: Braconidae)’, Journal of Balkan Ecology, 11, 99Á102. Greany, P.D, and Carpenter, J.E. (1999), ‘Use of Nuclear Techniques in Biological Control of Insects and Weeds’, Nuclear News, 42, 32Á34. Hamed, M., Nadeem, S., and Riaz, A. (2009), ‘Use of Gamma Radiation for Improving the Mass Production of Trichogramma chilonis and Chrysoperla carnea’, Biocontrol Science and Technology, this volume. Hamilton, G. (2000), ‘When Good Bugs Turn Bad’, New Scientist, 15 January pp. 30Á33. 20 J. Hendrichs et al.

Hasan, M.M., Uddin, M.R., Khan, M.A.R., and Reza, A.M.S. (2009), ‘Effects of Host Density, Host Age, Temperature and Gamma Irradiation on the Mass Production of Nesolynx thymus (Hymenoptera: ), an Endoparasitoid of Uzi Fly, Exorista sorbillans (Diptera: )’, Biocontrol Science and Technology, this volume. Henneman, M.L., and Memmott, J. (2001), ‘Infiltration of a Hawaiian Community by Introduced Biological Control Agents’, Science, 293, 1314Á1316. Hepdurgun, B., Turanli, T., and Zu¨mreog˘lu, A. (2009a), ‘Parasitism Rate and Sex Ratio of Psyttalia (Opius) concolor Szepl. (Hymenoptera: Brachonidae) Reared on Irradiated Ceratitis capitata (Wied.) Larvae (Diptera: Tephritidae)’, Biocontrol Science and Technology, this volume. Hepdurgun, B., Turanli, T., and Zu¨mreog˘lu, A. (2009b), ‘Control of the Olive Fruit Fly (Bactrocera oleae (Gmel.)) through Mass Trapping and Mass Releases of the Parasitoid (Psyttalia concolor Szepl.) Reared on Irradiated Mediterranean Fruit Flies’, Biocontrol Science and Technology, this volume. Hight, S.D., Carpenter, J.E., Bloem, S., and Bloem, K.A. (2005), ‘Developing a Sterile Insect Release Program for Cactoblastis cactorum Berg (Lepidoptera: Pyralidae): Effective Overflooding Ratios and Release-recapture Field Studies’, Environmental Entomology, 34, 850Á856. Hoch, G., Marktl, R.C., and Schopf, A. (2009a), ‘Gamma Radiation-induced Pseudopar- asitization as a Tool to Study Interactions between Host Insects and Parasitoids in the System Lymantria dispar (Lep., Lymantriidae)-Glyptapanteles liparidis (Hym., Braconidae)’, Biocontrol Science and Technology, this volume. Hoch, G., Solter, L.F., and Schopf, A. (2009b), ‘Treatment of Lymantria dispar (Lepidoptera, Lymantriidae) Host Larvae with Polydnavirus/Venom of a Braconid Parasitoid Increases Spore Production of Entomopathogenic Microsporidia’, Biocontrol Science and Technology, this volume. Howarth, F.G. (1991), ‘Environmental Impacts of Classical Biological Control’, Annual Review of Entomology, 36, 485Á509. (IAEA) International Atomic Energy Agency (2008), ‘Experts Tackle Shipment Issues for Beneficial Radiation Sources, Mediterranean Workshop Focuses on Radioactive Materials for Medicine, Industry’, Regional IAEA workshop held in Rome, Italy 14Á16 May 2008 http://www.iaea.org/NewsCenter/News/2008/shipmentissues.html. (IAEA) International Atomic Energy Agency (2009), Manual for the Use of Stable Isotopes in Entomology, Vienna, Austria: 75 pp. (IDIDAS) International Database on Insect Disinfestation and Sterilization (2009), http:// www-ididas.iaea.org/IDIDAS/default.htm. Jordao-Paranhos, B.A., Walder, J.M.M., and Papadopoulos, N.T. (2003), ‘A Simple Method to Study Parasitism and Field Biology of the Parasitoid Diachasmimorpha longicaudata

Downloaded By: [Hendrichs, Jorge] At: 16:18 4 November 2009 (Hymenoptera: Braconidae) on Ceratitis capitata (Diptera: Tephritidae)’, Biocontrol Science and Technology, 13, 631Á639. Kaspi, R., and Parrella, M.P. (2008), ‘Synergistic Interaction between Parasitoids and Sterile Insects. Integrated Control in Protected Crops. Temperate Climate’, IOBC/WPRS Bulletin, 32, 99Á102. Knipling, E.F. (1992), ‘Principles of Insect Parasitism Analyzed from New Perspectives: Practical Implications for Regulating Insect Populations by Biological Means’, USDA Agriculture Handbook No. 693, 337 pp. Kuswadi, A.N., Himawan, T., Indarwatmi, M., and Nasution, I.A. (2003), ‘The Use of Gamma Irradiation to Support the Colonization and Production of Natural Enemies of Bactrocera carambolae (Drew & Hancock)’, Report of 3rd Research Coordination Meeting, IAEA, Vienna, Austria. Lockwood, J.A., Purcell, M.F., and Howarth, F.G. (2001), ‘Balancing Nature: Assessing the Impact of Importing Non-Native Biological Control Agents (An International Perspec- tive)’, Entomological Society of America, USA, 130 pp. Louda, S.M., Pemberton, R.W., Johnson, M.T., and Follett, P.A. (2003), ‘Non-target Effects: the Achilles’ Heel of Biological Control?’, Annual Review of Entomology, 48, 365Á396. Luckey, T.D. (1991), Radiation Hormesis, Boca Raton, FL, USA: CRC Press. Biocontrol Science and Technology 21

Mehta, K. (2009), ‘Radiation Sources Supporting the Use of Natural Enemies for Biological Control of Agricultural Pests’, Biocontrol Science and Technology, this volume. Moeri, O. (2007), ‘Application of the F1 Sterile Insect Technique (F1 SIT) for Field Host Range Testing of Episimus utilis Zimmerman (Lepidoptera: Tortricidae), a Candidate Biological Control Agent of Brazilian Peppertree’, M.Sc. thesis, University of Florida, Department of Entomology and Nematology, Gainesville, FL, USA. Moeri, O., Cuda, J.P., Overholt, W.A., Bloem, S., and Carpenter, E. (2009), ‘F1 Sterile Insect Technique: A Novel Approach for Risk Assessment of Episimus unguiculus (Lepidoptera: Tortricidae), a Candidate Biological Control Agent of Schinus terebinthifolius in the continental USA’, Biocontrol Science and Technology, this volume. Nakashima, Y., Hirose, Y., and Kinjo, K. (1996), ‘Rearing Orius sauteri (Poppius) on Diet of Freeze-dried Larval Powder of Melon Fly, Bactrocera cucurbitae Coquillett’, Japanese Journal of Applied Entomology and Zoology, 40, 80Á82. Novotny, J., and Zubrik, M. (2003), ‘Sterile Insect Technique as a Tool for Increasing the Efficacy of Gypsy Moth Biocontrol’,inProceedings: Ecology, Survey and Management of Forest Insects, Krakow, Poland, September 1Á5 2002, eds. M.L. McManus and A.M. Liebhold, USDA Forest Service, Northeastern Research Station, General Technical Report NE-311, USA, pp. 80Á86. Saour, G. (2009), ‘Effect of Early Oviposition Experience on Host Acceptance in Trichogramma (Hymenoptera: Trichogrammatidae) and Application of F1 Sterility and T. principium to Suppress the Potato Tuber Moth (Lepidoptera: )’, Biocontrol Science and Technology, this volume. Schafellner, C., Marktl, R.C., and Schopf, A. (2007), ‘Inhibition of Juvenile Hormone Esterase Activity in Lymantria dispar (Lepidoptera, Lymantriidae) Larvae Parasitized by Glyptapanteles liparidis (Hymenoptera, Braconidae)’, Journal of Insect Physiology, 53, 858Á 868. Seth, R.K., and Barik, T.K. (2009), ‘Assessment of Infective Behaviour and Reproductive Potential over Successive Generations of Entomopathogenic Nematodes, Steinernema glaseri (Rhabditida: Steinernematidae), Reared within Radiosterilized Host Larvae, towards Spodoptera litura (Lepidoptera: )’, Biocontrol Science and Technology, this volume. Seth, R.K., Barik, T.K., and Chauhan, S. (2009), ‘Interaction of Entomopathogenic Nematodes, Steinernema glaseri (Rhabditida: Steinernematidae), Cultured in Irradiated Hosts, with ‘F1 Sterility’: Towards Management of a Tropical Pest, Spodoptera litura (Fabr.) (Lepidoptera: Noctuidae)’, Biocontrol Science and Technology, this volume. Steinberg, S., and Cayol, J.P. (2009), ‘Synergism between Biological Control and Sterile Insect Technique: Can Commercial Mass Production of Biocontrol Agents and Sterile Insects be Integrated within the Same Industrial Entity?’, Biocontrol Science and Technology, this volume. Downloaded By: [Hendrichs, Jorge] At: 16:18 4 November 2009 Tate, C.D., Carpenter, J.E., and Bloem, S. (2007), ‘Influence of the Dose of Radiation on the Level of F1 Sterility in the Cactus Moth, Cactoblastis cactorum (Lepidoptera: Pyralidae)’, Florida Entomologist, 90, 537Á544. Tate, C.D., Hight, S.D., and Carpenter, J.E. (2009), ‘Role of Inherited Sterility in Risk Assessment of Biological Control Agents of Weeds: Influence of Radiation Treatment on Cactoblastis cactorum (Lepidoptera: Pyralidae) Host Preference for Oviposition’, Biocontrol Science and Technology, this volume. Thomas, M.B., and Willis, A.J. (1998), ‘Biocontrol Á Risky but Necessary?’, Trends in Ecology and Evolution, 13, 325Á329. Tillinger, N.A., Hoch, G., and Schopf, A. (2004), ‘Effects of Parasitoid Associated Factors of the Endoparasitoid Glyptapanteles liparidis (Hymenoptera: Braconidae)’, European Journal of Entomology, 101, 243Á249. Tunc¸bilek, A.S., Canpolat, U., and Ayvaz, A. (2009a), ‘Effects of Gamma Radiation on Suitability of Stored Cereal Pest Eggs and the Reproductive Capability of the Egg Parasitoid Trichogramma evanescens (Trichogrammatidae: Hymenoptera)’, Biocontrol Science and Technology, this volume. Tunc¸bilek, A.S., Canpolat, U., and Sumer, F. (2009b), ‘Suitability of Irradiated and Cold- stored Eggs of Ephestia kuehniella (Pyralidae: Lepidoptera) and Sitotroga cerealella 22 J. Hendrichs et al.

(Gelechidae: Lepidoptera) for Stockpiling the Egg-parasitoid, Trichogramma evanescens (Trichogrammatidae: Hymenoptera) in Diapause’, Biocontrol Science and Technology, this volume. van Driesche, R., Hoddle, M., and Center, T. (2008), Control of Pests and Weeds by Natural Enemies, Hoboken, NJ, USA: Wiley-Blackwell. van Lenteren, J.C., Babendreier, D., Bigler, F., Burgio, G., Hokkanen, H.M.T., Kuske, S., Loomans, A.J.M., Menzler-Hokkanen, I., van Rijn, P.C.J., Thomas, M.B., Tommasini, M.G., and Zeng, Q.Q. (2003), ‘Environmental Risk Assessment of Exotic Natural Enemies Used in Inundative Biological Control’, Biocontrol, 48, 3Á38. van Lenteren, J.C., Bale, J., Bigler, F., Hokkanen, H.M.T., and Loomans, A.J.M. (2006), ‘Assessing Risks of Releasing Exotic Biological Control Agents of Arthropod Pests’, Annual Review of Entomology, 51, 609Á634. Vey, A., and Causse, R. (1979), ‘Effect de l’Exposition aux Rayons Gamma sur la Re´action Haemocytaire Multicellulare des Larves de Mamestra brassica (Lep.: Noctuidae)’, Entomophaga, 24, 41Á47. Viscarret, M.M., La Rossa, R., Segura, D.F., Ovruski, S.M., and Cladera, J. (2006), ‘Evaluation of the Parasitoid Diachasmimorpha longicaudata (Ashmead) (Hymenoptera: Braconidae) Reared on a Genetic Sexing Strain of Ceratitis capitata (Wied.) (Diptera: Tephritidae)’, Biological Control, 36, 147Á153. Vreysen, M.J.B., Robinson, A.S., and Hendrichs, J. (2007), Area-Wide Control of Insect Pests: From Research to Field Implementation, Dordrecht, The Netherlands: Springer, 789 pp. Wajnberg, E., Scott, J.K., and Quimby, P. (2001), Evaluating Indirect Ecological Effects of Biological Control, Oxon, UK: CABI Publishing, CAB International, 261 pp. Wang, E., Lu, D., Liu, X., and Li., Y. (2009), ‘Evaluating the Use of Nuclear Techniques for Colonization and Production of Trichogramma chilonis in Combination with Releasing Irradiated Moths for Control of Cotton Bollworm, Helicoverpa armigera’, Biocontrol Science and Technology, this volume. Wong, T.T.Y., Ramadan, M.M., Herr, J.C., and McInnis, D.O. (1992), ‘Suppression of the Mediterranean Fruit Fly Population with Concurrent Parasitoid and Sterile Fly Releases in Kula, Maui, Hawaii’, Journal of Economic Entomology, 85, 1671Á1681. Yudelman, M., Ratta, A., and Nygaard, D. (1998), ‘Pest Management and Food Production, Looking to the Future’, Food, Agriculture and the Environment Discussion Paper 25. International Food Policy Research Institute, Washington, DC, USA. Zapater, M.C., Andiarena, C.E, Pe´rez-Camargo, G., and Bartoloni, N. (2009), ‘Use of Irradiated Musca domestica Pupae to Optimize Mass Rearing and Commercial Shipment of the Parasitoid Spalangia endius (Hymenoptera: Pteromalidae)’, Biocontrol Science and Technology, this volume. Zubrik, M., and Novotny, J. (2009), ‘Impact of Gamma Radiation on the Developmental Characteristics of the Gypsy Moth, Lymantria dispar (Lepidoptera: Lymantriidae) Larvae

Downloaded By: [Hendrichs, Jorge] At: 16:18 4 November 2009 Preparatory to Their Use as Supplemental Hosts/Prey for Natural Enemy Enhancement’, Biocontrol Science and Technology, this volume. Biocontrol Science and Technology, Vol. 19, S1, 2009, 23Á34

Gamma radiation-induced pseudoparasitization as a tool to study interactions between host insects and parasitoids in the system Lymantria dispar (Lep., Lymantriidae) Glyptapanteles liparidis (Hym., Braconidae)Á Gernot Hoch*, Robert C. Marktl, and Axel Schopf

Department of Forest and Soil Sciences/Institute of Forest Entomology, Forest Pathology and Forest Protection, BOKU-University of Natural Resources and Applied Life Sciences, Vienna, Austria

Larvae of the koinobiont endoparasitoid Glyptapanteles liparidis (Hym., Braco- nidae) need to suppress the immune responses of parasitized Lymantria dispar host larvae while maintaining them at high nutritional quality. We used the method of g-radiation-induced pseudoparasitization to study the effects of the parasitoid’s polydnavirus and venom in these processes. To achieve pseudopar- asitization, G. liparidis females were irradiated in a cobalt-60 irradiator; such wasps injected during oviposition nonviable eggs along with polydnavirus and venom into the host. Glyptapanteles liparidis eggs or larvae were implanted into unparasitized or pseudoparasitized L. dispar larvae together with or without the parasitoid’s teratocytes. Eggs or larvae of G. liparidis implanted into unparasi- tized hosts were readily encapsulated by the host hemocytes. The further development of the hosts was not impaired. Implantation into pseudoparasitized hosts prevented encapsulation; complete endoparasitic development, however, was only possible when also teratocytes were implanted along with parasitoids into the L. dispar larva. These parasitoids required longer to emerge from the host compared to natural parasitization, but they were able to complete metamor- phosis into imagines. Analysis of trehalose levels in the host hemolymph and glycogen in host tissue revealed that G. liparidis polydnavirus/venom is responsible for an alteration of carbohydrate metabolism in L. dispar that is probably beneficial for the developing parasitoid. Downloaded By: [Hendrichs, Jorge] At: 16:21 4 November 2009 Keywords: parasitoids; host defense; polydnavirus; parasitoid nutrition; g-radiation; Glyptapanteles liparidis; Lymantria dispar

Introduction Interactions between insect parasitoids and their hosts have been shown to be a complex struggle between the two counterparts. Fine mechanisms have evolved in parasitic insects, particularly in endoparasitic koinobionts, to keep the balance between succeeding over the host’s defense reactions and maintaining the host at high quality for the developing parasitoid. Braconid endoparasitoids use parasitoid associated factors, such as symbiotic polydnaviruses (PDV), venom, and teratocytes, to manipulate physiological functions of the host. During oviposition, particles of PDV that replicate in the calyx region of the ’s ovary are injected into the

*Corresponding author. Email: [email protected]

First Published Online 16 October 2008 ISSN 0958-3157 print/ISSN 1360-0478 online # 2009 Taylor & Francis DOI: 10.1080/09583150802434059 http://www.informaworld.com 24 G. Hoch et al.

hemocoel of the host together with venom. One of the main functions of PDV, in many cases supported by venom, is the suppression of the host immune responses to prevent damage to the endoparasitoid (Lavine and Beckage 1995; Strand and Pech 1995; Shelby and Webb 1999; Beckage and Gelman 2004). PDV are injected as virions which infect cells in the lepidopteran host, most importantly hemocytes, where specific viral genes are expressed (Kroemer and Webb 2004). The cellular immune response is directly impaired by causing failure of hemocyte spreading behavior or induction of apapotosis of hemocytes (Beckage and Gelman 2004). Moreover, PDV have been shown to manipulate host development by interfering with its hormone metabolism (Beckage 1997; Cusson, Laforge, Miller, Cloutier, and Stoltz 2000; Beckage and Gelman 2004; Schafellner, Marktl, and Schopf 2007). Teratocytes are a third factor that support the parasitoids. These cells are released from the serosal membrane of the parasitoid egg. They live individually in the host hemocoel and grow to large size; teratocytes have been shown to aid nutrition of the parasitoid, exert immunological and antimicrobial functions, and play a role in altering the host’s endocrine system (Dahlman 1990; Dahlman and Vinson 1993; Nakamatsu, Fujii, and Tanaka 2002; Bell, Kirkbride-Smith, Marris, and Edwards 2004). The gregarious larvae of Glyptapanteles liparidis (Bouche´) (Hym., Braconidae) are endoparasitic koinobionts that maintain their host larva in suitable physiological condition until completion of the larval development and emergence from the host. The wasp parasitizes young to mid-stage larvae; depending on host size at time of oviposition, females lay five to 50 eggs into the host’s hemocoel. After 2Á3 weeks of endoparasitic development through two instars, the larvae emerge from the host while molting to the third instar, spin a cocoon and pupate (Schopf and Steinberger 1996; Schopf and Hoch 1997). Parasitism by G. liparidis leads to alterations in host development as well as in levels of certain nutrients, such as trehalose and fatty acids (Schopf and Steinberger 1996; Hoch and Schopf 2001; Hoch, Schafellner, Henn, and Schopf 2002). We explored a method to study the effect of PDV plus venom in the system G. liparidisÁL. dispar by sterilizing female wasps with g-radiation from a cobalt-60 source (Tillinger, Hoch, and Schopf 2004). Such irradiation is normally used in sterile insect (SIT) programs (Bakri, Mehta, and Lance 2005; Klassen and Curtis 2005). Our method followed the idea of Soller and Lanzrein (1996) who employed X-rays to sterilize female Chelonus inanitus to study effects of polydnavirus Downloaded By: [Hendrichs, Jorge] At: 16:21 4 November 2009 and venom. In reference to Jones (1985) who reported on parasitized larvae that showed symptoms typical for parasitization but did not contain obvious or living parasitoids at that time, we use the term ‘pseudoparasitization’ for this procedure. One advantage of this method in comparison to injection of purified PDV and venom is that it reflects the natural situation of parasitism as far as amount of injected parasitoid associated substances and minimal impact on the host larva during the process of injection are concerned. A disadvantage is that PDV and venom can only be studied together. Radiation-induced pseudoparasitization by G. liparidis was shown to affect both, host development Á resulting in prolonged development and frequently incomplete metamorphosis Á and immune competence (Tillinger et al. 2004). In this study, we used the technique of g-radiation-induced pseudoparasitization to study interactions in the hostÁparasitoid system, L. disparÁG. liparidis and complemented it with implantation experiments. The goal was to develop a technique that can be used as a basic tool for our future research. We established Biocontrol Science and Technology 25

methods to harvest G. liparidis at different developmental stages as well as teratocytes and to subsequently implant them into new host larvae. PDV and venom were administered via pseudoparasitization. The experiments demonstrate that implanted parasitoids require both PDV/venom and teratocytes to successfully complete their development. We report the effects of the various implantation treatments on host and parasitoid development. In the second part of the study, we applied radiation-induced pseudoparasitization to test whether PDV and venom of G. liparidis are responsible for observed alterations in carbohydrate metabolism of parasitized L. dispar larvae.

Materials and methods Insects Lymantria dispar larvae were obtained from egg masses provided by the USDA/ APHIS Otis Methods Development Center at Cape Cod, MA, USA. Larvae were reared on high wheat germ diet (Bell, Owens, Shapiro, and Tardiff 1981) in groups in 250-mL plastic cups at 20918C and 16 h L:8 h D photoperiod. Larvae that had received parasitoid implants were reared individually in 90-mm diameter Petri dishes under temperature and light regime as above and checked daily. Glyptapanteles liparidis were obtained from our laboratory colony, originating from parasitized L. dispar collected in oak forests in Burgenland, Austria. Adult wasps were reared on water and honey at 15/108C and 16 h L:8 h D photoperiod. To achieve controlled parasitization, host larvae were offered to wasps with a pair of forceps until the sting occurred. This procedure results in approximately 95% successful parasitism and avoids superparasitism.

Irradiation of parasitoid wasps Glyptapanteles liparidis females were g-irradiated based on the method outlined in Tillinger et al. (2004). Six-day-old female wasps with at least one oviposition experience were placed in ventilated 101015-cm plastic boxes and irradiated with 96 Gy (at a dose rate of 25.6 Gy/min) in a Gammacell 220 cobalt-60 irradiator Downloaded By: [Hendrichs, Jorge] At: 16:21 4 November 2009 (Atomic Energy of Canada Ltd.) at the FAO/IAEA Agriculture and Biotechnology Laboratories, Seibersdorf, Austria. Wasps were used for the experiments within 48 h post irradiation for a maximum of four ovipositions in order to prevent recovery of egg viability.

Implantation of G. liparidis into L. dispar larvae To study the effects of parasitoid associated factors on the development of G. liparidis and the L. dispar host larvae, we implanted different parasitoid stages into unparasitized or pseudoparasitized L. dispar larvae. Development of both, host and parasitoids was monitored. G. liparidis were harvested either from normally parasitized hosts (larvae hatched in vivo) or derived from parasitoid eggs that hatched in vitro. To harvest in vivo larvae, a normally parasitized L. dispar larva (donor host) was dissected at a certain stage post parasitization. Parasitoid larvae were washed three times in 1 mL TC100 insect medium (Sigma-Aldrich) to remove 26 G. Hoch et al.

teratocytes and adhering host hemolymph and immediately implanted into an unparasitized or pseudoparasitized gypsy moth larva. Both, recipient and donor larvae were surface sterilized with 70% ethanol. To harvest larvae that hatched in vitro and their released teratocytes, five parasitoid eggs were dissected out of regularly parasitized L. dispar larvae and incubated in 10 mL TC100 medium in well plates (96 well microtiter plates with V-bottom, Roth GmbH) at 278C in darkness. The parasitoid larvae and teratocytes were retrieved from the wells together with the culture medium 24 h after hatch and implanted into the unparasitized or pseudoparasitized host larva with micro capillary pipettes after thoroughly washing them in culture medium. To perform the implantations, recipient host larvae were anesthetized with CO2. A small incision was cut with ocular scissors at the base of an abdominal leg and the G. liparidis larvae were implanted through this opening with a micro capillary pipette. Recipient larvae were placed individually on filter paper in Petri dishes until bleeding ceased. Then they were supplied with wheat germ diet and reared as described above. Implantation of G. liparidis eggs or larvae into unparasitized recipient hosts: G. liparidis eggs or newly hatched larvae were harvested from donor hosts 4 or 5 days post-parasitization (dpp), respectively. Either five parasitoid eggs or five parasitoid larvae were implanted into L. dispar larvae on day 1 in the fifth instar as described above. One group of larvae received only 10 mL of TC100 insect medium; controls remained untreated. Implantation of G. liparidis larvae plus teratocytes into pseudoparasitized hosts: L. dispar larvae were pseudoparasitized in premolt to the fourth instar (indicated by slipped head capsule). One group received five newly hatched parasitoid larvae together with their teratocytes on day 3 of the host’s fourth instar. The other group of L. dispar larvae received five newly hatched G. liparidis larvae plus teratocytes on day 1 of the host’s fifth instar. In the first case, we harvested parasitoid larvae in vivo. For the second implantation we switched to harvesting in vitro (see above), which appeared to be more feasible because it offers controlled conditions. Implantation of G. liparidis larvae without teratocytes into pseudoparasitized recipient hosts: L. dispar larvae were pseudoparasitized in premolt to fourth instar. Glyptapanteles liparidis larvae that had been harvested in vivo were implanted on day 3 in fourth instar. The following treatments were tested: 10 larvae received one Downloaded By: [Hendrichs, Jorge] At: 16:21 4 November 2009 second instar G. liparidis larva (harvested from the donor host 11 dpp), 10 larvae were injected with 10 mL TC100 medium, controls were pseudoparasitized but did not receive any implants or injections. After rearing for 8 or 9 days, the recipient larvae were dissected and the status of the implanted parasitoids was evaluated under the dissecting microscope. Harvesting larvae without teratocytes for im- plantations was only possible when the parasitoids had molted into second instar. Younger parasitoids had always significant numbers of teratocytes attached that could not be completely removed by several washings in insect medium.

Effects of pseudoparasitization on host carbohydrate metabolism Lymantria dispar larvae were pseudoparasitized in premolt to the third instar. Unparasitized larvae served as controls. Larvae were reared in groups in 250-mL plastic cups and checked daily. On days 5, 7, 9 and 11 post-pseudoparasitization, samples for nutrient analysis were taken from larvae that were of the same age in the Biocontrol Science and Technology 27

respective instar. After cutting one proleg of a L. dispar larva, 20 mL of hemolymph were collected with a micro capillary pipette and transferred into a reaction tube containing 180 mL ice-cold 50% (v/v) aqueous methanol. After collecting the hemolymph sample, larvae were immediately dissected on parafilm (American National Can) in a wax dish on ice. The guts were removed and the carcass rinsed with ice-cold distilled water. The washed L. dispar carcasses were lyophilized and weighed. Hemolymph samples were prepared and analyzed for trehalose as described in Hoch et al. (2002). Briefly, after adding 30 mL 0.1% pentaerythrit (Sigma) as an internal standard, the samples were centrifuged to eliminate cellular debris. The supernatant was washed in chloroform. The methanolic phase was dried under vacuum and resuspended in Milli-Q ultrapure water (Millipore). Sugars were separated by HPLC (Hewlett-Packard 1050 with Biorad Aminex HPX-97P column) followed by refractive index detection (Hewlett-Packard 1047A). The lyophilized carcasses were prepared for glycogen analysis by homogenizing and washing with ethanol and water to eliminate low molecular weight carbohydrates. Glycogen was cleaved into glucose by incubating the sample with heat-stable a-amylase (500 U/mL) from Bacillus licheniformis (Sigma). After centrifugation, aliquots of the supernatant were incubated with amyloglucosidase (20 U/mL) from Aspergillus niger (Boehringer Mannheim) (Hoch et al. 2002). These samples were then analyzed for glucose contents by HPLC as described above.

Statistical analyses Statistical analyses were carried out with SPSS 12.0.1 for Windows (SPSS Inc., 2003). Data were analyzed for normal distribution with KolmogorovÁSmirnov Z- test. Homogeneity of variances was tested with Levene’s test. Means of normally distributed data were compared by one-way ANOVA, and post-hoc analyzed by Scheffe test or Tamhane’s T2. Comparisons of two means of normally distributed data were carried out by Student’s t-test.

Results Development of G. liparidis implanted into L. dispar larvae/effects on host larvae Downloaded By: [Hendrichs, Jorge] At: 16:21 4 November 2009 When G. liparidis were implanted into unparasitized L. dispar larvae, parasitoids were unable to develop. Dissections of recipient hosts revealed that all implanted parasitoids, both egg and larval stages, had been encapsulated by host hemocytes (Figure 1). Moreover, implantation of parasitoids did not alter the development of the host. Only a short but statistically significant prolongation of the fifth instar occurred. The same prolongation occurred also in larvae into which only TC100 medium was injected. Duration of the pupal stage did not differ among treatments (Table 1). Pseudoparasitization of hosts prevented lethal encapsulation of implanted G. liparidis larvae by host hemocytes. However, no parasitoids were able to emerge from the recipient hosts when they had been implanted without teratocytes. Hosts that had received second instar parasitoids were dissected 9 days post-implantation (equaling a total age of the parasitoid of 20 days). All implanted parasitoids were full grown, normally developed second instars and alive at the time of dissection but showed partially melanized surfaces indicating host immune response, although at a reduced 28 G. Hoch et al.

Figure 1. Newly hatched G. liparidis larvae implanted into unparasitized L. dispar larvae were encapsulated by a thick layer of host hemocytes (arrows).

level. Moreover, these larvae were in very close contact with lobes of the host’sfat body that could hardly be removed from the larvae upon dissection (Figure 2). Successful development into adults was only possible when G. liparidis larvae were implanted together with teratocytes into pseudoparasitized hosts. Ninety-four percent of G. liparidis implanted into fourth instar hosts and 85% of parasitoid larvae implanted into fifth instar hosts completed development; this was not significantly different from natural parasitism in the respective instars. However, the duration of endoparasitic development of implanted parasitoids as well as total development to adults was significantly longer than in natural parasitization (Table 2).

Effects of pseudoparasitization on host carbohydrate levels Pseudoparasitization by G. liparidis led to alterations in hemolymph and tissue carbohydrate levels in L. dispar larvae. Trehalose concentrations in the hemolymph Downloaded By: [Hendrichs, Jorge] At: 16:21 4 November 2009 showed a steady increase within the fourth instar. This increase was steeper in pseudoparasitized larvae; consequently, trehalose concentrations were significantly elevated 9 and 11 days post-pseudoparasitization (Figure 3). In a similar way, glycogen content of the larvae increased during the fourth instar after a pronounced

Table 1. Duration of fifth instar and pupal stage of unparasitized male L. dispar that received implants of five G. liparidis eggs (harvested 4 dpp), five G. liparidis larvae (harvested 5 dpp), or 10 mL of TC100 medium and of untreated controls.

Controls TC100 5 eggs 5 larvae

Days in fifth instar 12.091.6 a 13.691.3 b 12.591.2 ab 13.491.7 b Days in 16.490.8 a 16.891.3 a 16.490.8 a 15.990.9 a N 10 9 14 11

Means9SD in a row followed by different letters are significantly different at PB0.05 (one-way ANOVA, post-hoc test: Scheffe). Biocontrol Science and Technology 29

Figure 2. G. liparidis larvae implanted into pseudoparasitized L. dispar hosts without teratocytes were able to develop but showed melanization on their surface and were tightly surrounded by lobes of the host fat body (total age of parasitoids at dissection was 20 days).

decline during the molting process. Glycogen content per mg dry host tissue was significantly higher 9 and 11 days post-pseudoparasitization compared to unparasitized larvae (Figure 3). Pseudoparasitized larvae showed also modified development, manifested in significantly higher dry mass on day 11 post- pseudoparasitization than unparasitized controls (22.791.4 vs. 16.691.3 mg).

Discussion Implantations of G. liparidis were only successful when host larvae were treated with PDV/venom. Without immune suppression by these associated factors, the host responded with immediate encapsulation of the parasitoid by hemocytes, sometimes accompanied by melanization. The host immune system coped with the implanted parasitoids; no effect other than a short prolongation of the host instar following the

Downloaded By: [Hendrichs, Jorge] At: 16:21 4 November 2009 Table 2. Duration of endoparasitic and total development (in days) and developmental success of G. liparidis after regular parasitization or implantation of five parasitoid larvae together with teratocytes into pseudoparasitized L. dispar larvae.

Regular parasitization by Implantation of G. liparidis parasitoids

d2 in L31 d3 in L41 d1 in L51 d3 in L41 d1 in L51

Endoparasitic development [d] 14.890.8 b 12.290.6 a 16.391.4 c 18.391.3 d 18.190.7 d Total development [d] 21.290.8 b 18.090.6 a 23.191.5 c 25.490.9 d 25.191.1 d No. of hatched wasps per host 33.49 13.4 37.3914.4 34.1914.1 4.090.8 4.490.6 % Successful development2 96.9 a 95.1 ab 94.6 ab 84.6 b 93.9 a n 35 33 34 11 14

Means9SD within a row followed by different letters are significantly different (PB0.05; one-way ANOVA, post-hoc tests: Scheffe or Tamhane’s T2). 1Developmental stage of host at treatment: e.g. d2 in L3second day in third instar. 2Data were arcsin-transformed before statistical analysis. The table shows values before transformation. 30 G. Hoch et al.

* ***

Figure 3. Trehalose concentration in hemolymph and glycogen content of tissue (n10Á12) of L. dispar larvae 5, 7, 9, and 11 days post-pseudoparasitization by G. liparidis. Arrows indicate the molt of the host larvae to fourth instar. Values from pseudoparasitized (gray boxes) and unparasitized larvae (white boxes) within a day marked with an asterisk are significantly different (PB0.01; t-test).

implantation was noticed. This prolongation was apparently due to the trauma of the treatment rather than energetic cost of encapsulation. Parasitoids were killed before they could have caused any alterations of host development due to interference with the host endocrine system (Schafellner, Marktl, Nussbaumer, and Schopf 2004). Also, teratocytes are unable to escape the immune response in L. dispar; when injected into unparasitized larvae they are totally cleared from the hemolymph (Schafellner et al. 2007). When G. liparidis were implanted into pseudoparasitized hosts without terato- cytes, the parasitoid larvae were able to continue development. PDV/venom

Downloaded By: [Hendrichs, Jorge] At: 16:21 4 November 2009 prevented the fatal immune response. Previous work had already shown that pseudoparasitized L. dispar larvae show reduced encapsulation of implanted artificial objects and reduced hemolymph melanization (Tillinger et al. 2004). However, as is suggested by our findings from the dissections, G. liparidis larvae implanted without teratocytes still suffered some attack from the host immune system and were apparently unable to complete their development. Pseudoparasi- tization alone did not permanently reduce hemolymph melanization. This is demonstrated by substantial melanin deposition on G. liparidis larvae implanted without teratocytes. Also C. kariyai implanted into hosts without teratocytes showed melanized surfaces 4 days post-implantation (Nakamatsu et al. 2002). Teratocytes of other parasitoid species were shown to exert immune suppressive functions such as reduction of host phenoloxidase activity (Bell et al. 2004) or of cellular immune response (Andrew, Basio, and Kim 2006). Moreover, it was interesting to observe that implanted G. liparidis larvae were always closely attached to the fat body Á pieces of which would remain attached to the larvae upon dissection Á while they are Biocontrol Science and Technology 31

normally floating freely in the hemolymph. This is in agreement with the finding that larvae of the braconid endoparasitoid Cotesia kariyai consume the fat body of their hosts with the help of teratocytes (Nakamatsu et al. 2002). We also noticed that the fat body of L. dispar normally parasitized by G. liparidis was always clearly reduced in volume compared to unparasitized larvae (personal observations). Generally, important involvement of teratocytes in nutrition of parasitoids is known from other systems (Dahlman 1990; Beckage and Gelman 2004). These cells can synthesize and release a fatty acid binding protein believed to aid fatty acid nutrition of the parasitoid (Falabella et al. 2005), proteins that inhibit protein synthesis in the host (Rana, Dahlman, and Webb 2002), or interfere with host development (Dahlman 1990; Bell et al. 2004). The nutritional deficiency and lack of a hormonal cue may have prevented emergence of the developed parasitoid larvae in our case. Only the complete re-assemblage of all components, i.e. G. liparidis larvae, teratocytes, and PDV/venom allowed successful development in our study. The successful implantation of G. liparidis demonstrates that administration of PDV/ venom by pseudoparasitization functions as during the act of regular parasitization; it conditions the host such that implanted parasitoids can develop successfully. Teratocytes were shown to be necessary to allow successful completion of endoparasitic development and emergence from the host. Pseudoparasitized insects showed significantly elevated levels of trehalose in the hemolymph and glycogen in the tissue. This demonstrates that PDV/venom assists the developing parasitoid by altering the nutritional situation in the host, which seems to be an important feature in addition to the suppression of the immune system. These nutritional alterations are in accordance with previous findings from normally parasitized L. dispar. There, trehalose titers are elevated in the early stage of parasitism but are significantly reduced towards the end of the endoparasitic phase. In contrast, glycogen content is maintained at the same levels as in unparasitized larvae (Schopf and Nussbaumer 1996; Hoch et al. 2002). On the other hand, fatty acid levels in the hemolymph as well as total lipids of normally parasitized L. dispar are significantly reduced (Bischof and Ortel 1996; Hoch et al. 2002). This indicates some action of the parasitoid to redirect the host’senergy metabolism for the parasitoid’s advantage. Alterations of the host’s metabolism towards increased trehalose titers in the hemolymph have been shown in several host Downloaded By: [Hendrichs, Jorge] At: 16:21 4 November 2009 parasitoid systems and were interpreted as beneficial for the hemolymph feeding parasitoids (Thompson, Lee, and Beckage 1990; Vinson 1990; Thompson and Dahlman 1999). We were able to demonstrate that PDV/venom of G. liparidis are involved in such alterations of host metabolism. Thereby, PDV/venom could assist the developing parasitoids by having higher levels of nutrients easily available in the hemolymph. The finding that treatment of Pseudaletia separata larvae with C. kariyai PDV/venom led to increased approximate digestibility but decreased efficiency of conversion of digested food by the host larva (Nakamatsu, Gyotoku, and Tanaka 2001) further supports this assumption of an important supportive function of parasitoid associated factors in parasitoid nutrition. Overall, our study demonstrates the use of g-radiation-induced pseudoparasiti- zation as a tool to investigate interactions between host insects and their parasitoids. PDV and venom administered to L. dispar host larvae by pseudoparasitization suppressed the host immune response and altered its metabolism such that implanted G. liparidis larvae can develop successfully. This tool will be highly useful in future 32 G. Hoch et al.

studies on the delicate physiological interaction between this parasitoid and its host. We were able to use a g-radiation source at the FAO/IAEA laboratories for our study, where it is used for research and development in sterile insect technique (SIT) programs. Sterilization of insects is normally done with either cobalt-60 and caesium-137 irradiators. Both must meet the very high standards for nuclear safety and must be licensed by national atomic energy authorities (Bakri et al. 2005). Moreover, transportation of nuclear material is becoming increasingly problematic. Therefore, alternatives such as electron beams or X-rays are being explored also for SIT programs (Robinson and Hendrichs 2005; Mehta, in press). Such X-ray sources that can be used for insect sterilization as well as pseudoparasitization may eventually become more accessible to researchers than g-radiation sources.

Acknowledgements We are grateful to Dr Alan Robinson of the FAO/IAEA Agriculture and Biotechnology Laboratory, Seibersdorf, Austria, for irradiation of the parasitoid wasps. We thank Dr D.L. Dahlman, University of Kentucky, Lexington for introducing R.C.M. to in vitro techniques with teratocytes. Gypsy moth eggs were kindly provided by USDA/APHIS Otis Methods Development Center at Cape Cod, MA, USA. We thank Ms Andrea Stradner and Mr Petr Zabransky for their technical assistance. Funding was provided from grant P13603-BIO by the Austrian Science Fund (FWF) to A.S. This work is part of the FAO/IAEA coordinated research project CRP D4.30.02 (project coordinator: Dr Jorge Hendrichs).

References Andrew, N., Basio, M., and Kim, Y. (2006), ‘Additive Effect of Teratocytes and Calyx Fluid from Cotesia plutellae on Immunosuppression of Plutella xylostella’, Physiological Entomology, 31, 341Á347. Bakri, A., Mehta, K., and Lance, D.R. (2005), ‘Sterilizing Insects with Ionizing Radiation’,in Sterile Insect Technique. Principle and Practice in Area-Wide Integrated Pest Management, eds. V.A. Dyck, J. Hendrichs and A.S. Robinson, Dordrecht: Springer, pp. 233Á268. Beckage, N.E. (1997), ‘New Insights: How Parasites and Pathogens Alter the Endocrine Physiology and Development of Insect Hosts’,inParasites and Pathogens: Effects on Host Hormones and Behavior, ed. N.E. Beckage, New York: Chapman and Hall, pp. 3Á36. Beckage, N.E., and Gelman, D.B. (2004), ‘Wasp Parasitoid Disruption of Host Development: Implications for New Biologically Based Strategies for Insect Control’, Annual Review of Downloaded By: [Hendrichs, Jorge] At: 16:21 4 November 2009 Entomology, 49, 299Á330. Bell, R.A., Owens, C.D., Shapiro, M., and Tardiff, J.R. (1981), ‘Mass Rearing and Virus Production’,inThe Gypsy Moth: Research toward Integrated Pest Management, eds. C.C. Doane, and M.L. McManus, Washington, DC: USDA [U.S. Dept. Agric. For. Serv. Tech. Bull. 1584], pp. 599Á600. Bell, H.A., Kirkbride-Smith, A.E., Marris, G.C., and Edwards, J.P. (2004), ‘Teratocytes of the Solitary Endoparasitoid Meteorus gyrator (Hymenoptera: Braconidae): Morphology, Numbers and Possible Functions’, Physiological Entomology, 29, 335Á343. Bischof, C., and Ortel, J. (1996), ‘The Effects of Parasitism by Glyptapanteles liparidis (Braconidae: Hymenoptera) on the Hemolymph and Total Body Composition of Gypsy Moth Larvae (Lymantria dispar, Lymantriidae: Lepidoptera)’, Parasitology Research, 82, 687Á692. Cusson, M., Laforge, M., Miller, D., Cloutier, C., and Stoltz, D. (2000), ‘Functional Significance of Parasitism-Induced Suppression of Juvenile Hormone Esterase Activity in Developmental Delayed Choristoneura fumiferana Larvae’, General and Comparative Endocrinology, 117, 343Á354. Dahlman, D.L. (1990), ‘Evaluation of Teratocyte Functions: An Overview’, Archives of Insect Biochemistry and Physiology, 13, 159Á166. Biocontrol Science and Technology 33

Dahlman, D.L., and Vinson, S.B. (1993), ‘Teratocytes: Developmental and Biochemical Characteristics’, in eds. N.E. Beckage, S.N. Thompson, and B.A. Federici, Parasites and Pathogens of Insects. Vol. 1, San Diego: Academic Press, pp. 145Á165. Falabella, P., Perugino, G., Caccialupi, P., Riviello, L., Varricchio, P., Tranfaglia, A., Rossi, M., Malva, C., Graziani, F., Moracci, M., and Pennacchio, F. (2005), ‘A Novel Fatty Acid Binding Protein Produced by Teratocytes of the Aphid Parasitoid Aphidius ervi’, Insect Molecular Biology, 14, 195Á205. Hoch, G., and Schopf, A. (2001), ‘Effects of Glyptapanteles liparidis (Hym.: Braconidae) Parasitism, Polydnavirus, and Venom on Development of Microsporidia-Infected and Uninfected Lymantria dispar (Lep.: Lymantriidae) Larvae’, Journal of Invertebrate Pathology, 77, 37Á43. Hoch, G., Schafellner, C., Henn, M.W., and Schopf, A. (2002), ‘Alterations in Carbohydrate and Fatty Acid Levels of Lymantria dispar Larvae Caused by a Microsporidian Infection and Potential Adverse Effects on a Co-Occurring Endoparasitoid, Glyptapanteles liparidis’, Archives of Insect Biochemistry and Physiology, 50, 109Á120. Jones, D. (1985), ‘Parasite Regulation of Host Insect Metamorphosis: A New Form of Regulation in Pseudoparasitized Larvae of Trichoplusia ni’, Journal of Comparative Physiology B, 155, 583Á590. Klassen, W., and Curtis, C.F. (2005), ‘History of the Sterile Insect Technique’,inSterile Insect Technique. Principle and Practice in Area-Wide Integrated Pest Management, eds. V.A. Dyck, J. Hendrichs and A.S. Robinson, Dordrecht: Springer, pp. 3Á36. Kroemer, J.A., and Webb, B.A. (2004), ‘Polydnavirus Genes and Genomes: Emerging Gene Families and New Insights into Polydnavirus Replication’, Annual Review of Entomology, 49, 431Á456. Lavine, M.D., and Beckage, N.E. (1995), ‘Polydnaviruses: Potent Mediators of Host Insect Immune Dysfunction’, Parasitology Today, 11, 368Á378. Mehta, K. (2008), ‘Radiation Sources Supporting the Use of Natural Enemies for Biological Control of Agricultural Pests’, Biocontrol Science and Technology (in press). Nakamatsu, Y., Gyotoku, Y., and Tanaka, T. (2001), ‘The Endoparasitoid Cotesia kariyai (Ck) Regulates the Growth and Metabolic Efficiency of Pseudaletia separata Larvae by Venom and Ck Polydnavirus’, Journal of Insect Physiology, 47, 573Á584. Nakamatsu, Y., Fujii, S., and Tanaka, T. (2002), ‘Larvae of an Endoparasitoid, Cotesia kariyai (Hymenoptera: Braconidae), Feed on the Host Fat Body Directly in the Second Stadium with the Help of Teratocytes’, Journal of Insect Physiology, 48, 1041Á1052. Rana, R.L., Dahlman, D.L., and Webb, B.A. (2002), ‘Expression and Characterization of a Novel Treatocyte Protein of the Braconid, Microplitis croceipes (cresson)’, Insect Biochemistry and Molecular Biology, 32, 1507Á1516. Robinson, A.S., and Hendrichs, J. (2005), ‘Prospects for the Future Development and Application of the Sterile Insect Technique’,inSterile Insect Technique. Principle and Downloaded By: [Hendrichs, Jorge] At: 16:21 4 November 2009 Practice in Area-Wide Integrated Pest Management, eds. V.A. Dyck, J. Hendrichs and A.S. Robinson, Dordrecht: Springer, pp. 727Á760,. Schafellner, C., Marktl, R.C., Nussbaumer, C., and Schopf, A. (2004), ‘Parasitism-Induced Effects of Glyptapanteles liparidis (Hym., Braconidae) on the Juvenile Hormone Titer of its Host, Lymantria dispar: The Role of the Parasitoid Larvae’, Journal of Insect Physiology, 50, 1181Á1189. Schafellner, C., Marktl, R.C., and Schopf, A. (2007), ‘Inhibition of Juvenile Hormone Esterase Activity in Lymantria dispar (Lepidoptera, Lymantriidae) Larvae Parasitized by Glyptapanteles liparidis (Hymenoptera, Braconidae)’, Journal of Insect Physiology, 53, 858Á 868. Schopf, A., and Hoch, G. (1997), ‘Bionomics and the Significance of Glyptapanteles liparidis (Hym., Braconidae) as a Regulator of Lymantria dispar (Lep., Lymantriidae) in Different Host Population Densities (in German, English Abstract)’, Journal of Applied Entomology, 121, 195Á203. Schopf, A., and Nussbaumer, C. (1996), ‘Influence of Parasitism by Glyptapanteles liparidis (Hym., Braconidae) on the Hemolymph Carbohydrate and Glycogen Content of its Host Larva, Lymantria dispar (Lep., Lymantriidae)’, Journal of Applied Entomology, 120, 357Á 362. 34 G. Hoch et al.

Schopf, A., and Steinberger, P. (1996), ‘The Influence of the Endoparasitic Wasp Glypta- panteles liparidis (Hymenoptera: Braconidae) on the Growth, Food Consumption, and Food Utilization of its Host Larva, Lymantria dispar (Lepidoptera: Lymantriidae)’, European Journal of Entomology, 93, 555Á568. Shelby, K.S., and Webb, B.A. (1999), ‘Polydnavirus-Mediated Suppression of Insect Immunity’, Journal of Insect Physiology, 45, 507Á514. Soller, M., and Lanzrein, B. (1996), ‘Polydnavirus and Venom of the Egg-Larval Parasitoid Chelonus inanitus (Braconidae) Induce Developmental Arrest in the Prepupa of its Host Spodoptora littoralis (Noctuidae)’, Journal of Insect Physiology, 42, 471Á481. Strand, M.R., and Pech, L.L. (1995), ‘Immunological Basis for Compatibility in Parasitoid- Host Relationships’, Annual Review of Entomology, 40, 31Á56. Thompson, S.N., and Dahlman, D.L. (1999), ‘Aberrant Nutritional Regulation of Carbohy- drate Synthesis by Parasitized Manduca sexta L.’, Journal of Insect Physiology, 44, 745Á753. Thompson, S.N., Lee, R.W.-K., and Beckage, N.E. (1990), ‘Metabolism of Parasitized Manduca sexta Examined by Nuclear Magnetic Resonance’, Archives of Insect Biochemistry and Physiology, 13, 127Á143. Tillinger, N.A., Hoch, G., and Schopf, A. (2004), ‘Effects of Parasitoid Associated Factors of the Endoparasitoid Glyptapanteles liparidis (Hymenoptera: Braconidae)’, European Journal of Entomology, 101, 243Á249. Vinson, S.B. (1990), ‘Physiological Interactions between the Host Genus Heliothis and its Guild of Parasitoids’, Archives of Insect Biochemistry and Physiology, 13, 63Á81. Downloaded By: [Hendrichs, Jorge] At: 16:21 4 November 2009 Biocontrol Science and Technology, Vol. 19, S1, 2009, 3542

Treatment of Lymantria dispar (Lepidoptera, Lymantriidae) host larvae with polydnavirus/venom of a braconid parasitoid increases spore production of entomopathogenic microsporidia Gernot Hocha*, Leellen F. Solterb, and Axel Schopfa

aDepartment of Forest and Soil Sciences, BOKU University of Natural Resources and Applied Life Sciences, Vienna, Austria; bCenter for Economic Entomology, Illinois Natural History Survey, Champaign, IL, USA

Female Glyptapanteles liparidis (Hym., Braconidae) were irradiated in a caesium- 137 irradiator; these wasps oviposit nonviable eggs along with polydnavirus and venom into the host (pseudoparasitization). When Lymantria dispar larvae were infected with microsporidian species for which they are permissive or semi- permissive hosts, spore production was higher in pseudoparasitized than in unparasitized larvae. Keywords: microsporidia; Nosema portugal; Vairimorpha necatrix; host suitability; polydnavirus; immune suppression

Introduction When Lymantria dispar larvae are parasitized by the endoparasitoid Glyptapanteles liparidis (Hym., Braconidae) and infected with the microsporidium Vairimorpha disparis they die more quickly than unparasitized, infected hosts. Moreover, reproduction of the microsporidium, measured in numbers of mature spores, is higher in the parasitized hosts (Hoch, Schopf, and Maddox 2000). In a follow-up study we showed that treatment of the hosts with the parasitoid’s polydnavirus (PDV) and venom also induced higher spore yield of V. disparis (Hoch and Schopf 2001). We hypothesized that a reduced immune response in PDV/venom treated

Downloaded By: [Hendrichs, Jorge] At: 16:22 4 November 2009 hosts could be responsible for this effect. Like other braconids and ichneumonids, G. liparidis possesses a symbiotic PDV that occurs in the calyx region of its ovary (Krell 1991). PDV is injected together with venom into the host hemocoel during oviposition. In several other host-parasitoid systems, these substances were shown to play a crucial role in preventing encapsulation of the parasitoid egg by the host immune system (e.g. Edson, Vinson, Stoltz, and Summers 1981; Lavine and Beckage 1995; Strand and Pech 1995; Shelby and Webb 1999) and to manipulate host development by interfering with hormonal metabolism (reviewed in Beckage 1997). Glyptapanteles liparidis PDV, e.g. is responsible for suppressed juvenile hormone esterase activity in parasitized L. dispar larvae (Schafellner, Marktl, and Schopf 2007). To study the effects of PDV and venom of G. liparidis we modified a method described by Soller and Lanzrein (1996) and used gamma radiation to sterilize eggs inside the ovaries of wasps; these wasps then insert nonviable eggs into the host

*Corresponding author. Email: [email protected]

First Published Online 10 October 2008 ISSN 0958-3157 print/ISSN 1360-0478 online # 2009 Taylor & Francis DOI: 10.1080/09583150802364868 http://www.informaworld.com 36 G. Hoch et al.

hemocoel together with active PDV and venom (pseudoparasitization) (Tillinger, Hoch, and Schopf 2004). Compared to injection of purified PDV and venom, pseudoparasitization has the advantage that a physiological dose is administered during oviposition and that the host is not affected by the application procedure itself. Pseudoparasitization by irradiated G. liparidis altered host development and interfered with larval-pupal molt as well as metamorphosis. Moreover, it suppressed the immune response of L. dispar larvae; encapsulation of implanted artificial objects was reduced and hemolymph melanization occurred at lower level (Tillinger et al. 2004). The responses of the insect immune system to microsporidian infections are not well known. A typical microsporidian life cycle in a lepidopteran host larva begins with the invasion of the midgut tissues after oral ingestion of infective spores. The pathogen then spreads to target tissues, e.g. other midgut cells, fat body or silk glands, where a secondary developmental cycle leads to production of persistent environmental spores (Maddox et al. 1999). In the laboratory, L. dispar is a semi- permissive host for a variety of entomopathogenic microsporidia isolated from other lepidopteran hosts. These infections, however, are often atypical and result in very low spore production (Solter and Maddox 1998). Hemocytic nodulation was regularly observed in one microsporidian species for which L. dispar was barely permissive. The most intense melanization of hemolymph, however, occurred as a consequence of heavy infections by microsporidia for which L. dispar is permissive and was, therefore not interpreted as sign of successful immune response (Hoch, Solter, and Schopf 2004). In the present study, we used pseudoparasitization by irradiated G. liparidis to test whether polydnavirus/venom-induced changes in host physiology enhance reproduction of microsporidia for which L. dispar larvae are permissive or semi- permissive hosts. Therefore, we measured microsporidian spore load in infected hosts. In addition, a qualitative evaluation was done by observing the progress and quality of infection under the light microscope.

Materials and methods Lymantria dispar egg masses were provided by the USDA/APHIS Otis Method Downloaded By: [Hendrichs, Jorge] At: 16:22 4 November 2009 Development Center, Cape Cod, MA. Larvae were reared on high wheat germ diet (Bell, Owens, Shapiro, and Tardiff 1981) at 24918C, 5560% relative humidity, and 16 h L:8 h D photoperiod. G. liparidis were obtained from the laboratory colony at BOKU University, Vienna. Adult wasps were reared on water and honey and L. dispar larvae were used as hosts. All microsporidian isolates were available from the microsporidia germplasm collection at the Illinois Natural History Survey and were used in previous studies. We used the following microsporidia isolated from L. dispar: (1) Nosema portugal and (2) Nosema sp., hereafter referred to as N. sp. [Schweinfurth]. These species were propagated in L. dispar larvae; spores were harvested and processed as described in Hoch et al. (2004). The following microsporidia, originally isolated from other lepidopteran hosts were also used in the experiments: (1) Vairimorpha necatrix, originally isolated from Pseudaletia unipuncta; (2) Vairimorpha sp. from Hyphantria cunea, hereafter referred to as V. sp. [H. cunea]; (3) Nosema sp. from Malacosoma americanum, hereafter referred to as N. sp. [M. americanum]. Details on these species Biocontrol Science and Technology 37

as well as production of inoculum are given in Hoch et al. (2004). All microsporidia were stored in liquid nitrogen (Maddox and Solter 1996) until used in the experiments, but no longer than 2 months. To achieve pseudoparasitization, wasps were temporarily sterilized by irradiation with 50 Gy at a dose rate of 7.6 Gy/min in the caesium-137 irradiator at the Department of Molecular and Integrative Physiology, University of Illinois. All female wasps were approximately 1-week-old, mated, and had previous oviposition experience. They were put in ventilated 30-mL plastic cups for irradiation. The day after irradiation, the wasps were allowed to sting up to four L. dispar larvae in premolt to third instar as described in Tillinger et al. (2004). On the second day in the third instar, larvae were inoculated with microsporidia at dosages of 1104 spores per larva (except N. sp. [M. americanum], for which the dosage was 1105). Exact dosages were administered via 4-mm3 diet cubes that were contaminated with 1 mL of spore suspension (Hoch et al. 2000, 2004). The larvae were then reared in groups of three to four in 30-mL plastic cups. Both, L. dispar larvae pseudoparasitized with G. liparidis and unparasitized larvae were infected with each microsporidian species from the same batch out of liquid nitrogen storage. Spore production in infected L. dispar was quantified following the method described in Hoch et al. (2000). Larvae were weighed and frozen 18 dpi unless otherwise indicated. Dead larvae were decapitated and homogenized in water in a tissue grinder. The homogenate was cleaned by filtration through cheese cloth and centrifugation in tap water. The spore pellet was re-suspended in water by filling vials to a total volume of 20 mL. Spores were counted in a Neubauer hemacytometer; four counts were carried out per sample. Infections with N. sp. [Schweinfurth], which reproduces mainly in the silk glands of L. dispar larvae, were additionally assessed by measuring the percentage of infested silk gland area on fresh preparations. The silk glands were dissected out of unparasitized and pseudoparasitized larvae 12 and 16 dpi (10 larvae per day and treatment) and photographed under a dissecting microscope. After coloring infested areas that can easily be recognized by their opaque color (as opposed to the clear uninfested areas), the digital images were analyzed using Lucia 4.21 software (Laboratory Imaging, Ltd.). The percentage of infested area of the projection of the silk gland was calculated. Besides these Downloaded By: [Hendrichs, Jorge] At: 16:22 4 November 2009 quantitative methods, infection levels in unparasitized and pseudoparasitized larvae were compared qualitatively by examination of fresh preparations of infected tissues under phase contrast microscopy. Statistical analysis was done using SPSS 12.0.1 for Windows (SPSS, Inc.). Data were tested for normal distribution with KolmogorovSmirnov Z-test. Means of normally distributed data were compared between unparasitized and pseudopar- asitized larvae by independent samples t-test. Data lacking normal distribution were compared by MannWhitney U-tests. Percent values were arcsin transformed before analysis. Spore production of N. portugal in unparasitized and pseudoparasitized hosts at different points in time was analyzed by two-way ANOVA using the GLM procedure of SPSS. The relationship between fresh mass and spore load of larvae was analyzed by computing Pearson’s correlation coefficient or Spearman’s r when data were not normally distributed. Relative frequencies were compared with x2 cross table analysis. 38 G. Hoch et al.

Table 1. Number of microsporidian spores produced per host, fresh mass of the infected host, and correlation between number of spores and fresh mass in unparasitized (np) and pseudoparasitized (psp) L. dispar larvae measured 18 days post-infection.

No of spores per host$ Fresh mass (mg)$ Correlation %

np psp np psp np psp

N. portugal 6. 910893.5107 8.210895.0107 593931 613928 0.466** 0.699** P0.036 P0.624 N.sp. 4.510795.6106 7.710799.9106 328926 464928 0.627** 0.674** [Schweinfurth] P0.007 P0.001 V.sp. 1.310892.2107 3.910896.0107 478939 561944 0.154 0.356* [H. cunea] P0.000 P0.166 V. necatrix 1.510893.7107 2.610895.2107 337933 383928 0.268 0.132 P0.045 P0.293

$Means9SE, n3435. Probability values refer to independent samples t-test (except spores/host in V. necatrix. MannWhitney U-test) comparing unparasitized and pseudoparasitized hosts within microsporidian species. %Pearson’s correlation coefficient (except V. necatrix: Spearman’s r); asterisks denote significant correlations at the 0.05 and 0.01 level, respectively.

Results and discussion The total spore production after 18 days of infection was generally higher in pseudoparasitized L. dispar larvae than in unparasitized hosts (Table 1). We counted significantly more spores in pseudoparasitized larvae infected with the L. dispar microsporidia N. portugal and N. sp. [Schweinfurth] as well as with microsporidia from other Lepidoptera, V. necatrix and V. sp. [H. cunea], for which L. dispar has been shown to be a semi-permissive host. The spore loads of the two L. dispar microsporidia, N. portugal and N. sp. [Schweinfurth], in the host were positively correlated with the fresh mass of the larva, both with or without pseudoparasitiza- tion, indicating optimal utilization of the host tissue by the microsporidium.

Downloaded By: [Hendrichs, Jorge] At: 16:22 4 November 2009 Nevertheless, pseudoparasitization led to a further increase in total spore produc- tion. For N. sp. [Schweinfurth] that reproduces mainly in the silk glands the higher spore load appears to be due to increased larval weight of pseudoparasitized hosts. There were no significant differences between pseudoparasitized and unparasitized larvae in spore load per mg fresh mass (t-test: P0.287). It is known from previous studies that PDV/venom of G. liparidis alters host growth and development, resulting in bigger larvae and heavier pupae (Tillinger et al. 2004). Analysis of the photographs of infected silk glands revealed no significant difference in the percentage of infected area in unparasitized (12 dpi: 70.594.6%; 16 dpi: 79.49 2.2%) and pseudoparasitized larvae (12 dpi: 72.993.5%; 16 dpi: 71.294.5%; t-test: P0.740 and P0.146, respectively). For N. portugal, which develops in the fat body as well as in the silk glands, the situation was different; pseudoparasitization also led to higher spore load per unit fresh weight. The increase in spore production was analyzed in more detail for N. portugal on days 10, 12, and 18 post-infection. A two-way ANOVA with date of count and parasitization as factors and fresh mass as Biocontrol Science and Technology 39

Figure 1. Number of Nosema portugal spores (logarithmic scale) produced in unparasitized and pseudoparasitized L. dispar larvae quantified 10, 12 and 18 days post-infection. Two-way ANOVA was performed with parasitization treatment and day of count as factors and fresh mass (FM) of the infected larva as covariate.

covariate showed a significantly higher amount of spores produced in pseudopar- asitized larvae together with a significant increase in spore load over time (Figure 1).

Downloaded By: [Hendrichs, Jorge] At: 16:22 4 November 2009 Gypsy moth is a semi-permissive host for V. necatrix; infection of the fat body is light and variable, and infection of silk glands remains localized (Hoch et al. 2004). There was no correlation between spore load in the host and fresh mass (Table 1) indicating that the microsporidium was not able to utilize the semi-permissive host optimally. The lack of correlation was due to a high percentage of larvae that showed only low infestation levels; 60% of the unparasitized and 43% of pseudoparasitized larvae had a spore load of less than 5107 spores. Spore counts at an earlier date (13 dpi) did not reveal any significant difference (P0.319) but pseudoparasitization led to a slight but significant increase in total spore load compared to unparasitized hosts 18 dpi (Table 1). The V. sp. [H. cunea] was likewise promoted by pseudoparasitization of the L. dispar larvae, which are also semi-permissive hosts for this species (Hoch et al. 2004); total spore load was significantly higher in pseudoparasitized larvae than in unparasitized L. dispar (Table 1). While unpar- asitized hosts showed no relationship between fresh mass and spore load there was a moderate but significant correlation in pseudoparasitized larvae infected with V. sp. 40 G. Hoch et al.

[H. cunea]. Apparently, the microsporidium was able to better utilize the tissue of the pseudoparasitized hosts. We could not quantify reproduction of N. sp. [M. americanum], the third microsporidium, for which L. dispar is only a semi-permissive host. Although it was fed to larvae at 10 times higher dosages than the other microsporidia, infectivity of this species was clearly below 100%. Moreover, only few and frequently atypically shaped spores were produced in L. dispar larvae. Of unparasitized larvae, 52.2% (n46) were diagnosed positive when dissected between 17 and 25 dpi. For pseudoparasitized larvae the percentage increased to 70.2% (n 47). This difference was not statistically significant (x2 3.188, P0.05). Based on our microscopic observations, the level of the infections did not differ in the two host types. We found very few spores and only in infected silk glands of both host types. Also, atypically shaped spores (Solter and Maddox 1998) occurred regularly in both unparasitized and pseudoparasitized hosts. Microscopy likewise revealed no qualitative differences of infections with V. necatrix and V. sp. [H. cunea] in the semi-permissive L. dispar larvae with or without pseudoparasitization. Progression of infection and infection of the respective tissues was not affected. In pseudopar- asitized hosts even the strongest infections remained localized. Thus, the improve- ment in host quality through pseudoparasitization seems to occur mostly on a quantitative level; a host classified as semi-permissive did not become permissive after pseudoparasitization. This appears to be different from baculovirus infections in lepidopteran larvae. Refractory hosts, such as Helicoverpa zea and Manduca sexta react to a systemic infection with Autographa californica multienveloped nucleopo- lyhedrovirus by encapsulation of viral foci in the tracheae and thereby block a subsequent spread of the infection. In hosts that were immunosuppressed either by chemicals or parasitization by hymenopteran parasitoids, virus infection caused higher mortality and was more widespread than in infected, immune competent hosts (Washburn, Kirkpatrick, and Volkman 1996; Washburn, Haas-Stapleton, Tan, Beckage, and Volkman 2000). Infection with microsporidia following oral inocula- tion differs also from injection of microorganisms into the hemocoel of larvae, in which case even normally non-pathogenic yeasts were shown to cause lethal infections in PDV-treated insects because the hemolymph was not cleared of infection (Stoltz and Guzo 1986). Pseudoparasitization of L. dispar suppressed hemolymph melanization after Downloaded By: [Hendrichs, Jorge] At: 16:22 4 November 2009 infection with the virulent microsporidium N. portugal. Melanization levels were significantly lower in pseudoparasitized, infected larvae than in unparasitized, infected larvae 7 dpi; levels further decreased to levels in unparasitized, uninfected controls 9 dpi (unpublished data). But overall, our findings suggest that it is not the level of host defense reaction alone that determines host suitability for microspor- idia. The gradual increase in suitability of semi-permissive hosts and of permissive hosts may also be caused by better nutritional quality of pseudoparasitized larvae. Microsporidia are obligate intracellular parasites and are thus completely dependent on resources provided by the host cell. We showed that infection of L. dispar larvae by V. disparis leads to total depletion of carbohydrates from the host as well as significantly reduced levels of lipids (Hoch, Schafellner, Henn, and Schopf 2002). A redirection of the host’s metabolism for the parasitoid’s advantage by its PDV/venom could, likewise, be beneficial for a developing microsporidian infection. Such alterations of the host’s metabolism by several parasitic wasps have been demon- strated (e.g. reviewed in Nakamatsu, Gyotoku, and Tanaka 2001), and we showed Biocontrol Science and Technology 41

that PDV/venom of G. liparidis leads to increased trehalose titers in the hemolymph and increased glycogen levels in the tissue of L. dispar larvae (Hoch et al., submitted). Hence, such changes in host metabolism may well be profitable for a developing microsporidium by supplying higher amounts of nutrients or energy for the parasite.

Acknowledgements We are grateful to Dr H.S. Ducoff, Department of Molecular and Integrative Physiology, University of Illinois, for irradiation of G. liparidis and dosimetry. Mr J. Tanner, USDA/ APHIS Otis Method Development Center, kindly provided the L. dispar used in this study. This research was funded by the Austrian Science Fund FWF (Erwin Schro¨dinger Fellowship J2027 to G.H.), the Illinois Natural History Survey, and the Office of Research/Illinois Agricultural Experiment Station Project No. 65-344_S301. The work is part of the FAO/IAEA coordinated research project CRP D4.30.02 (project coordinator: Dr Jorge Hendrichs).

References Beckage, N.E. (1997), ‘New Insights: How Parasites and Pathogens Alter the Endocrine Physiology and Development of Insect Hosts’,inParasites and Pathogens: Effects on Host Hormones and Behavior, ed. N.E. Beckage, New York: Chapman and Hall, pp. 336. Bell, R.A., Owens, C.D., Shapiro, M., and Tardiff, J.R. (1981), ‘Mass Rearing and Virus Production’,inThe Gypsy Moth: Research toward Integrated Pest Management, eds. C.C. Doane and M.L. McManus, Washington, DC: USDA (US Dept. Agric. For. Serv. Tech. Bull. 1584), pp. 599600. Edson, K.M., Vinson, S.B., Stoltz, D.B., and Summers, M.D. (1981), ‘Virus in a : Suppression of the Cellular Immune Response in the Parasitoid’s Host’, Science, 211, 582583. Hoch, G., and Schopf, A. (2001), ‘Effects of Glyptapanteles liparidis (Hym.: Braconidae) Parasitism, Polydnavirus, and Venom on Development of Microsporidia Infected and Uninfected Lymantria dispar (Lep.: Lymantriidae) Larvae’, Journal of Invertebrate Pathology, 77, 3743. Hoch, G., Schopf, A., and Maddox, J.V. (2000), ‘Interactions between an Entomopathogenic Microsporidium and the Endoparasitoid Glyptapanteles liparidis within their Host, the Gypsy Moth Larva’, Journal of Invertebrate Pathology, 75, 5968. Hoch, G., Schafellner, C., Henn, M.W., and Schopf, A. (2002), ‘Alterations in Carbohydrate and Fatty Acid Levels of Lymantria dispar Larvae Caused by a Microsporidian Infection Downloaded By: [Hendrichs, Jorge] At: 16:22 4 November 2009 and Potential Adverse Effects on a Co-Occurring Endoparasitoid, Glyptapanteles liparidis’, Archives of Insect Biochemistry and Physiology, 50, 109120. Hoch, G., Solter, L.F., and Schopf, A. (2004), ‘Hemolymph Melanization and Alterations in Hemocyte Numbers in Lymantria dispar Larvae Following Infections with Different Entomopathogenic Microsporidia’, Entomologia Experimentalis et Applicata, 113, 7786. Krell, P.J. (1991), ‘Polydnaviridae’,inAtlas of Invertebrate Viruses , eds. J.R. Adams and J.R. Bonami, Boca Raton, FL: CRC Press, pp. 321338. Lavine, M.D., and Beckage, N.E. (1995), ‘Polydnaviruses: Potent Mediators of Host Insect Immune Dysfunction’, Parasitology Today, 11, 368378. Maddox, J.V., and Solter, L.F. (1996), ‘Long-Term Storage of Viable Microsporidian Spores in Liquid Nitrogen’, Journal of Eukaryotic Microbiology, 43, 221225. Maddox, J.V., Baker, M.D., Jeffords, M.R., Kuras, M., Linde, A., Solter, L.F., McManus, M.L., Vavra, J., and Vossbrinck, C.R. (1999), ‘Nosema portugal, n.sp., Isolated from Gypsy Moths (Lymantria dispar L.) Collected in Portugal’, Journal of Invertebrate Pathology, 73, 114. Nakamatsu, Y., Gyotoku, Y., and Tanaka, T. (2001), ‘The Endoparasitoid Cotesia kariyai (Ck) Regulates the Growth and Metabolic Efficiency of Pseudaletia separata Larvae by Venom and Ck Polydnavirus’, Journal of Insect Physiology, 47, 573584. 42 G. Hoch et al.

Schafellner, C., Marktl, R.C., and Schopf, A. (2007), ‘Inhibition of Juvenile Hormone Esterase Activity in Lymantria dispar (Lepidoptera, Lymantriidae) Larvae Parasitized by Glyptapanteles liparidis (Hymenoptera, Braconidae)’, Journal of Insect Physiology, 53, 858 868. Shelby, K.S., and Webb, B.A. (1999), ‘Polydnavirus-Mediated Suppression of Insect Immunity’, Journal of Insect Physiology, 45, 507514. Soller, M., and Lanzrein, B. (1996), ‘Polydnavirus and Venom of the Egg-Larval Parasitoid Chelonus inanitus (Braconidae) Induce Developmental Arrest in the Prepupa of its Host Spodoptora littoralis (Noctuidae)’, Journal of Insect Physiology, 42, 471481. Solter, L.F., and Maddox, J.V. (1998), ‘Physiological Host Specificity of Microsporidia as an Indicator of Ecological Host Specificity’, Journal of Invertebrate Pathology, 71, 207216. Stoltz, D.B., and Guzo, D. (1986), ‘Apparent Haemocytic Transformations Associated with Parasitoid-Induced Inhibition of Immunity in Malacosoma disstria Larvae’, Journal of Insect Physiology, 32, 377388. Strand, M.R., and Pech, L.L. (1995), ‘Immunological Basis for Compatibility in Parasitoid- Host Relationships’, Annual Review of Entomology, 40, 3156. Tillinger, N.A., Hoch, G., and Schopf, A. (2004), ‘Effects of Parasitoid Associated Factors of the Endoparasitoid Glyptapanteles liparidis (Hymenoptera: Braconidae)’, European Journal of Entomology, 101, 243249. Washburn, J.O., Kirkpatrick, B.A., and Volkman, L.E. (1996), ‘Insect Protection against Viruses’, Nature, 383, 767. Washburn, J.O., Haas-Stapleton, E.J., Tan, F.F., Beckage, N.E., and Volkman, L.E. (2000), ‘Co-infection of Manduca sexta Larvae with Polydnavirus from Cotesia congregata Increases Susceptibility to Fatal Infection by Autographa californica M Nucleopolyhedro- virus’, Journal of Insect Physiology, 46, 179190. Downloaded By: [Hendrichs, Jorge] At: 16:22 4 November 2009 Biocontrol Science and Technology, Vol. 19, S1, 2009, 43Á48

Use of gamma radiation for improving the mass production of Trichogramma chilonis and Chrysoperla carnea Muhammad Hamed*, Sajid Nadeem, and Asia Riaz

Nuclear Institute for Agriculture & Biology (NIAB), Faisalabad, Pakistan

Gamma radiation studies were conducted to improve the existing mass rearing capabilities for an egg parasitoid, Trichogramma chilonis Ishii and a predator, Chrysoperla carnea Stephens as part of a program to achieve an area-wide control of cotton and sugarcane pests. The suitability of host (Sitotroga cerealella Olivier) eggs for parasitization by T. chilonis was restored and prolonged from 3 to 7 days pre-hatching with the application of gamma radiation in the range of 5Á55 Gy. The findings indicated that all treatment doses were effective but varied significantly in parasitization of eggs. This effect was similar during the first 2 days ranging from 78 to 94% parasitization, but it decreased drastically at lower doses (5 and 15 Gy) in succeeding days. High doses were better than low doses. For the highest treatment dose (55 Gy), successful parasitization declined only gradually, resulting in 45% parasitization even after 7 days. This finding would allow for reliable supply of viable host eggs to small insectaries in remote areas. Studies on C. carnea showed that feeding of irradiated prey eggs increased larval survival, fecundity and female sex ratio. Larval survival improved by 89% over the control when C. carnea was fed eggs irradiated with a dose of 45 Gy. The effects of radiation-treated prey eggs on survival and fecundity of C. carnea persisted in successive generations, but were considerably less in the F2 than in the F1 and P generations. At 45 Gy, fecundity was highest with 444 eggs/female in the parent, 397 in F1, and 311 in F2 generations, whereas it decreased significantly at lower doses and in untreated eggs. Keywords: gamma radiation; Trichogramma chilonis Ishii; Chrysoperla carnea Stephens; Sitotroga cerealella Olivier; viability; fecundity; sex ratio; generations Downloaded By: [Hendrichs, Jorge] At: 15:45 6 November 2009 Introduction The egg parasitoid, Trichogramma chilonis Ishii (Hymenoptera: Trichogrammatidae) and the predator, Chrysoperla carnea Stephens (Neuroptera: Chrysopidae) are prevalent and widely used in Pakistan in biological control (Cheema, Muzaffar, and Ghani 1980; Mohyuddin et al. 1997). Rearing of these parasitoids and predators has been done mostly on eggs of the , Sitotroga cerealella Olivier, because this host is easy and inexpensive to produce (Marston and Ertle 1973; Mark, Jenings, Welty, and Southard 1983). An essential aspect improving the economics of the mass production and area-wide release of beneficial insects is to use some means to extend the shelf-life of host eggs to ensure their continuous supply. It is known that suitability of host eggs for parasitization by Trichogramma species is decreased significantly with the increase in their age (Reznik and Umarova 1985;

*Corresponding author. Email: [email protected]

First Published Online 17 October 2008 ISSN 0958-3157 print/ISSN 1360-0478 online # 2009 Taylor & Francis DOI: 10.1080/09583150802433846 http://www.informaworld.com 44 M. Hamed et al.

Somchoudhury and Dutt 1989; Ruberson and Kring 1993; Takada, Kawamura, and Tanaka 2000). Therefore, a sufficient number of fresh eggs should be preserved to use as hosts for production of parasitoids. A variety of options may be used for egg preservation, including gamma radiation, cold storage, liquid nitrogen storage, etc. (Smith 1996). The early hatching of fresh host eggs during transportation to small insectaries at outstations highlighted the need for a critical evaluation of the effect of gamma radiation on the hatching of hosts and their acceptance and suitability for parasitization. The other objective of using radiation was to improve the production capacity of the predator Chrysoperla carnea under mass rearing conditions.

Materials and methods The egg parasitoid T. chilonis, the predator C. carnea, and their factitious host S. cerealella were obtained from stock cultures maintained separately in small insectaries for running mass rearing laboratories. Laboratory conditions were set at 27918C, 6595% RH and 14 L: 10 D h period. To determine the percent T. chilonis parasitization, freshly laid host eggs (ageB12 h) were glued on cards, each with 50 eggs. Immediately thereafter, these eggs were given gamma radiation doses of 5, 15, 25, 35, 45 or 55 Gy through a cobalt-60 gamma irradiator with a dose rate of 872 Gy/ h (Brower and Tilton 1973; Diop, Marchioni, Doudou, and Hasselman 1973). Three replicates were made for each dose according to methods by Zhengdong, Ping, Zulin, and Jianqing (1993) and Brower, Tilton, and Cogburn (1971). The host eggs of each dose were exposed, starting at the same day, to two male and two female newly emerged parasitoids in glass jars (117 cm) with honey streaks inside and covered on top with muslin cloth. This procedure was continued daily for all treatments up to 7 days with 1 day older eggs each day. After 24-h exposure to T. chilonis, the parasitized eggs on the cards were taken out of jars and placed in Petri dishes for incubation. After complete development of parasitoids, the host eggs were examined under a binocular microscope to determine percent parasitization. For experiments on C. carnea, 1-day-old untreated or irradiated (5, 15, 25, 35 and 45 Gy) S. cerealella host eggs were fed to the larvae in parent, F1 and F2 generations (0.1 g eggs/day per larva), each in a separate glass vial (7.71.3 cm) till pupation. Downloaded By: [Hendrichs, Jorge] At: 15:45 6 November 2009 Five replicates were made for each radiation treatment with 25 larvae per replicate. Observations were taken on percent survival of larvae, fecundity and sex ratio. Data were subjected to least significant difference, and one- and a three-way factorial analysis of variance (ANOVA) with Duncan’s multiple range tests. The associated statistics are presented in the legends of the appropriate figures.

Results and discussion Trichogramma studies The percent parasitization by T. chilonis of gamma irradiated host eggs varied significantly (for day 1, F8.0, P0.0007; day 2, F5.721, P0.0034; day 3, F119.311, P0.00; day 4, F35.161, P0.00; day 5, F41.740, P0.00; day 6, F68.346, P0.00; day 7, F94.275, P0.00 and df6 for all treatments) among radiation doses and age at which host eggs were exposed for parasitization (Table 1). The untreated (control) eggs were most heavily parasitized on the first Biocontrol Science and Technology 45

Table 1. Comparison between means (9SE) for daily parasitization (%) by T. chilonis on un-irradiated (control) and irradiated host (S. cerealella) eggs of increasing age in successive 24-hour periods up to the seventh day against radiation doses.

Day of parasitization

Radiation Doses (Gy) 1 2 3 4 5 6 7

Control 94.790.66 a 86.791.76 ab 78.091.15 a 0.00 0.00 0.00 0.00 5 92.091.15 a 84.790.66 b 33.391.76 d 30.093.05 c 28.091.15 d 25.390.66 d 16.091.15 d 15 94.091.15 a 90.792.90 a 44.093.05 c 42.093.05 b 40.093.05 c 42.091.15 bc 18.091.15 d 25 92.092.31 a 88.091.15 ab 80.091.15 a 60.092.31 a 52.091.15 b 45.391.76 b 34.791.76 c 35 92.791.57 a 87.391.76 ab 70.091.15 b 65.392.90 a 50.791.76 b 40.092.00 c 40.091.76 b 45 92.790.66 a 86.091.15 ab 69.391.33 b 60.091.15 a 54.091.15 ab 56.091.15 a 44.791.76 a 55 82.092.00 b 78.091.76 c 76.091.15a 66.791.76 a 58.091.03 a 56.091.15 a 44.790.66 a

Means in the same column sharing the same letter are statistically non-significant (df 6, PB0.05) according to Duncan’s multiple range test.

day followed by the second and third day. Thereafter, as usual, they started to hatch on the fourth day. Irradiated eggs proved somewhat suitable as hosts through the seventh day after irradiation in a dose-dependent fashion, whereas none of the untreated eggs were suitable after day 3. This finding, that irradiation extends host eggs’ suitability for parasitization and that the viability of host eggs for parasitization gradually decreases over time, agrees partly with Brower (1982), who found that irradiated host eggs of the Indian meal moth, Plodia interpunctella (Hubner) were preferred by T. pretiosum over eggs from irradiated females. Our data completely agree with the findings of Zhengdong et al. (1993) that the preservation time of Antheraea pernyi (Gue´rin-Me´neville) eggs for parasitization with Trichogramma species was remarkably extended as a result of irradiation. There is a negative correlation between the age of the eggs and their sensitivity to radiation treatment (Chand and Sehgal 1978). Seal and Tilton (1986) observed that radio-sensitivity of eggs of hide beetles decreased with increasing embryonic development. The reasons of the minimum 33 and 44% parasitism of eggs at 5 and 15 Gy in our results might be the incomplete arrest of the embryo development at day 3 and onward, whereas the increases in doses and the resulting

Downloaded By: [Hendrichs, Jorge] At: 15:45 6 November 2009 decreases in embryo development encouraged parasitism.

Chrysoperla studies The assessment of the effects of feeding the predator C. carnea on control or irradiated prey eggs in terms of larval survival (Figure 1) indicate that the percent survival to the adult stage of larvae fed treated eggs increased significantly (P0.05) in parent, F1 and F2 generations (LSD2.676). Even though the overall interactions between radiation doses and generation means were non-significant, with untreated eggs, the survival was only 56%, whereas feeding of irradiated eggs significantly enhanced larval survival. The subsequent effects of feeding on a diet of control or irradiated prey eggs on the fecundity of C. carnea for P, F92293.53, P0.00; for F1, F43896.59, P0.00; for F2, F16041.68, P0.00 and df5 for all treatments varied significantly (PB0.05) among radiation doses and generations (Table 2). Fecundity increased in the parent generation with increased radiation 46 M. Hamed et al.

Figure 1. Effect of feeding on control or irradiated prey (S. cerealella) eggs on percent survival of C. carnea larvae in the parental and two successive generations.

doses to prey eggs, whereas it decreased at 5 Gy. There was a successive decrease in

fecundity in F1 and F2 than the parent generation. However, it was comparatively high in F1 at 25 Gy and onward and in F2 at 45 Gy. The resulting C. carnea male to female sex ratio (Figure 2) was non-significant in successive generations, whereas it Downloaded By: [Hendrichs, Jorge] At: 15:45 6 November 2009 Table 2. Comparison among means (9SE) for the effect of larval feeding on irradiated prey (S. cerealella) eggs on fecundity of C. carnea in successive generations against radiation treatments.

Generations

Radiation doses (Gy) Parent F1 F2

Control 271.890.63 e 273.290.98 d 273.691.30 b 5 216.090.43 f 192.590.54 f 209.290.66 d 15 332.090.89 d 222.890.70 e 178.090.67 f 25 400.290.43 b 336.991.33 c 196.890.93 e 35 349.690.09 c 371.190.66 b 240.791.03 c 45 443.690.81 a 396.690.84 a 311.290.49 a

Means in the same column sharing similar letters are non-significant (df=5, PB0.05) according to Duncan’s multiple range test. Biocontrol Science and Technology 47

LSD value for: Generation means = Non-significant Radiation doses = 0.9797 5.0 Interaction = Non-significant

4.5 P F1 F2 4.0

3.5

3.0

2.5

2.0

Male to female ratio 1.5

1.0

0.5

0.0 0 5 15 25 35 45 Radiation doses

Figure 2. Effect of larval feeding on irradiated prey (S. cerealella) eggs on sex ratio of C. carnea in successive generations.

varied significantly (LSD0.9797) among radiation doses. The overall mean values for male to female ratios were significantly higher with 3:1 at 5 Gy followed by 2:1 at 15, 2:1 at 25 Gy and 1:1 with the control. At the highest doses (35 and 45 Gy), the sex ratio decreased even further in all generations. The overall mean values for male to female ratio for 35 and 45 Gy treated eggs were 0.9:1 and 0.3:1, respectively. In the present studies, radiation doses, fecundity and female to male ratio were directly correlated to each other. There are few studies on insects reared for several generations on irradiated diets and irradiated prey with respect to predator. Ye and Cheng (1986) reared Chrysopa sinica on UV-irradiated eggs of Corcyra cephalonica and reported that all stages of the predator developed normally for three successive generations. Brower et al. (1971) conducted research on the effects of irradiated diets on Indian meal moth on several

Downloaded By: [Hendrichs, Jorge] At: 15:45 6 November 2009 generations and found statistically non-significant but biologically significant and increasing effects on progeny, fecundity and sex ratio in successive generations. In similar studies on the rearing of three stored product insects on irradiated wheat, fecundities were slightly but consistently higher than that of the control (Hodges and Guyer 1958).

Conclusions Gamma radiation prolonged the viability of host eggs for T. chilonis parasitization up to 7 days. High doses proved better to prolong egg viability than low doses. Treatment of 55 Gy resulted in similar levels of parasitization as the control during the first 3 days, and gave 45% parasitization on the seventh day. Supply of irradiated host eggs enabled us to run insectaries in remote areas with low rearing cost and fulfill the requirement of area-wide releases of T. chilonis. Feeding of irradiated prey eggs (45-Gy dose) to C. carnea increased significantly the percent larval survival (89%), fecundity 48 M. Hamed et al.

(444 eggs/female) and female to male ratio (1:0.5) in the parental generation. The effect on some of these parameters was less pronounced in the F2 generation.

References Brower, J.H. (1982), ‘Parasitization of Irradiated Eggs and Eggs from Irradiated Adults of the Indian Meal Moth (Lepidoptera: Pyralidae) by Trichogramma pretiosum (Hymenoptera: Trichogrammatidae)’, Journal of Economic Entomology, 75, 939Á944. Brower, J.H., and Tilton, E.W. (1973), ‘Comparative Gamma Radiation Sensitivity of Tribolium madens (Charpentier) and T. castaneum (Herbst)’, Journal of Stored Product Research,9,93Á100. Brower, J.H., Tilton, E.W., and Cogburn, R.R. (1971), ‘Effects of Irradiated Diets on Production of Progeny by Successive Generations of the Indian Meal Moth, Plodia interpunctella (Hubner)’, Radiation Research, 48, 283Á290. Chand, A.T., and Sehgal, S.S. (1978), ‘Sensitivity of Gamma Radiation of Corcyra cephalonica Eggs Related to Their Age’, Entomon,3,7Á9. Cheema, M. A., Muzaffar, N., and Ghani, M.A. (1980), ‘Investigation on Phenology, Distribution, Host Range and Evaluation of Predators of Pectinophora gossypiella (Sunders) in Pakistan’, The Pakistan Cotton, 24, 140Á176. Diop, Y.M., Marchioni, E., Doudou, B.A., and Hasselman, C. (1973), ‘Radiation Disinfesta- tion of Cowpea Seeds Contaminated by Callosobruchus maculates’, Journal of Food Processing and Preservation, 21, 69Á81. Hodges, R., and Guyer, G. (1958), ‘The Effects of an Irradiated Wheat Diet on the Confused Flour Beetle, Granary Weevil, and the Angoumois Grain Moth’, Journal of Economic Entomology, 51, 674Á675. Mark, W.H., Jenings, D.T., Welty, E., and Southard, S.G. (1983), ‘Progeny Production of Trichogramma minutum, Utilizing Eggs of Choristoneura fumiferana and Sitotroga cerealella’, Canadian Entomologist, 115, 1245Á1252. Marston, N., and Ertle, L.R. (1973), ‘Host Influences on the Bionomics of Trichogramma minutum’, Annals of the Entomological Society of America, 66, 1155Á1162. Mohyuddin, A.I., Gilani, G., Khan, A.G., Hamza, A., Ahmad, I., and Mahmood, Z. (1997), ‘Integrated Pest Management of Major Cotton Pests by Conservation, Redistribution and Augmentation of Natural Enemies’, Pakistan Journal of Zoology, 29, 293Á298. Reznik, S. Ya., and Umarova, T. Ya. (1985), ‘The Reaction of Females of Trichogramma cacoeciae (Hymenoptera, Trichogrammatidae) to the Duration of Development of the Eggs of the Host’, Zoologicheskii Zhurnal, 64, 709Á714. Ruberson, J.R., and Kring, T.J. (1993), ‘Parasitism of Developing Eggs by Trichogramma pretiosum (Hymenoptera: Trichogrammatidae): Host Age Preference and Suitability’, Biological Control,3,39Á46. Downloaded By: [Hendrichs, Jorge] At: 15:45 6 November 2009 Seal, D.R., and Tilton, E.W. (1986), ‘Effect of Gamma Radiation on the Metamorphic Stages of Dermestes maculates DeGeer (Coleoptera: Dermestidae)’, International Journal of Radiation and Applied Instruments [A], 37, 531Á535. Smith, S.M. (1996), ‘Biological Control with Trichogramma: Advances, Successes, and Potential of Their Use’, Annual Review of Entomology, 41, 375Á406. Somchoudhury, A.K., and Dutt, N. (1989), ‘Influence of Hosts and Hostages on the Bionomics of Trichogramma perkinsi Girault and Trichogramma australicum Girault’, Indian Journal of Entomology, 50, 374Á379. Takada, Y., Kawamura, S., and Tanaka, T. (2000), ‘Biological characteristics: Growth and Development of the Egg Parasitoid Trichogramma dendrolimi (Hymenoptera: Trichogram- matidae) on the Cabbage Armyworm Mamestra brassicae (Lepidoptera: Noctuidae)’, Applied Entomolohy and Zoology, 35, 369Á379. Ye, Z.C., and Cheng, D.F. (1986), ‘Rearing the Green Lacewing, Chrysopa sinica on Ultraviolet-Irradiated Eggs of the Rice Moth’, Chinese Journal of Biological Control,2, 133Á134. Zhengdong, D.L.W., Ping, J., Zulin, C., and Jianqing, S. (1993), ‘Radiation Preservation Study on Middle Host Eggs of Trichogramma Species’, Radiation Physics and Chemistry, 42, 647Á 650. Biocontrol Science and Technology, Vol. 19, S1, 2009, 49Á79

Colonization and domestication of seven species of native New World hymenopterous larval-prepupal and pupal fruit fly (Diptera: Tephritidae) parasitoids Martı´n Alujaa*, John Sivinskib, Sergio Ovruskic, Larissa Guille´na, Maurilio Lo´peza, Jorge Cancinod, Armando Torres-Anayaa, Guadalupe Gallegos-Chana and Lı´a Ruı´zd

aInstituto de Ecologı´a, A.C., Xalapa, Veracruz, Me´xico; bCenter for Medical, Agricultural and Veterinary Entomology, USDA-ARS, Gainesville, FL, USA; cPlanta Piloto de Procesos Industriales Microbiolo´gicos y Biotecnologı´a (PROIMI), Divisio´n Control Biolo´gico de Plagas, San Miguel de Tucuma´n, Argentina; dSubdireccio´n de Desarrollo de Me´todos, Campan˜a Nacional Contra Moscas de la Fruta, Tapachula, Chiapas, Me´xico

We describe the techniques used to colonize and domesticate seven native New World species of hymenopterous parasitoids that attack flies within the genus Anastrepha (Diptera: Tephritidae). All parasitoid species successfully developed on artificially reared Mexican fruit fly, Anastrepha ludens (Loew) larvae or pupae. The parasitoid species colonized were the following: Doryctobracon areolatus (Sze´pligeti), Doryctobracon crawfordi (Viereck), Opius hirtus (Fischer), Utetes anastrephae (Viereck) (all Braconidae, Opiinae), Aganaspis pelleranoi (Bre´thes) and Odontosema anastrephae Borgmeier (both Figitidae, Eucoilinae) (all larval- pupal parasitoids), and the pupal parasitoid Coptera haywardi (Ogloblin) (Diapriidae, Diapriinae). We provide detailed descriptions of the different rearing techniques used throughout the domestication process to help researchers elsewhere to colonize local parasitoids. We also describe handling procedures such as number of hosts in parasitization units and compare optimal host and female age, differences in parasitism rate, developmental time, life expectancy and variation in sex ratios in each parasitoid species over various generations. In the case of D. crawfordi and C. haywardi we also provide partial information on mass- rearing techniques such as cage type, parasitization unit, larval irradiation dose

Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009 and adult handling. Keywords: hymenoptera; Braconidae; Figitidae; Diapriidae; Tephritidae; Anastrepha; biological control; parasitoids; rearing

Introduction Historically the release of exotic (i.e. non-native) parasitoid species to deal with fruit fly pests has been the norm (Wharton 1989; Aluja 1994; Purcell 1998; Ovruski, Aluja, Sivinski, and Wharton 2000). In comparison, native parasitoids of indigenous pestiferous species have received little attention except for systematic studies and surveys of parasitoids of flies in the economically important genera Anastrepha (e.g. Wharton, Gilstrap, Rhode, Fischel, and Hart 1981; Aluja et al. 1990, 2003; Katiyar, Camacho, Geraud, and Matheus 1995; Lo´pez, Aluja, and Sivinski 1999; Canal and

*Corresponding author. Email: [email protected]

First Published Online 10 October 2008 ISSN 0958-3157 print/ISSN 1360-0478 online # 2009 Taylor & Francis DOI: 10.1080/09583150802377373 http://www.informaworld.com 50 M. Aluja et al.

Zucchi 2000; Ovruski, Schliserman, and Aluja 2004), Bactrocera (e.g. Wharton and Gilstrap 1983) and Rhagoletis (e.g. Wharton and Marsh 1978; AliNiazee 1985; Hoffmeister 1990; Gut and Brunner 1994; Feder 1995). The unstated perception has perhaps been that the long-standing co-existence of native parasitoids with flies that have remained pests was evidence that they were unable to exert economically significant levels of control. However, recent interest in the augmentative release of parasitoids (e.g. Sivinski et al. 1996; Purcell 1998; Montoya et al. 2000), with the possibility of strategically increasing the mortality inflicted by native species (Sivinski, Aluja, and Lo´pez 1997; Lo´pez et al. 1999; Sivinski, Pin˜ero, and Aluja 2000), has given new impetus to studies of their colonization and mass rearing. Around 205 species of the Neotropical genus Anastrepha have been described to date (Norrbom 2004). In Mexico, 37 species have been reported to date (Herna´ndez- Ort´ız and Aluja 1993; Herna´ndez-Ort´ız 1998, 2004; Herna´ndez-Ort´ız, Manrique- Sade, Delf´ın-Gonza´lez, and Novelo-Rinco´n 2002) and the larvae and/or pupae of these species are hosts for a diversity of parasitoids (Aluja et al. 1990, 2003; Herna´ndez-Ort´ız et al. 1994; Lo´pez et al. 1999). Species such as Diachasmimorpha longicaudata (Ashmead), Psyttalia incisi (Silvestri), P. concolor (Sze´pligeti), Fopius arisanus (Sonan), F. vandenboschi (Fullaway), Aceratoneuromyia indica (Silvestri) and Pachycrepoideus vindemiae (Rondani) were introduced into Mexico as biological control agents, beginning in 1954 in an attempt to curb populations of the Mexican fruit fly, Anastrepha ludens (Loew) (Jime´nez-Jime´nez 1961; Wharton 1989; Ovruski et al. 2000). With similar intentions, non-native parasitoids were released in El Salvador, Nicaragua, Costa Rica, Panama´, Colombia, Peru´, Brazil and Argentina (Wharton, Gilstrap, Rhode, Fischel, and Hart 1981; Ovruski et al. 2000). However, despite the large numbers of individuals introduced, few parasitoid species have successfully established (Ovruski et al. 2000). The shortcomings of this handful of exotic species has turned attention to the many native parasitoid candidates for augmentative release. Their diversity suggests that suitable species would be available for programs faced with an assortment of pests occurring in a variety of environments (Sivinski et al. 1997; Aluja, Lo´pez, and Sivinski 1998; Sivinski and Aliya 2003). To facilitate native parasitoid colonization efforts in other parts of the world, we describe the colonization and domestication of the following seven native Anastrepha Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009 parasitoids found in Mexico and various other countries in Latin America (in some cases the US) (updates on exact distribution can be found in Ovruski et al. 2000; Ovruski, Wharton, Schliserman, and Aluja 2005): Doryctobracon areolatus (Sze´- pligeti), Doryctobracon crawfordi (Viereck), Opius hirtus (Fischer), Utetes anastre- phae (Viereck) (all Braconidae, Opiinae), Aganaspis pelleranoi (Bre`thes) and Odontosema anastrephae Borgmeier (both Figitidae, Eucoilinae) (all larval-prepupal parasitoids), and the pupal parasitoid Coptera haywardi (Ogloblin) (Diapriidae, Diapriinae). Recent findings on the biology, ecology, and behavior of the latter parasitoid species have been reported by Sivinski (1991), Sivinski et al. (1996, 1997, 2000), Sivinski, Aluja, Holler, and Eitam (1998a), Sivinski, Vulinec, Menezes, and Aluja (1998b), Sivinski, Aluja, and Holler (1999), Sivinski, Vulinec, and Aluja (2001), Aluja et al. (1998), Aluja et al. (2003), Lo´pez et al. (1999), Guille´n, Aluja, Equihua, and Sivinski (2002), Eitam, Holler, Sivinski, and Aluja (2003), Eitam, Sivinski, Holler, and Aluja (2004), Ovruski and Aluja (2002), Ovruski et al. (2004), Ovruski et al. (2005) and Guimara˜es and Zucchi (2004). Biocontrol Science and Technology 51

The most common and widely distributed Anastrepha native parasitoid species in the Neotropics and subtropics is D. areolatus (Ovruski et al. 2000). It is a larval- prepupal braconid parasitoid and broadly distributed from Mexico to Argentina (Wharton and Marsh 1978). When introduced into Florida in 1969, it became one of the most common parasitoids of A. suspensa (Loew) (Sivinski et al. 1998a; Eitam et al. 2004). In Mexico, D. areolatus and U. anastrephae are among the most numerous native species parasitizing larvae of A. obliqua (Macquart), a fruit fly that is an economically important pest of mango (Mangifera indica L.) and tropical plum (Spondias purpurea L.) (Aluja et al. 1996). Utetes anastrephae is also a larval-prepupal braconid parasitoid, but in comparison to D. areolatus, has the shortest ovipositor of any of the braconids sampled (Sivinski et al. 2001; Sivinski and Aluja 2003). This parasitoid species occurs naturally from Florida to Argentina (Ovruski et al. 2000). Doryctobracon crawfordi is a larval-prepupal opiine parasitoid commonly associated with A. ludens (Plummer and McPhail 1941; Lo´pez et al. 1999). Reported for the first time by L. de la Barrera (see Herrera 1905), this species apparently prefers more temperate climates (Aluja et al. 1998) and higher altitudes (Sivinski et al. 2000). Opius hirtus is another larval-prepupal parasitoid that commonly attacks the relatively rare A. cordata Aldrich in Tabernaemontana alba Mill. (Apocynaceae) (Herna´ndez-Ort´ız et al. 1994). It has also been reported attacking A. obliqua in Tapirira mexicana Marchand and Spondias mombin L (both Anacardiaceae) (Herna´ndez-Ort´ız et al. 1994; Sivinski et al. 2000), A. alveata Stone in Ximenia americana L. (Olacaceae) (Lo´pez et al. 1999), Toxotrypana curvicauda Gerstaecker and Ceratitis capitata (Wiedemann) (Wharton 1983). Aganaspis pelleranoi and O. anastrephae are two figitid larval-prepupal parasitoids that gain access to A. striata Schiner and A. fraterculus (Wiedemann) in guavas through wounds or holes in the fruit (Ovruski 1994; Sivinski et al. 1997; Ovruski et al. 2004). A. pelleranoi is more widely distributed and has a broader host range than O. anastrephae (Wharton, Ovruski, and Gilstrap 1998). One native, pupal endoparasitoid that has potential for fruit fly biological control is C. haywardi (Baeza-Larios, Sivinski, Holler, and Aluja 2002a; Guille´n et al. 2002). It was originally discovered in Argentina attacking A. fraterculus and A. schultzi Blanchard pupae (Loia´cono 1981). In 1994, C. haywardi was found in Veracruz, Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009 Mexico, attacking A. ludens pupae (Lo´pez et al. 1999). More recently, this diapriine species was recovered from A. striata and A. serpentina (Wiedemann) pupae in Venezuela (Garc´ıa and Montilla 2001) and from A. fraterculus and A. sororcula Zucchi pupae in Brazil (Aguiar-Menezes, Menezes, and Loia´cono 2003). Unlike many other pupal parasitoids of Diptera, it has a relatively restricted host range and is known only to parasitize Tephritidae (Sivinski et al. 1998b).

Materials and methods Source of insects In every case with the exception of C. haywardi, we obtained parasitoids by harvesting mature fruit from the tree or retrieving fallen fruit from the ground and transporting it to our laboratories in Xalapa, Veracruz, where they were processed following the methods described in Aluja et al. (1998), Lo´pez et al. (1999) and Sivinski et al. (2000). In the case of C. haywardi, specimens stemmed from pupae that 52 M. Aluja et al.

Table 1. Location and host plant from which the individuals stemmed that were used to establish the first successful colonies.

Locality Host plant Fruit fly host Parasitoid species

Llano Grande1 and Tejer´ıa2, Spondias mombin L. Anastrepha Doryctobracon areolatus Municipality of Teocelo, (Anacardiaceae) obliqua Utetes anastrephae State of Veracruz, Mexico A. obliqua Coptera haywardi pupae Psidium guajaba L. A. fraterculus D. crawfordi, Aganaspis (Myrtaceae) and/or pelleranoi, Odontosema A. striata anastrephae Citrus sinensis L. A. ludens D. crawfordi (Rutaceae) La Mancha3, Santiago P. guajaba L. A. fraterculus O. anastrephae Tuxtla4 and San Andre´s (Myrtaceae) and/or Tuxtla5, State of Veracruz, A. striata Mexico Vicinity of Tapachula6, Ximenia americana L. A. alveata O. anastrephae State of Chiapas, Mexico (Olacaceae) Playa Escondida7 and Tabernaemantana A. cordata Opius hirtus Sontecomapan8, Los alba Mill. larvae Tuxtlas, State of Veracruz, (Apocynaceae) Mexico

1(19822’08’’ N, 96851’57’’ W), 2(19822’07’’ N, 96854’59’’ W), 3(19835’23’’ N, 96822’49’’ W), 4(18828’31’’ N, 95818’40’’ W), 5(18826’42’’ N, 95811’53’’ W), 6(14854’21’’ N, 92815’33’’ W), 7(18836’47’’ N, 95803’45’’ W), 8(18825’07’’ N, 95812’48’’ W).

were collected underneath fruit naturally infested in the field or from lab reared pupae artificially exposed to parasitization in the field (details in Lo´pez et al. 1999). Details on fruit fly (fruit) and parasitoid host (fruit fly larvae) species and the geographical location where the specimens for founding the colonies were collected are provided in Table 1.

Laboratory conditions

Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009 During the initial phases of the colonization and domestication processes, we maintained parasitoid colonies at the Fruit Fly and Parasitoid Laboratory of the Instituto de Ecolog´ıa, A.C., Xalapa, Me´xico, at 25918C, 7095% RH, and a photoperiod of 12:12 h. Over time (i.e. several years of observations), much insight into the particular idiosyncrasies of each species was gained, and as a result, we moved established colonies of D. crawfordi and C. haywardi into a laboratory maintained at a lower temperature (23928C). As previously noted, both species are common in areas above 800 m, with lower year-round temperatures. All the other species, typically found in warmer climates, were maintained in laboratories at 259 18C. A separate laboratory, kept at 27918C, 7095% RH, 12:12 h photoperiod) was used to rear A. ludens adults, while larvae and pupae were kept at 30918C, 7595% RH in an additional room without light (i.e., full darkness). This species was used as a host for all the parasitoid species. Yet another laboratory was used to mass-rear D. crawfordi in Metapa de Dom´ınguez, Chiapas (24928C, 70910% RH, 12:12 h photoperiod). Biocontrol Science and Technology 53

Rearing of A. ludens larvae as parasitoid hosts Our A. ludens strain was originally provided by the Comite´ Estatal de Sanidad Vegetal (DGSV-SAGARPA) in Xalapa, Veracruz, where it had been kept for over 200 generations. We placed 200 mL of A. ludens pupae in 303060-cm Plexiglas cages. Between 2,500 and 3,000 adults emerged 1Á2 days later and were fed ad libitum with a mixture of hydrolyzed protein (Greif Bros. Corporation, Delaware, OH) and locally available refined sugar (no particular brand). Water was provided ad libitum by using 300-mL plastic bottles with a cotton wick. After 8 days, flies were provided with an artificial oviposition medium placed inside the cage, which originally consisted of a 10-cm dome-like, hollow, dark green hemisphere made of green cheesecloth (dyed with commercial fabric dye (Mariposa†, Colorantes Importados, S.A. de C.V., Me´xico D.F., Mexico) and paraffin (McPhail and Guiza 1956). This oviposition device was later replaced by a 12-cm diameter Petri-type plastic dish covered with green linen cloth and filled with transparent silicon or ‘fuseleron’ (Devcon†, Junta Flex, ITW Poly Mex SA de CV, Mexico). The plastic dish was placed upside down on top of the fly-holding cage so that females could insert their aculeus through the cloth and lay eggs into the ‘fuseleron’. Once flies reached 8 days of age, eggs were collected daily over an 8-day period and washed in a solution of 2 g of sodium benzoate (Baker, J.T. Baker S.A. de C.V., Xalostoc, Edo. de Me´xico) dissolved in 1 L of purified water. After washing, eggs were placed on pieces of filter paper (Whatman No. 1, Whatman Int., Ltd., Maidstone, England) in Petri dishes, incubated for 4 days and then placed (2 mL per unit) in a 112632-cm plastic washbowl containing an artificial diet (ingredients in Appendix 1). Once the desired larval stage was reached (2nd and 3rd stage depending on parasitoid species), exposure to parasitoids was carried out according to the technique used for each particular species (details follow). In the particular cases of the D. crawfordi and C. haywardi strains sent from Xalapa, Veracruz to the Laboratorios de Desarrollo de Me´todos, Campan˜a Nacional Contra Moscas de la Fruta in Metapa de Dom´ınguez, Chiapas, Mexico for mass-rearing purposes, parasitoids were exposed to irradiated A. ludens larvae (pupae in case of C. haywardi) produced locally (Dom´ınguez, Herna´ndez, and Castellanos 2002). For D. crawfordi we used larvae irradiated at 40 Gy and in the

Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009 case of C. haywardi, irradiation dose for pupae was 30 Gy (Cancino, Ruiz, Sivinski, Ga´lvez, and Aluja 2008). Since irradiated larvae support parasitoid development but do not mature into fertile flies, removal of unattacked hosts from the colony is greatly simplified (Sivinski and Smittle 1990). Larvae (32,000) and pupae (25,000) were placed in 1-L containers and irradiated, in an atmosphere containing oxygen, using a Gammacell irradiator with a cobalt-60 source (Cancino et al. in press) located at the Medfly mass rearing facility in Metapa de Dom´ınguez, Chiapas.

Cages for holding parasitoids Various sizes of Plexiglas cages, covered with fiberglass and aluminum screen, were used to house parasitoids. Screen mesh size and cage size depended on the size of the parasitoid species kept inside (details in Table 2). In the case of Plexiglas cages, one side of each cage was covered with plastic wrap (Kleen Pack†; Kimberly Clark de Me´xico S.A. de C.V.) held in place by three strips of masking tape (Shurtape†, Table 2. Summary of rearing procedures and handling conditions used during the domestication and colonization of seven native Anastrepha 54 parasitoid species (all parasitoid colonies were maintained at 25918C, 7095% RH, 12:12 h photoperiod) (see Figures 1Á4 for further details on rearing cages and parasitization devices such as FF, SD and M-PD).

No. of parasitoids per rearing cage

Rearing Type of parasitization Host No. of exposed hosts Plexiglas cage Host stage Host age devices (and No. exposure per parasitoid female Species1 size Female Male attacked (days) hosts per unit) periods (h) and per hour

Doryctobracon 3030301,2 30 15 Larva 8 Fruit filled with guava 36 0.05 larvae crawfordi Á FF (50) Sandwich- type oviposition device one Á SD1 36 0.23 larvae Aluja M. (250) Sandwich type oviposition Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009 6 November 15:46 At: Jorge] [Hendrichs, By: Downloaded device two Á SD2 7 19 larvae tal et (250) 1 D. areolatus 252525 30 15 Larva 8 FF (50) 36 0.05 larvae . SD1 (250) 36 0.23 larvae Utetes anastrephae 2525251 40 20 Larva 7 Á 8 FF (50) 48 0.03 lavae Modified Petri dish Á 24 0.26 larvae M-PD (250) SD2 (250) 7 0.91 larvae Opius hirtus 3030601 40 20 Larva 8 FF (50) 36 0.04 larvae SD1 (250) 24 0.26 larvae SD2 (250) 7 0.91 larvae Aganaspis 3030301 30 15 Larva 9 Uncovered Petri dish 24 0.35 larvae pelleranoi Á UP (250) UP (250) 7 1.19 larvae Odontosema 3030301 30 15 Larva 9 UP (250) 24 0.35 larvae anastrephae Table 2 (Continued) Technology and Science Biocontrol No. of parasitoids per rearing cage

Rearing Type of parasitization Host No. of exposed hosts Plexiglas cage Host stage Host age devices (and No. exposure per parasitoid female Species1 size Female Male attacked (days) hosts per unit) periods (h) and per hour Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009 6 November 15:46 At: Jorge] [Hendrichs, By: Downloaded Coptera haywardi 3030301 30 15 Pupa 1 Á 2 Covered pupae Á CP 168 0.10 pupae (500) 125 125 Pupa 2 Naked pupae Á NP 72 0.09 pupae (800)

1The fiberglass screen that covered the cage frame had a 0.3-mm mesh size. 2When D. crawfordi was mass-reared (details in text) we used an aluminum cage frame covered with a metallic screen (1-mm mesh size). 55 56 M. Aluja et al.

0.30 M 0.40 M

0.3

0 M

0.25 M 0.055M

0.14 M

0.11M

0.165 M 0.24 M

Figure 1. ‘Metapa‘ cage used in initial D. crawfordi mass-rearing efforts. ‘Cassette-type’ oviposition units (compact disk cases) filled with larvae (2000 third instar larvae mixed with a small amount of rearing diet) were slid into cage openings in walls. Each cage contained 1,500 parasitoids.

Shurtape Technologies, Inc., Hickory, NC). A 150-mL container holding one or two orange, mango or guava (depending on availability) branches with five to eight leaves each, was placed in every cage to provide resting sites and adequate conditions for mating activities. In the case of C. haywardi,1010-cm pieces of black paper were used to form small (58 cm) resting shelters that were placed on cage floors (1Á2 per cage). In each clean, sealed cage, we placed a predetermined number of newly emerged males and females from a given parasitoid species (details in Table 2). For mass-rearing purposes (case of D. crawfordi), we used a 403030-cm cage with an aluminum frame, covered with a metallic mesh (1 mm) known as the ‘Metapa’ cage (Figure 1). In the cage front, there are two 151.5-cm openings that project inside of the cage by means of two 1711.5-cm hollow aluminum squares (width of 2 cm) covered with the same 1-mm metallic mesh used to cover all cage

Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009 walls. Inside the hollow squares, we slid the oviposition units, which consisted of empty compact disk cases (1051 cm, lengthwidthdepth) in which the top had been replaced by organdy cloth held tightly to the frame. Between the disk case bottom and the cloth cover, we placed 2,000 third instar A. ludens larvae mixed with some of the diet the larvae had been reared in (Figure 1). Each cage contained 1,500 parasitoids (sex ratio close to 1:1) that were allowed to parasitize larvae over a period of 4 h daily over 10 days. After this, they were replaced with a new cohort.

Feeding, and handling of adults Adults were fed with diluted honey (70% honey, 30% water) (Miel Carlota†; Herdez S.A. de C.V., Cuernavaca, Morelos, Mexico). Pieces of cotton (Zuum†; Universal Productora S.A. de C.V., Me´xico D.F.) saturated with this liquid diet were placed in Petri dishes (10 cm in diameter) and offered to the parasitoids ad libitum (see Bautista, Harris, and Vargas 2001). Food was changed on a weekly basis. Water was Biocontrol Science and Technology 57

also administered on a piece of cotton and was changed two times per week. At the same time that food and water were changed, dead parasitoids were removed from the cages to avoid problems with fungi, bacteria, mites, and other insect pathogens. To keep parasitoids from escaping the cages while maneuvering objects within them, we temporarily shut the lights in the laboratory and used a 22-W lamp to attract the parasitoids towards the light.

Diagnostic features for quick recognition of the sexes To facilitate quick recognition of the sexes, the following diagnostic features were used. In the case of braconid species, differences among the sexes were obvious because the female, besides being larger than the male, has an exerted ovipositor that is clearly visible (Sivinski et al. 2001; Sivinski and Aluja 2003). In the case of figitids, the most obvious character for identifying the sexes is the size and shape of the antenna, since the ovipositor is not apparent in females. Male antennae are filiform and 1.6Á1.8 times longer than female antennae which are moniliform (Ovruski and Aluja 2002). In the case of C. haywardi, sex can also be distinguished by clearly different antennal lengths. Female and male antennae measure, respectively (mean9 SE), 1.790.1 mm (N20) and 3.090.2 mm (N20).

General conditions for the reproduction, management and care of parasitoids Once field-collected larvae had pupated and adult parasitoids emerged, the domestication phase ensued. It initially consisted of adapting adults of each species to the artificial housing and rearing conditions associated with the laboratory. The first step was to identify and manipulate environmental conditions, such as temperature, required by each species. In addition, preliminary observations of mating and oviposition behaviors were conducted to determine which species parasitized larvae and which attacked pupae and what circumstances enhanced mating. To confirm that C. haywardi exclusively parasitized pupae (and not late third larval instars), females were offered two guavas containing 50, third instar A. ludens larvae. These fruit were removed before the larvae had pupated. At the same time, parasitoids were exposed to pupae (0Á2 days old) for 7 days (168 h). Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009

Description of oviposition units utilized to colonize each species of parasitoid We tried to fabricate the cheapest and most natural oviposition devices to entice females to accept the artificial laboratory conditions (details in Figures 2Á4). In what follows we describe the oviposition devices that worked best for us after several failed attempts.

Oviposition substrates for larval-prepupal parasitoids Fruit filled with larvae (FF). Our objective was to simulate a naturally infested fruit that would be attractive to wild parasitoids, particularly in the initial stages of the domestication process. Commercial guava (Psidium guajava) was chosen as the preferred parasitization unit because: (a) almost all species of larval-prepupal parasitoids described in this work were found parasitizing fruit fly larvae in guavas in 58 M. Aluja et al. Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009

Figure 2. Description of the ‘fruit filled with larvae (FF)’ oviposition substrate used during the initial colonization stages of larval-prepupal parasitoids. (1Á3) Cutting of fruit, with proximal quarter functioning as ‘lid’ and rest as ‘base’.(4Á6) Removal of pulp to create cavity (hollow ‘base’). (7Á8) Filling of hollowed ‘base’ with larvae mixed with diet. (9) Joining of ‘base’ and ‘lid’ with aid of 1.510-cm parafilm strip (‘belt’). (10) Pricking of holes into of fruit. (11Á12) Paper clip inserted into parafilm ‘belt’ to hang fruit from cage roof. (13) Fruit hanging from cage roof. (14) Parasitoids ovipositing in FF unit. Biocontrol Science and Technology 59 Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009

Figure 3. Modified Petri dish (M-PD) oviposition unit used to rear U. anastrephae, the parasitioid species with the shortest ovipositor (left). For comparative purposes (i.e. distinguish differences in thickness of oviposition unit), a ‘sandwich-type oviposition device’ is also shown (right).

the field (Lo´pez et al. 1999) and (b) because guava can be obtained year round in local markets and supermarkets at a reasonable price. Guavas were cut open transversally along the peduncle, about one-quarter down the length of the fruit as measured from the proximal end (Figure 2). The proximal quarter sections functioned as ‘lids’ for the filled fruits and the remainder of the fruit served as ‘bases’ for filling. Mesocarp and endocarp (pulp) were extracted in the bases to 60 M. Aluja et al. Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009

Figure 4. Preparation of the ‘sandwich-type oviposition devices (SD)’. (A) Exposure of naked larvae without fruit skin. (B) One-mm (thickness) guava epicarp (skin) pieces placed on top of chiffon cloth covering larvae placed to entice female parasitoids to land on oviposition unit and parasitize larvae.

create cavities that could be filled with larvae and diet. Guavas had to be mature (yellow and soft, but not watery) and emit the characteristic odor associated with this fruit (i.e. not sealed with wax). However, if wax residues were encountered, they were removed by gently washing the fruit with diluted soap. The optimal size for guavas was 45Á55 g and 4Á5 cm in diameter. Larger fruit typically yielded smaller Biocontrol Science and Technology 61

numbers of parasitoids because females were unable to reach larvae feeding deep within the fruit (Sivinski 1991). The short ovipositor of U. anastrephae (Sivinski et al. 2001) restricts females to parasitizing larvae in small fruit such as Spondias mombin (Lo´pez et al. 1999). As a consequence, we were forced to use small (25Á30 g and 3Á4 cm in diameter) larvae-filled guavas to colonize this species. We filled each fruit with ca. 50, laboratory-reared, second or third instar A. ludens larvae, and hung three or four guavas per rearing cage (Figure 2). Larval stage was associated to parasitoid species as described in Table 2. Once guava ‘bases’ were filled with larvae, they were covered with their corresponding ‘lids’ and the different parts tightly joined with 1.510-cm strips of parafilm (Parafilm ‘belts’) (Parafilm† Laboratory Film, American National Can Tm, Chicago, IL). Four to five holes were pricked into the fruit with a 1-mm metal needle to allow for aeration. Plastic paper clips were inserted into the Parafilm ‘belts’ to hang fruit from the cage ceilings where parasitoid density was usually highest. A variant of this technique was used in the case of O. anastrephae and A. pelleranoi, whose females prefer to enter into fruit interiors to search for fruit fly larvae (Ovruski 1994; Sivinski et al. 1997). For these species a 2- mm orifice was left in the upper portion of each guava (between the ‘lid’ and the ‘base’) to serve as an entrance for female parasitoids. Because adults of these two figitid species prefer to forage on the ground (Ovruski et al. 2004), fruit were not hung, but rather placed on cage floors.

Modified Petri dish (M-PD). This technique was only used in the case of U. anastre- phae, which as noted before, has the shortest ovipositor of the species we were attempting to colonize. The oviposition unit consisted of 10-cm diameter Petri dishes, which we made shallower by scraping down ca. 50% of the walls (height was lowered from 0.9 to 0.4 cm) (Figure 3). We placed A. ludens larvae mixed with the diet on which they had been reared on the lowered Petri dish ‘bottom plate’ and tightly covered it with a stretched-out piece of Parafilm (original size was 55 cm). We chose to use Parafilm, because we had observed that the organdy cloth, which worked well in the case of other species, apparently did not provide the necessary mechanical aculeus stimulation that U. anastrephae females needed before parasitizing larvae. Sandwich-type oviposition devices (SD). Once the parasitoids had reproduced for

Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009 several generations using the ‘fruit filled with larvae’ technique (FF), the next step in the colonization process was to develop an artificial oviposition substrate for parasitoid females that was inexpensive and easy to handle. For this reason, we began to adapt adult parasitoids to ‘sandwich-type devices’ (SD) which were similar to the Petri dish methodology employed for mass rearing exotic opine parasitoids such as D. longicaudata and D. tryoni (Cameron) in Hawaii (Wong and Ramadan 1992). We used two kinds of SD devices (Figure 4). Sandwich-type oviposition device one (SD1). This parasitization unit was suitable during the initial rearing stages of D. crawfordi, D. areolatus, and O. hirtus (Figure 2). It consisted of a 11.51.6-cm (diameterheight) plastic ‘dish’ with a bottom made of a 1515-cm piece of chiffon cloth. On the cloth surface we placed ca. 250 A. ludens larvae mixed with the diet on which they had been reared. The age of the larvae depended on the species of parasitoid being reared (details in Table 2). The dish containing larvae and diet was covered with another 1515-cm piece of chiffon cloth that was tied to the base by a 11.70.8-cm (diameterheight) plastic ‘ring’ 62 M. Aluja et al.

put in place by pushing against the base (i.e. pressure exerted with index fingers). After the ‘sandwich’ was built, we completely covered the chiffon cloth top with a layer of guava epicarp (skin) ca. 1 mm in thickness. The thin skin pieces were obtained by finely slicing the guava epicarp with a razorblade or sharp knife. The ultimate goal was to entice females to oviposit by mechanical and olfactory stimulation with the fragrant guava epicarp. Sandwich-type oviposition device two (SD2). The parasitization unit was the same as described under SD1, but in this case larvae were exposed in naked form (i.e. not mixed with diet). Furthermore, we did not place a layer of guava epicarp but instead soaked the chiffon cloth with liquid guava pulp. This method turned out suitable to entice wild D. crawfordi, D. areolatus, O. hirtus, and U. anastrephae females to oviposit. Uncovered Petri dish (UP). We discovered that the females of the figitids A. pelleranoi and O. anastrephae were suffering severe ovipositor damage while attempting to parasitize larvae in the oviposition units covered with chiffon cloth. Furthermore, because females of these species like to enter fruit in search of the larvae feeding inside, we used an uncovered unit. We used the bottom part of a Petri dish half filled with diet mixed with larvae. At the same time, half a guava was added to the artificial diet with larvae. The fruit, including seeds, was macerated into pieces and thoroughly mixed with the diet. In general, endocarp and mesocarp were utilized because the fruit’s fragrance appeared to attract females and stimulate oviposition behavior.

Oviposition substrate for pupal parasitoid Initial exposure of A. ludens pupae to C. haywardi was done in 500-mL plastic containers containing a ca. 10-cm layer of moistened soil (50Á70% water content) and some leaf litter. Soil was brought from the original collection locality of Tejer´ıa, Veracruz (Lo´pez et al. 1999), and was predominantly clay (Guille´n et al. 2002). Approximately 500 recently formed pupae (1Á2 days from pupation) were placed in the plastic container and mixed with the soil (referred to as CP method, i.e. covered pupae, in the text). Then, a mature guava placed on a galvanized wire screen was inserted into the container to lure parasitoids to the pupae underneath. The wire Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009 screen measured 1010 cm with 11-cm mesh openings. Pupae were exposed to parasitism over a period of 7 days (Table 2). The guava was only inserted into the oviposition unit during the first three generations, after which time the parasitoids seemed to respond well to A. ludens pupae alone. After the 21st generation, soil and leaf litter were eliminated and only ‘naked’ pupae (referred to as NP-method, i.e. naked pupae, in the text) were exposed during 3 days on a 11.51.6-cm (diameter height) plastic dish (Table 2).

Maintenance of parasitized larvae and pupae In the case of M-PD (modified Petri dish), SD1 (sandwich-type oviposition device one (larvae mixed with diet)), and UP (uncovered Petri dish) exposures, larvae were cleaned of diet and guava residues by placing them in a fine mesh plastic colander and rinsing them under running tap water. Once clean, larvae were placed in 500-mL plastic containers with 2.5 cm3 of moistened vermiculite where they formed puparia. Biocontrol Science and Technology 63

All containers were labeled, protected with a top made of chiffon cloth, and maintained under laboratory conditions (25918C, 7095% RH) until fly or parasitoid adults emerged. In the case of FF (fruit filled with larvae) exposures, fruit was placed in 200-mL plastic vials, which in turn were placed inside 500-mL plastic containers with 2.5 cm3 of moistened vermiculite. This was done to allow larvae to exit the fruit, a process that many times caused the fruit to disintegrate, spilling larvae and diet onto the floor of the 200-mL vial. On day 4, any diet or fruit residues were rinsed from pupae and larvae as described above and transferred to a 500-mL plastic container with moistened vermiculite, where they remained until adult emergence. The double container technique allowed us to avoid fungal and bacterial contamination that usually ensues if the vermiculite is mixed with fruit and diet residues.

Handling of emerged parasitoids and flies Once parasitoids and flies had emerged, they were transferred to a clean, empty Plexiglas cage and provided with food and water. The size of the cage and the number of males and females per cage depended on the species (see Table 2). Daily inspection of containers with pupae was critical to make sure that emerging adults did not escape or suffer stress because of lack of food and water. Length of pupal period and associated timing of parasitoid emergence was species-specific and may occur before, after, or in synchrony with host emergence. In the case of parasitoids that emerge before their host (i.e. U. anastrephae), there was no need to separate adult parasitoids from adult flies since unemerged A. ludens pupae were simply removed and discarded once the adult parasitoids had emerged. In the case of parasitoid species whose emergence is more synchronous with host emergence (i.e. O. hirtus, D. crawfordi, and D. areolatus), we were forced to separate adult parasitoids from adult flies. This was done utilizing a standard aspirator. Adult parasitoids that were very sensitive to ‘rough’ handling (i.e. aspirator) like D. areolatus, or that were destined for behavioral studies, were separated using 10-mL glass vials into which insects walked. When parasitoids had a more prolonged pupation interval than their hosts (i.e. A. pelleranoi, O. anastrephae, and C. haywardi), the emerged adult flies and empty puparia were separated to leave only parasitized pupae. Separation of flies and empty puparia was critical to avoid fungal growth and to significantly lower the risk Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009 of contamination by mites.

Determination of percent parasitism, sex ratio, and pupal viability per generation To measure percent parasitism, two 10-mL samples of parasitized A. ludens larvae (approximately 220 larvae) were processed per generation. The first sample was taken when parasitoid females reached, 4 and the second one when they reached 10 days of age. In the case of C. haywardi, instead of larvae, two random samples of 100 pupae were processed. The handling procedure for these larvae and pupae was the same as that described earlier. Once parasitoid adults had emerged, number and sex were recorded. Relative percent parasitism was estimated by dividing the total number of parasitoids that emerged by the total number of larvae exposed in the parasitization unit as we were not interested in an exact determination of the ‘killing power’ of each parasitoid species at this juncture (i.e. a certain proportion of larvae/ pupae were parasitized and killed and therefore ended up not yielding an adult 64 M. Aluja et al.

parasitoid). Pupal viability was determined as the total number of pupae that yielded flies and parasitoids divided by the total number of unemerged and emerged pupae.

Demographic studies. Doryctobracon areolatus, D. crawfordi, and O. hirtus Adults used in these tests stemmed from colonies that were 14 generations old. Utetes anastrephae, A. pelleranoi and O. anastrephae had been reared over nine generations, and C. haywardi over 24 generations. For these studies, braconid larval- prepupal species were only reared using A. ludens larvae in the FF (fruit filled with larvae) method, while figitid larval-pupal species were reared using A. ludens in the UP (uncovered Petri dish) method (Table 2). In all cases, 30 host larvae were exposed daily to 15 parasitoid pairs (i.e. 15 females and 15 males totaling 30 individuals per cage) for 24 h during their entire adult lifespan in Plexiglas rearing cages containing water and honey (details on size in Table 2). After exposure to parasitoid attack, host larvae were placed in plastic trays (500 mL) and provided with fresh larval diet. Three days after, formed pupae were separated from diet and transferred to other 500-mL trays with 150 mL of moistened vermiculite. All trays were taken into a room at 25918C, 7095% RH, and full darkness, where they remained until fly and parasitoid adults emerged. After all died (no food or water was provided), they were counted and sexed. In the case of the pupal parasitoid C. haywardi, 30 pairs (30 females and 30 males totaling 60 individuals per cage) were exposed daily to 20 two- day-old A. ludens pupae in 51.5 (diameter  height) plastic Petri dishes covered with 1 cm of vermiculite. Cages in this case, were 101010 cm in size, with glass walls and aluminum frame. Each study (i.e. one per species) was replicated five times. Life table parameters (lx, fraction of the original cohort surviving to age x; px, period survival; qx, period mortality; dx, fraction of the original cohort dying at age x; ex, expectation of life; Mx, average number of male and female offspring produced by female at age x; mx, female offspring per female at age x; Carey 1993, 1995) were calculated from daily mortality records and offspring data for cohorts of all larval- prepupal and pupal parasitoids. These values were used to determine reproductive parameters such as gross fecundity rate (GFR in text), net fecundity rate (NFR in text), cohort lifespans, and offspring sex ratios (as female proportions) and population parameters such as Ro (net reproductive rate), r (intrinsic rate of Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009 increase), l (finite rate of increase), and T (mean generation time) (Carey 1993; Vargas et al. 2002). These demographic parameters helped us to estimate the relative population growth vigor of the first colonized cohorts.

Experiments to determine optimal pupal age to rear C. haywardi We conducted two types of experiments with mated, 7-day-old females: no choice and choice tests. In each case we tested six treatments, each corresponding to an age class of the host (A. ludens pupae). Pupal age classes (days) tested were: 0Á2, 3Á5, 6Á8, 9Á11, 12Á14, and 15Á17. We used 30 pupae per age class, 10 per age included in every age class (i.e. in age class 0Á2, there were 10 pupae each of ages 0 (B24 h or prepupa), 1 and 2 days). In the no choice experiment, we released 15 C. haywardi females together with 30 pupae of a determined age class in a 500-mL plastic container that was halfway filled with sterilized clayey soil (pupae were superficially buried). Exposure period was 24 h and the experiment was replicated five times for every age Biocontrol Science and Technology 65

Table 3. Parasitization rates (mean percent parasitism), proportion of emerged females, and pupal viability in all seven native Anastrepha fruit fly parasitoids as the domestication and colonization process proceeded.

% Parasitisme % Emerged females % Pupal viabilitye Parasitoid species Rearing method (Mean9SEM) (Mean9SEM) (Mean9SEM)

D. crawfordia FF 38.792.9 54.592.8 55.992.9 SD1 20.891.4 47.992.5 58.893.3 SD2 37.992.1 44.792.6 43.692.3 U. anastrephaeb FF 26.194.3 58.792.9 72.892.6 M-PD 20.494.1 45.493.3 50.494.9 SD2 25.293.9 50.693.7 39.794.4 O. hirtusa FF 24.792.1 55.692.7 61.093.6 SD1 16.591.0 45.492.2 55.591.4 SD2 13.791.3 56.893.9 64.792.2 D. areolatusa FF 24.391.6 60.191.9 56.192.8 SD1 11.191.2 58.592.6 54.491.6

A. pelleranoia UP-24h 26.491.8 58.392.7 50.092.4 UP-7h 35.692.9 46.892.9 65.694.3

O. anastrephaec UP-24h (bisexual) 30.592.6 61.193.2 52.994.1 UP-24h (unisexual) 24.491.2 100.0 56.291.8 C. haywardid CP 3.890.3 57.191.3 8.490.3 NP 4.590.3 48.292.8 8.390.4

aData from first 14 generations; bdata from first nine generations; cdata from first 20 generations, ddata from first 12 generations. FF (fruit filled with larvae), M-PD (modified Petri dish), SD1 (sandwich-type oviposition device one [larvae mixed with diet]), (sandwich-type oviposition device two [naked larvae]), UP (uncovered Petri dish filled with larvae mixed with diet and fruit pulp; 7 and 24 h refer to exposure period), CP (pupae covered with soil), NP (pupae exposed naked [without soil cover]).

class. In all cases (each replicate) we used a new cohort of females (i.e. no repeated measures on same cohort). In the multiple choice experiment, we released 90 females together with 180 pupae encompassing all age classes (30 pupae per age class) in a 1510-cm (diameterheight) plastic container that was also half-filled with Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009 sterilized clayey soil. To distinguish pupae of every age class, they were individually marked with a dot of acrylic paint (six colors used) (Colores Acr´ılicos Indelebles Politec, Distribuidora Rodin, Mexico). Exposure period in this case was 36 h and we replicated each experiment five times. The pupae were handled as already described before until all parasitoids emerged and were counted.

Statistical analyses Owing to the fact that colonization efforts where not simultaneous and that we typically only had access to a small number of individuals of any given species at any particular time, we could not run any formal statistical analyses comparing the performance of the various rearing methods. Nevertheless, overall trends can be ascertained by visually comparing data summarized in Tables 3 and 5. In the case of the experiment to determine optimal A. ludens pupal age to rear C. haywardi, we ran a one-way ANOVA comparing percent parasitism, sex ratio and proportion of 66 M. Aluja et al.

unemerged puparia (sometimes the host is killed due to single or multiple parasitoid stings). Post-hoc mean comparisons were done by means of a Tukey honest significant difference test (HSD) at an a of 0.05. Proportions were arcsine square root transformed prior to analysis, but untransformed means are presented in the text.

Results Colonization and adult handling conditions A summary of parasitoid rearing and handling procedures is provided in Table 2. In what follows, we report the most relevant results of the colonization efforts on a per species basis to facilitate domestication and colonization efforts in other parts of the world. We place emphasis on sex ratios, percent parasitism and mean proportion of pupae yielding a parasitoid given that these parameters greatly influence the success rate of the domestication/colonization process early on.

Doryctobracon crawfordi. The domestication process of this species was initiated in October 1994, using the FF (fruit filled with larvae) method over 10 generations. Then gradually, between the 10th and 15th generations, we exposed the parasitoids to the SD1 (sandwich-type oviposition device using larvae mixed with diet) method. The length of the larval exposure period was the same in both cases (Table 2). The sex ratio for both FF and SD1 parasitoids varied throughout the colonization process. For example, for FF parasitoids, the smallest proportion of females occurred in the first four generations (0.4Á0.9:1). From generation 5 to 42 and with only one exception (generation six, 0.7:1), the sex ratio consistently favored females (1.1Á7.0:1). Similarly, the lowest proportion of SD1 females was observed in the first eight generations (0.3Á0.9:1), whereas the highest appeared after generation 9 (1.1Á2.6:1; generations 9Á14). Starting with generation 14, the SD1 technique was replaced by method SD2 (sandwich-type oviposition device using naked larvae). The sex ratio in the SD2 strain varied sharply from generation to generation over the 44 generations recorded (most likely due to variations in host quality). The lowest proportions of SD2 females were 0.2:1, whereas the highest proportions were 6:1 (mean values in Table 3). Percent parasitism levels during the first 14 generations using the three rearing methods varied between 15.5Á62.3% (FF), 9.1Á41.8% (SD1) Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009 and 20.0Á56.8% (SD2) (mean values in Table 3).

Doryctobracon areolatus. This parasitoid species presented various challenges during the early stages of the domestication/colonization process. Among the most difficult ones to overcome was a propensity to enter what appeared to be a reproductive diapause from late November until almost March (coldest time of the year), despite the fact that we controlled temperature and lighting conditions inside the laboratory. As a result, from 1993 to 1997 we were only able to keep temporary colonies (all eventually died out) by using plums (Spondias purpurea and S. mombin) and mangos (Mangifera indica) naturally infested by A. obliqua (collected in the field) and parasitized by D. areolatus. Later, in July of 1997, we were able to successfully establish two D. areolatus colonies using artificially reared A. ludens larvae as a host, taking advantage of an unusually high parasitism rate in A. obliqua developing in the above mentioned fruit species. One of the colonies was maintained employing the SD1 (sandwich-type oviposition device using larvae mixed with diet) method while the Biocontrol Science and Technology 67

other colony was maintained using the FF (fruit filled with larvae) technique (see Table 2 for details). As was the case with D. crawfordi, sex ratios in both successfully colonized strains tended to be initially male-skewed. However, in subsequent generations the proportion of males and females was gradually equalized or favored females. The lowest proportion of FF females was observed in first and second generations (0.7Á0.8:1) and thereafter (up to generation 69) it reached a maximum of 3.5:1. Overall, sex ratios of SD1 parasitoids were female skewed but intergenerational variation was greater than that observed in FF parasitoids (Table 3). Parasitism rates during generations 1Á69 (68 in the case of the SD1 method) using the FF method varied from 8.2 to 36.4% between first and 69th generation, whereas employing the SD1 technique varied from 1.4 to 25.9% between first and 68th generation (Table 3). Opius hirtus. Domestication of the first strain of this species was initiated in October 1994, through FF (fruit filled with larvae) exposures. However, the colony was lost in generation 6 (March 1995). We believe that failure hinged principally on the fact that females were probably not mating because of saturation of the environment with sexual pheromones (avery strong fruit-like bouquet was perceived near the cage). Wetherefore doubled cage size and introduced citrus brancheswith ample foliage as resting sites (tips of branches were inserted into 60-mL glass vials covered with cotton to prevent the parasitoids from drowning). After the original failure, a new colonization attempt was initiated in January 1996 with a few (B20) parasitoids obtained from a rare Anastrepha species (A. cordata) collected in the few remaining patches of tropical evergreen rainforest in Southern Veracruz, Mexico. Due to the difficulties involved in finding parasitoids in nature and considering our initial failure, we maintained three strains along the domestication/colonization process. Initially, we used the FF technique and then (generation six), started a new line using the SD1 (sandwich-type oviposition device using larvae mixed with diet) method (Table 2). Three generations later, we started a third line, by switching to the SD2 (sandwich-type oviposition device using naked larvae) method. In the latter case, we reduced the exposure period 5-fold with respect to the other rearing methods. Because in nature O. hirtus females are faced with very low host densities, we wanted to reduce the risk of larvae being marked with a marking pheromone that would have caused females to quickly leave the ‘resource patch’. Sex ratios in the FF strain tended to be initially (generations 1Á7) male-skewed

Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009 (0.4Á0.9:1), but in subsequent generations (8Á37), favored females (1.3Á5.6:1). In general, sex ratios of SD1 parasitoids were more male-skewed than FF parasitoids. In the case of the SD2, sex ratios were highly variable over time (0.2Á5.3:1 over 115 generations). Mean parasitism was highest under the FF rearing method (Figure 2). Parasitization rates varied between 9.5Á59.1, 7.2Á26.4, and 6.8Á31.4% in the FF, SD1, and SD2 lines, respectively.Pupal viability was highest in the FFand SD2 lines (Table 3). Utetes anastrephae. This parasitoid presented a particularly difficult challenge because of its extremely short ovipositor and the fact that it is usually reared from only very small fruit in nature (e.g. S. mombin, Tapirira mexicana;Lo´pez et al. 1999; Sivinski et al. 2000, but see Eitam et al. 2004 for exceptions to the rule). The first unsuccessful colonization attempt was made in October 1996 using the FF rearing method (after fourth generation no adults emerged). In September 1999, another attempt was made using ca. 800 female parasitoids collected from A. obliqua larvae infesting S. mombin. The original colony was divided into FF (fruit filled with larvae) and M-PD (modified Petri dish) strains. After four generations, we initiated a third strain (SD2 68 M. Aluja et al.

(sandwich-type oviposition device using naked larvae)) with M-PD material. Exposure periods in the M-PD and SD2 strains were reduced 2Á7-fold with respect to the FF strain to avoid superparasitism caused by easier access to larvae (Table 2). Sex ratios in the FF strain were slightly male-skewed in the first two generations (0.8Á0.9:1), but then remained relatively stable over the next 11 generations, with a consistent tendency for more females to emerge than males (1.1Á6.5:1). In contrast, sex ratios in the M-PD and SD2 strains were highly variable between generations. The lowest proportion of females fluctuated between 0.3 and 0.9:1 in both M-PD and SD2 strains, while the greatest proportions fluctuated between 1.1Á2.3:1 and 1.0Á7.0:1 in the M-PD (first 11 generations) and SD2 (first nine generations) (mean 9 SE values in Table 3). Parasitization rates varied between 6.8Á51.8, 1.8Á56.4, and 3.6Á60% in the FF, M-PD, and SD2 rearing methods, respectively (mean 9 SE values in Table 3). Finally, we found that pupal viability in insects stemming from FF lines was higher than those stemming from M-PD and SD2 lines (Table 3).

Aganaspis pelleranoi. A colony of this figitid parasitoid was initiated in September of 1994, using adults obtained from field-infested P. guajava. At first, parasitoids were reared with the variant of the FF (fruit filled with larvae) technique described in Section 2, but few individuals were obtained per generation. Therefore, beginning with the fifth generation, this technique was replaced by the UP (uncovered Petri dish) rearing method, allowing us to reduce exposure periods 3-fold (Table 2). In general, and with few exceptions (e.g. generation one), sex ratio in UP-24h (24 h refers to the exposure period in hours) parasitoids favored females over the first 14 generations. In the case of the UP-7h strain, sex ratios were highly variable, ranging between 0.2 and 8:1 (78 generations considered). Parasitization rates varied between 20.0Á68.2 and 11.8Á43.6% in the UP-7h and UP-24 h lines, respectively. Also, pupal viability in UP-7h lines was higher than in UP-24h lines (Table 3).

Odontosema anastrephae. The first unsuccessful attempt at colonization was started in November of 1995. For the first two generations, we employed the variant FF (fruit filled with larvae) method, but extremely low yields forced us to switch to the UP (uncovered Petri dish) technique using 36-h exposure periods. However, extremely low oviposition activity by females and an extremely male-biased sex- Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009 ratio (as low as 0.2:1), lead to the demise of the colony after eight generations. After a 3-year search for sufficient wild material, we were finally able to start a new colony between September and November 1998, using the UP rearing technique. A second O. anastrephae colony was started in February 2000, with wild material stemming from guavas. Interestingly, starting with generation 11 (December 2000) essentially only females emerged (such a pattern has remained steady over more than 75 generations). On occasion, one or two males emerged (sex ratio of 1: 0.008), but when such was the case, we immediately removed them given our interest in maintaining a theliotokous line. In both lines, exposure period was gradually reduced to 6 h (details in Table 2). In the case of the bisexual O. anastrephae colony, sex ratios varied greatly between generations, fluctuating between 0.1:1 (first four generations) and 1.1Á8.0:1 in the remaining generations (29 generations considered). Parasitism rates varied between 7.7Á79.5 and 12.0Á37.8%, in the unisexual and bisexual colonies, respectively. Pupal viability in the bisexual and unisexual O. anastrephae colonies was similar (Table 3). Biocontrol Science and Technology 69

Coptera haywardi. This endoparasitic pupal parasitoid was first colonized in Novem- ber of 1994 by means of the CP-method (covered pupae (artificially buried in soil)). Starting with the 21st generation, we replaced this rearing technique with the NP- method (naked pupae (soil removed)), which is still currently used because of its practicality. Sex ratio in CP-line favored females in all generations (1.1Á2: 1), except one (generation 15, 0.7:1; data stem from generations 4 to 20). In the case of the NP- line, sex ratios varied more, fluctuating between 0.5 and 2.5:1 (34 generations considered). Parasitism rates varied initially between 2.2Á6.0 and 2.8Á7.4% in the CP and NP lines (first 12 generations obtained using each rearing method). Currently (generations 35Á42 in NP method), parasitism rates have reached 21.491.1% (range 11.3Á27.1%, n16), and sex ratios fluctuate between 0.4 and 2.5:1. Data on mean parasitism rates, mean proportion of emerged females and pupal viability for the first 12 generations are shown in Table 3. Results of the experiments to determine optimal pupal age are summarized in Table 4. Under choice conditions, parasitism in pupal age classes 0Á2, 3Á5, and 6Á8 was significantly higher than in age classes 9Á11, 12Á14, and 15Á17 (one-way ANOVA, F5,24 78.43, PB0.0001). Similar results were obtained in the no-choice experiment (one-way ANOVA, F5,24 31.46, PB0.0001). Mean parasitism in the optimal pupal age class varied between 60 and 70% in the no-choice experiment and between 36 and 55% in the choice one (further details in Table 4). With respect to sex ratios, in both choice and no-choice experiments, mean proportion of females was similar in parasitoids emerging from pupae within the first 5 age classes (i.e. 0Á2, 3Á5, 6Á8, 9Á 11, 12Á14) but different when compared to the sixth age class (15Á17 days) (one-way ANOVA, F5,24 3.50, P0.0161 and F5,24 6.26, P0.0008 for the choice and no- choice conditions, respectively) (Table 4). There were no statistically significant differences among age classes with respect to the proportion of unviable (i.e.

unemerged) pupae in the choice experiment (F5,24 0.99, P0.4425). The situation

Table 4. Percent parasitism, sex ratio (proportion of females) and proportion of uneclosed pupae in the experiments designed to determine the optimal host age (Anastrepha ludens pupae) for Coptera haywardi. Experiments conducted under choice (pupae of varying ages offered simultaneously to ovipositing females) and no-choice conditions (females offered pupae

Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009 of only one age class).

Choice experiment (mean9SEM) No-choice experiment (mean9SEM)

Age class of A. ludens % Unemerged % Unemerged pupae % Parasitism % Females pupae % Parasitism % Females pupae

0Á2 55.392.3a 52.794.1a 37.395.8a 60.795.0a 45.197.0a 30.795.9ab 3Á5 46.093.9ab 52.297.2a 44.795.9a 68.095.1a 55.2912.8a 17.394.6a 6Á8 36.795.5b 32.395.9a 57.396.4a 60.998.0a 55.796.2a 28.798.1ab 9Á11 20.092.9c 42.2917.4a 58.099.6a 43.3910.7a 76.192.2a 41.395.5ab 12Á14 3.391.1d 20.0920.1a 66.798.8a 12.792.7b 48.3920.5a 52.095.7b 15Á17 0.090.0d 0.090.0b 52.7916.1a 0.090.0b 0.090.0b 51.792.2b

Means within a column followed by the same letter are not significantly different (Tukey HSD test, P 0.05). 70 M. Aluja et al.

changed in the case of the no-choice experiment, since significant differences were detected (F5,24 5.91, P0.0011) (Table 4).

Demographic parameters Reproductive and population parameters for larval-prepupal and pupal parasitoid species are summarized in Table 5. Highest GFR, NFR, Ro, r,andl were recorded in the diaprid C. haywardi and in the braconid D. crawfordi. Mean generation time (T) was longest in the case of C. haywardi and A. pelleranoi, while it was short and similar in D. areolatus, D. crawfordi, and O. hirtus. Mean life spans in all larval- prepupal parasitoid species were quite short (B15 days). In contrast, in the pupal parasitoid C. haywardi lifespan was almost twice as long (Table 5). Survivorship curves for all species are shown in Figure 5.

Discussion Our multiyear effort aimed at domesticating and colonizing various native fruit fly parasitoids resulted in many practical lessons that will hopefully facilitate similar efforts elsewhere in the world. Clearly, there were a number of major hurdles to overcome before successful establishment of stable colonies was achieved: (1) availability of large enough numbers of wild parasitoids to start a colony in the cases of rare species like O. hirtus. (2) Availability of a stable supply of high quality larval or pupal hosts. (3) Finding a fruit species that is available year round and that emits volatiles attractive to as many parasitoid species as possible and that can therefore be used to entice females to lay eggs under highly artificial laboratory conditions (e.g. guava in our case). (4) Building oviposition units that expose sufficient larvae to the attack of females with varying ovipositor sizes. (5) Overcoming the initially highly male-biased sex ratio, presumably due to lack of mating that in many cases led to the demise of the incipient colony. (6) Overcoming apparent pheromone saturation in the small rearing cages that can lead females to not mate or do so reluctantly. (7) Finding ideal environmental conditions to suit the idiosyncrasies of each species. (8) Cost considerations as the domestication and colonization processes are labor and material intensive and therefore end up being Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009 expensive. As noted by Vargas et al. (2002), knowledge on parasitoid demographic parameters is critical when trying to select candidate species for fruit fly biological control. Our data here, added to the wealth of knowledge already accumulated on the basic biology and ecology of native Anastrepha parasitoids (e.g. Sivinski et al. 1997, 2000; Aluja et al. 1998, 2003; Eitam et al. 2003, 2004) highlights the potential that species such as D. crawfordi, O. hirtus and C. haywardi have for augmentative release programs in regions with variable climatic and host density conditions. Furthermore, in many Latin American countries, in addition to dealing with pestiferous Anastrepha species, the presence of C. capitata is often the main concern for growers. Two of the species colonized here (i.e. A. pelleranoi and C. haywardi) are the only native parasitoids shown so far to be able to attack this important agricultural pest (Ovruski et al. 2004, 2005). Being able to choose among many parasitoid species opens up the possibility to release the one best adapted to the particular climatic and ecological conditions of a Table 5. Basic demographic parameters for seven native Anastrepha larval-prepupal and pupal parasitoids successfully colonized.

Parasitoid species

Demographic Technology and Science Biocontrol Parameter (Mean9SEM) D. areolatus1 D. crawfordi1 U. anastrephae2 O. hirtus1 A. pelleranoi2 O. anastrephae2 C. haywardi3

Offspring sex ratio (female 58.5896.45 50.4392.98 49.4192.58 60.2593.84 55.0293.79 30.9094.97 56.4192.47 proportion) Cohort lifespan (days) 9.8290.41 11.0990.11 10.5091.37 11.2393.02 7.9491.11 5.3490.35 28.0491.87 GFR (gross fecundity rate) 6.6191.75 29.1598.30 2.8790.40 6.2792.04 13.5791.84 13.2692.20 85.1396.63

Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009 6 November 15:46 At: Jorge] [Hendrichs, By: Downloaded (offspring/female) NFR (net fecundity rate) 2.1990.41 10.6891.40 2.6490.34 2.1490.37 5.1790.74 5.5490.81 63.7090.76 (offspring/female) Ro (net reproductive rate) (female 1.3990.16 5.3690.66 1.3490.20 1.2790.13 2.8490.53 1.4490.21 35.2490.78 offspring/generation) r (intrinsic rate of increase) (per 0.0390.01 0.2490.04 0.0790.04 0.0390.01 0.1390.03 0.0990.03 0.2590.01 female per day) l (finite rate of increase) (per day) 1.0490.01 1.2790.05 1.0890.04 1.0390.01 1.1590.03 1.0990.03 1.2890.01 T (mean generation time) (days) 8.6590.87 7.6991.44 3.0890.39 8.4690.68 7.4990.45 4.0790.73 14.3790.39

1Individuals stemmed from colonies that were 14 (1), 9 (2) and 24 (3) generations old, respectively. 71 72 M. Aluja et al.

C. haywardi 1 D. areolatus 0.9 U. anastrephae 0.8 D. crawfordi O. hirtus 0.7 A. pelleranoi 0.6 O. anastrephae 0.5 0.4 Survivorship 0.3 0.2 0.1 0 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 Age (days)

Figure 5. Survivorship (lx) curves for Doryctobracon areolatus, D. crawfordi, Opius hirtus, Utetes anastrephae, Aganaspis pelleranoi, Odontosema anastrephae and Coptera haywardi.

fruit growing region (e.g. temperature, rainfall, host density, larval host) which can greatly influence the efficacy of the control agent released or strategy implemented (Ovruski et al. 2000; Sivinski et al. 2000). In Mexico, a good example of the latter is represented by the native D. crawfordi and the recently introduced (1954Á1955, quoted in Jime´nez-Jime´nez 1956) exotic species D. longicaudata which, given the short time of their interaction (B50 years), have not been able to partition the niche in which they forage in nature (Miranda 2002). Both have long ovipositors (Sivinski et al. 2001) and thus are able to attack third instar A. ludens larvae in large fruit such as Citrus sinensis, C. paradisi and M. indica in perturbed environments (Lo´pez et al. 1999) where they exhibit similar distributions in tree canopies (Sivinski et al. 1997). Of the two species, D. longicaudata has already been successfully released augmentatively to reduce populations of A. ludens and A. obliqua in mango plantations in warm, lowland areas of the Soconusco region in Chiapas, Mexico (Montoya et al. 2000). Interestingly, here we found that D. crawfordi was not only the species exhibiting the highest r values of all larval-prepupal parasitoids studied (Table 5), but its intrinsic rate of population increase was twice as high as the one Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009 reported for D. longicaudata reared on Bactrocera dorsalis (Hendel) under laboratory conditions (Vargas et al. 2002). As documented by Sivinski et al. (2000), D. crawfordi, in contrast to D. longicaudata, prefers more humid, temperate environ- ments and does not enter diapause (which is the case with D. longicaudata; Aluja et al. 1998). According to Miranda (2002), each species should be released singly in different environments owing to the fact that they compete for the same resource. A particularly interesting potential release site for D. crawfordi is in areas where the native A. ludens host (Casimiroa greggii [S. Watts]) is abundant (e.g. canyons and mountain slopes in Tamaulipas and Nuevo Leo´n, Mexico), allowing fly populations to increase and cause damage to commercial citrus groves planted nearby. D. crawfordi is indigenous to those areas (Gonza´lez-Herna´ndez and Tejada 1979), rendering augmentative releases of this native species instead of the exotic D. longicaudata, more environmentally friendly (Simberloff and Stiling 1996). Despite the fact that D. areolatus was one of the native species with one of the lowest r values, it nevertheless exhibits certain ecological advantages over D. Biocontrol Science and Technology 73

crawfordi. For example, it is the most widely distributed native fruit fly parasitoid in the Neotropics (i.e. Florida to Argentina) and exhibits a close association with A. obliqua in native fruit species within the Anacardiaceae (Ovruski et al. 2000). As is the case with the exotic D. longicaudata, D. areolatus also prefers warm and drier environments at lower altitudes (Sivinski et al. 2000; but see below). Based on the fact that Vargas et al. (2002) reported a 4-fold higher intrinsic rate of increase in D. longicaudata when compared to what we found here for D. areolatus, the logical inference would be that the exotic species is a better candidate for augmentative releases. But recent evidence gathered in Florida where both species coexist (Eitam et al. 2004), indicates that at least in that part of the world, the distribution of D. longicaudata was negatively related to variance in monthly temperatures (it was most abundant in southern Florida along the Atlantic and Gulf coasts). These authors also reported that D. longicaudata may depend on a constant supply of hosts. In contrast, D. areolatus, a species that is able to diapause over extended periods (11 months; D. longicaudata did so only over a 7-month period) (Aluja et al. 1998), was the dominant species in most interior locations (Eitam et al. 2004). Based on the findings of Eitam et al. (2004), in Florida D. areolatus is apparently a superior searcher, while D. longicaudata a superior intrinsic competitor. So, as was the case with the previous example (D. crawfordi/D. longicaudata), augmentative releases of D. longicaudata need to be tailored to local conditions and are not warranted in every location. Opius hirtus exhibited similar r and fecundity values as D. areolatus, but together with D. crawfordi, was one of the larval-prepupal species that lived longest. Of all the braconid species that we successfully colonized, it is the least common and most specialized parasitoid (Sivinski et al. 2000; Aluja et al. 2003). Recently, Garc´ıa- Medel, Sivinski, D´ıaz-Fleischer, Ram´ırez-Romero, and Aluja (2008) showed that it is very effective at parasitizing hosts at very low densities and that it is able to coexist with other species such as D. longicaudata. As indicated by LaSalle (1993), many times rare parasitoid species exert a significant regulatory effect on pests. All the above renders O. hirtus an interesting candidate for more wide scale tests. The fourth species of native braconid parasitoid that we were able to colonize was U. anastrephae. In nature, this species is specialized at attacking A. obliqua and A. fraterculus in small fruit within the Anacardiaceae (e.g. Spondias spp.) and Myrtaceae (e.g. Psidium spp., Eugenia spp., Myrcianthes spp.), respectively (Sivinski Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009 et al. 1997; Lo´pez et al. 1999; Ovruski et al. 2004). The detailed studies by Sivinski et al. (1997) discovered an apparent partitioning of the niche in S. mombin trees, with U. anastrephae being most abundant in interior parts of the canopy preferentially infesting smaller fruit, while D. areolatus was most abundant in exterior parts of the canopy and infested larger fruit. Program managers would have to ascertain if any of these characteristics are of interest when deciding about new potential candidates for augmentative releases. Of the two figitid species we were able to successfully colonize, A. pelleranoi offers various interesting attributes. On the one hand, and in contrast to the braconid species, it preferentially forages on the ground where it attacks larvae in fallen fruit (Sivinski et al. 1997; Ovruski et al. 2004). It does so in a wide range of hosts that varies greatly with respect to physical and chemical characteristics (Wharton et al. 1998; Ovruski et al. 2000). On the other hand, it is one of the few native parasitoid species in the New World that is able to attack C. capitata (Baeza-Larios et al. 2002a; Ovruski et al. 2004, 2005). 74 M. Aluja et al.

Coptera haywardi (the only pupal parasitoid the colonization of which we describe here), was the species exhibiting the longest survival and highest fecundity, and exhibited r values similar to those found in D. crawfordi. We also show here that it can attack pupae of highly contrasting age classes (i.e. 0Á2to9Á11 days of age). Furthermore, C. haywardi produced high rates of pupal mortality (85Á92%). Similar observations were previously reported by Sivinski et al. (1998b) with A. suspensa (Loew) and Guille´n et al. (2002) with A. ludens pupae. Considering all the above, and the fact that C. haywardi is an endoparasitoid that only attacks tephritid flies (Sivinski et al. 1998b), among them C. capitata and several species within Anastrepha,it represents an ideal candidate to substitute generalist, cosmopolitan species such as P. vindemiae, Spalangia endius Walter and S. cameroni Perkins, which are known primarily as parasitoids of synantropic flies (e.g. in poultry sheds) (Morgan 1986). We conclude that, given the relatively fast adaptation of these organisms to laboratory conditions, it is feasible to mass rear most of them. As a matter of fact, in the case of D. crawfordi, A. pelleranoi, and C. haywardi, successful attempts at mass- rearing have already taken place in the fruit fly and parasitoid mass-rearing facilities of the Medfly Program in Metapa de Dom´ınguez, Chiapas, Mexico and, in the case of C. haywardi, the La Aurora rearing facility in Guatemala City, Guatemala (see Baeza- Larios, Sivinski, Holler, and Aluja 2002b). Furthermore, as reported by Cancino et al. (2008), with the exception of A. pelleranoi and O. anastrephae, native parasitoids can be successfully reared using irradiated larvae or pupae. As discussed above, demographic parameters from well-established colonies such as ours might guide mass-rearing and control programs. They indicate, all other things being equal, which parasitoids might increase at the greater rate and thus are cheaper to produce. Furthermore, the effectiveness of the parasitoid species successfully colonized here should not be limited to Mexico, but rather they should be amenable to introduction, augmentation and conservation in many other tropical areas (e.g. Costa Rica, Colombia, Venezuela, Brazil, Bolivia, Argentina) where fruit flies such as A. fraterculus, A. obliqua, A. ludens, A. serpentina, A. striata, and A. sororcula are important pests. Gates et al. (2002) have highlighted three important benefits of the use of native parasitoids in biological control: (1) avoidance of costly and prolonged trips abroad in search of candidate species, (2) avoidance of cumbersome importation and quarantine protocols, and (3) avoidance of potential non-target effects on local Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009 fauna (also see Simberloff and Stiling 1996). Importantly, and given the massive rate of deforestation prevalent in Latin America, on top of searching for native species as potential fruit fly biological agents, we also need to foster the conservation of natural habitats, to enhance local parasitoid reservoirs and prevent the local extinction of rare species such as O. hirtus (Aluja 1996, 1999; Aluja et al. 2003).

Acknowledgements We thank two anonymous reviewers and the editor for helping us produce a more polished final product. We also wish to acknowledge Juan Rull, Jaime Pin˜ero, Diana Pe´rez-Staples, Astrid Eben, Francisco D´ıaz-Fleischer and Ricardo Ram´ırez (all Instituto de Ecolog´ıa, A.C.) and Patrick Dohm for revisions and suggestions on improving the manuscript. We thank Drs Robert Wharton (Texas A&M University) for identifying all the braconid and figitid parasitoids described here and Lubomir Masner (Canadian Bureau of Land Resources), for identifying C. haywardi. Thanks are due also to Jorge A. Mu¨ller and his collaborators (Comite´ Estatal de Sanidad Vegetal, Xalapa, Veracruz, Me´xico) for donating the A. ludens laboratory strain we used to rear all parasitoids. We appreciate the important technical support of Isabel Ja´come, Biocontrol Science and Technology 75

Emmanuel Herrera, Cesar Ru´ız, Gloria Lagunes, Guadalupe Trujillo, Cecilia Mart´ınez, Alejandro Va´zquez, Graciano Blas, Alberto J. Mata, and Braulio Co´rdoba (all Instituto de Ecolog´ıa, A.C.). We especially thank Faustino Cabrera Cid and family (Tejer´ıa), Otho´n Herna´ndez (Llano Grande) and Pablo Ventura (Los Tuxtlas) for allowing us to work in their orchards in order to collect parasitoids. Thanks are due to Nicoletta Righini and Alberto Anzures (both INECOL) for formatting and final preparation of this manuscript, and to Dar´ıo Garc´ıa-Medel for preparing Figures 2Á4. This work was principally financed with resources from the Mexican Campan˜a Nacional Contra Moscas de la Fruta (Secretar´ıa de Agricultura, Ganader´ıa, Desarrollo Rural y Pesca Á Instituto Interamericano de Cooperacio´n para la Agricultura [SAGARPA-IICA]) and the US Department of Agriculture, Agricultural Research Service (USDA-ARS Á Agreement No. 58-6615-3-025). Complementary resources were provided by the USDA Office of International Cooperation and Development (OICD Á Project No. 198-23), the Mexican Comisio´n Nacional para el Conocimiento y Uso de la Biodiversidad (Project No. H-296), and the Consejo Nacional de Ciencia y Tecnolog´ıa Á Sistema Regional del Golfo de Me´xico (Project 96-01-003-V). Sergio Ovruski acknowledges the CONICET- Argentina for support during his research stays in Me´xico during 2003 and 2004. MA also acknowledges support from CONACyT through a Sabbatical Year Fellowship (Ref. 79449) and thanks Benno Graf and Jo¨rg Samietz (Forschungsanstalt Agroscope Changins-Wa¨denswil ACW), for providing ideal working conditions to finish writing this paper.

References Aguiar-Menezes, El, Menezes, E.B., and Loia´cono, M.S. (2003), ‘First Record of Coptera haywardi Loia´cono (Hymenoptera: Diapriidae) as a Parasite of Fruit-Infesting Tephritidae (Diptera) in Brazil’, Neotropical Entomology (Brazil), 32, 355Á358. AliNiazee, M.T. (1985), ‘Opiine Parasitoids (Hymenoptera: Braconidae) of Rhagoletis pomonella and R. zephyria (Diptera: Tephritidae) in the Willamette Valley, Oregon’, Canadian Entomologist, 117, 163Á166. Aluja, M. (1994), ‘Bionomics and Management of Anastrepha’, Annual Review of Entomology, 39, 155Á178. Aluja, M. (1996), ‘Future Trends in Fruit Fly Management’,inFruit fly Pests: A World Assessment of Their Biology and Management, eds. B.A. McPheron and G.J. Steck, DelRay Beach, FL: St. Lucie Press, pp. 309Á320. Aluja, M. (1999), ‘Fruit fly (Diptera: Tephritidae) Research in Latin America: Myths, Realities and Dreams’, Anais da Sociedade Entomologica do Brasil, 28, 565Á594. Aluja, M., Celedonio-Hurtado, H., Liedo, P., Cabrera, M., Castillo, F., Guille´n, J., and R´ıos, E. (1996), ‘Seasonal Population Fluctuations and Ecological Implications for Management of Anastrepha Fruit Flies (Diptera: Tephritidae) in Comercial Mango Orchards in Southern Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009 Mexico’, Journal of Economic Entomology, 89, 654Á667. Aluja, M., Guille´n, J., Liedo, P., Cabrera, M., R´ıos, E., de la Rosa, G., Celedonio-Hurtado, H., and Mota, D. (1990), ‘Fruit Infesting Tephritids (Diptera: Tephritidae) and Associated Parasitoids in Chiapas, Mexico’, Entomophaga, 35, 39Á48. Aluja, M., Lo´pez, M., and Sivinski, J. (1998), ‘Ecological Evidence for Diapause in Four Native and One Exotic Species of Larval-Pupal Fruit Fly (Diptera: Tephritidae) Parasitoids in Tropical Environments’, Annals of the Entomological Society of America, 91, 821Á833. Aluja, M., Rull, J., Sivinski, J., Norrbom, A.L., Wharton, R.A., Mac´ıas-Ordo´n˜ez, R., D´ıaz- Fleischer, F., and Lo´pez, M. (2003), ‘Fruit Flies of the Genus Anastrepha (Diptera: Tephritidae) and Associated Native Parasitoids (Hymenoptera) in the Tropical Rainforest Biosphere Reserve of Montes Azules, Chiapas, Mexico’, Environmental Entomology, 32, 1377Á1385. Baeza-Larios, G., Sivinski, J., Holler, T., and Aluja, M. (2002a), ‘The Ability of Coptera haywardi (Ogloblin) (Hymenoptera: Diapriidae) to Locate and Attack the Pupae of the Mediterranean Fruit Fly, Ceratitis capitata (Wiedemann) (Diptera: Tephritidae) under Seminatural Conditions’, Biological Control, 23, 213Á218. Baeza-Larios, G., Sivinski, J., Holler, T., and Aluja, M. (2002b), ‘The Effects of Chilling on the Fecundity and Life Span of Mass-Reared Parasitoids (Hymenoptera: Braconidae) of the 76 M. Aluja et al.

Mediterranean Fruit Fly, Ceratitis capitata (Wiedemann) (Diptera: Tephritidae)’, Biocontrol Science & Technology, 12, 205Á215. Bautista, R.C., Harris, E.J., and Vargas, R.I. (2001), ‘The Fruit Fly Parasitoid Fopius arisanus: Reproductive Attributes of Pre-Released Females and the Use of Added Sugar as a Potential Food Supplement in the Field’, Entomologia Experimentalis et Applicata, 101, 247Á255. Canal, N.A., and Zucchi, R.A. (2000), ‘Parasito´ides Á Braconidae’,inMoscas-das-frutas de Importaˆncia Econoˆmica no Brasil: Conhecimento Ba´sico e Aplicado, eds. A. Malavasi and R.A. Zucchi, Ribera˜o Preto, Brasil: Holos Editora, pp. 119Á126. Cancino, J., Ruiz, L., Sivinski, J., Ga´lvez, F.O., and Aluja, M., ‘Rearing of Five Hymenopterous Larval-Prepupal (Braconidae, Figitidae) and Three Pupal (Diapriidae, Chalcidoidea, Eurytomidae) Native Parasitoids of the Genus Anastrepha (Diptera: Tephritidae) on Irradiated A. ludens Larvae and Pupae’, Biocontrol Science and Technology (in press). Carey, J.R. (1993), Applied Demography for Biologists with Special Emphasis on Insects,New York: Oxford University Press. Carey, J.R. (1995), Insect Demography. Encyclopedia of Environmental Biology 2, San Diego, CA: Academic Press. Dom´ınguez, J., Herna´ndez, E., and Castellanos, D. (2002), Cr´ıa masiva de moscas de la fruta. Memorias del XIV Curso Internacional sobre Moscas de la Fruta, CICMF, Metapa de Dom´ınguez, Chiapas, Me´xico, pp. 172Á180. Eitam, A., Holler, T., Sivinski, J., and Aluja, M. (2003), ‘Use of Host Fruit Chemical Cues for Laboratory Rearing of Doryctobracon areolatus (Hymenoptera: Braconidae), a Parasitoid of Anastrepha spp. (Diptera: Tephritidae)’, Florida Entomologist, 86, 211Á216. Eitam, A., Sivinski, J., Holler, T., and Aluja, M. (2004), ‘Biogeography of Braconid Parasitoids of the Caribbean Fruit Fly (Diptera: Tephritidae) in Florida’, Annals of the Entomological Society of America, 97, 928Á939. Feder, J.L. (1995), ‘The Effects of Parasitoids on Sympatric Host Races of Rhagoletis pomonella’, Ecology, 76, 801Á813. Garc´ıa, J.L., and Montilla, R. (2001), ‘Coptera haywardi Loia´cono (Hymenoptera: Diapriidae) parasitoide de de Anastrepha spp. (Diptera: Tephritidae) en Venezuela’, Entomo- tro´pica (Venezuela), 16, 191Á195. Garc´ıa-Medel, D., Sivinski, J., D´ıaz-Fleischer, F., Ram´ırez-Romero, R., and Aluja, M. (2007), ‘Foraging Behavior by Six Fruit Fly Parasitoids (Hymenoptera: Braconidae) Released as Single- or Multiple-species Cohorts in Field Cages: Influence of Fruit Location and Host Density’, Biological Control, 43, 12Á22. Gates, M.W., Heraty, J.M., Schauff, M.E., Wagner, D.L., Whitfield, J.B., and Wahl, D.B. (2002), ‘Survey of the Parasitic Hymenoptera on Leafminers in California’, Journal of

Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009 Hymenoptera Research, 11, 213Á270. Gonza´lez-Herna´ndez, A., and Tejada, L.O. (1979), ‘Fluctuacio´n de la poblacio´nde Anastrepha ludens (Loew) y de sus enemigos naturales en Sargentia greggii S. Watts’, Folia Entomologica Mexicana, 41, 49Á60. Guille´n, L., Aluja, M., Equihua, M., and Sivinski, J. (2002), ‘Performance of Two Fruit Fly (Diptera: Tephritidae) Pupal Parasitoids (Coptera haywardi [Hymenoptera: Diapriidae] and Pachycrepoideus vindemiae [Hymenoptera: Pteromalidae]) under Different Environmental Soil Conditions’, Biological Control, 23, 219Á227. Guimara˜es, J.A., and Zucchi, R.A. (2004), ‘Parasitism Behavior of Three Species of Eucolinae (Hymenoptera: Figitidae) Fruit Fly Parasitoids (Diptera) in Brazil’, Neotropical Entomol- ogy, 33, 217Á224. Gut, L.J., and Brunner, J.F. (1994), ‘Parasitism of the Apple Maggot, Rhagoletis pomonella, Infesting Hawthorns in Washington’, Entomophaga, 39, 41Á49. Herna´ndez-Ort´ız, V. (1998), ‘Nueva especie de Anastrepha Schiner del grupo spatulata en Me´xico (Diptera: Tephritidae)’, Folia Entomologica Mexicana, 104, 121Á127. Herna´ndez-Ort´ız, V. (2004), ‘Two Remarkable New Species Related to Anastrepha tripunctata (Diptera: Tephritidae) with a Discussion of the Affinities of the cryptostrepha Group’, The Canadian Entomologist, 136, 759Á770. Biocontrol Science and Technology 77

Herna´ndez-Ort´ız, V., and Aluja, M. (1993), ‘Listado de especies del ge´nero neotropical Anastrepha (Diptera: Tephritidae) con notas sobre su distribucio´n y plantas hospederas’, Folia Entomolo´gica Mexicana, 88, 89Á105. Herna´ndez-Ort´ız, V., Manrique-Sade, P., Delf´ın-Gonza´lez, H., and Novelo-Rinco´n, L. (2002), ‘First Report of Anastrepha compressa in Mexico and New Records for other Anastrepha Species in the Yucata´n Peninsula (Diptera: Tephritidae)’, Florida Entomologist, 85, 389Á391. Herna´ndez-Ortiz, V., Pe´rez-Alonso, R., and Wharton, R.A. (1994), ‘Native Parasitoids Associated with the Genus Anastrepha (Dip.: Tephritidae) in Los Tuxtlas, Veracruz, Mexico’, Entomophaga, 39, 171Á178. Herrera, A.L. (1905), ‘El gusano de la naranja’, Bolet´ın de la Sociedad Agr´ıcola Mexicana, 14, 61Á69. Hoffmeister, T. (1990), ‘Zur Struktur und Dynamik des Parasitoidenkomplexes der Kirschenfliege Rhagoletis cerasi L. (Diptera: Tephritidae) auf Kirschen und Hecken- kirschen’, Mitteilungen der Deutschen Gesellschaft fu¨r Allgemeine und Angewandte En- tomolgie, 7, 546Á551. Jime´nez-Jime´nez, E. (1956), ‘Las moscas de la fruta y sus enemigos naturales’, Fito´filo (Me´xico),9,4Á11. Jime´nez-Jime´nez, E. (1961), ‘Resumen de los trabajos de control biolo´gico que se efectu´an en Me´xico para el combate de las plagas agr´ıcolas’, Fito´filo (Mexico), 32, 9Á15. Katiyar, K.P., Camacho, J., Geraud, F., and Matheus, R. (1995), ‘Parasitoides hymeno´pteros de moscas de las frutas (Diptera: Tephritidae) en la regio´n occidental de Venezuela’’, Revista Facultad Agronom´ıa (Venezuela), 12, 303Á312. LaSalle, J. (1993), ‘Parasitic Hymenoptera, Biological Control, and Biodiversity’,in Hymenoptera and Biodiversity, eds. J. LaSalle and I. Gauls, Wallingford, UK: CABI, pp. 197Á215. Loia´cono, M.S. (1981), ‘Notas sobre Diapriinae neotropicales (Hymenoptera: Diapriidae)’, Revista de la Sociedad Entomolo´gica Argentina, 40, 237Á241. Lo´pez, M., Aluja, M., and Sivinski, J. (1999), ‘Hymenopterous Larval-Pupal and Pupal Parasitoids of Anastrepha Flies (Diptera: Tephritidae) in Mexico’, Biological Control, 15, 119Á129. McPhail, M., and Guiza, F.E. (1956), ‘An Oviposition Medium for the Mexican Fruit Fly’, Journal of Economic Entomology, 49, 570. Miranda, M. (2002), ‘Patrones demogra´ficos y de comportamiento de dos endoparasitoides (Hymenoptera: Braconidae) de moscas de la fruta del ge´nero Anastrepha (Diptera: Tephritidae)’, PhD Thesis, Me´xico D.F.: Instituto de Ecolog´ıa, 175 p. UNAM. Montoya, P., Liedo, P., Benrey, B., Cancino, J., Barrera, J.F., Sivinski, J., and Aluja, M. (2000), ‘Biological Control of Anastrepha spp. (Diptera: Tephritidae) in Mango Orchards through Augmentative Releases of Diachasmimorpha longicaudata (Ashmead) (Hymenoptera: Braconidae)’, Biological Control, 18, 216Á224. Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009 Morgan, P.B. (1986), ‘Mass Culturing Microhymenopteran Pupal Parasites (Hymenoptera: Pteromalidae) of Filth Breeding Flies’, Miscelaneous Publication of Entomological Society of America, 61, 77Á87. Norrbom, A.L. (2004), ‘Host Plant Database for Anastrepha and Toxotrypana (Diptera: Tephritidae: Toxotrypanini)’, Diptera Data Dissemination Disk (CD-ROM)2. Ovruski, S.M. (1994), ‘Comportamiento de deteccio´n del hue´sped de Aganaspis pelleranoi (Bre`thes) (Hymenoptera: Cynipoidea, Eucoilidae) parasitoide de de Ceratitis capitata (Wied.)(Diptera: Tephritidae)’, Revista de la Sociedad Entomolo´gica Argentina, 53, 121Á127. Ovruski, S.M., and Aluja, M. (2002), ‘Mating Behavior of Aganaspis pelleranoi (Bre`thes) (Hymenoptera: Figitidae, Eucoilinae), a Fruit Fly (Diptera: Tephritidae) Larval Parasitoid’, Journal of Insect Behavior, 15, 139Á151. Ovruski, S., Aluja, M., Sivinski, J., and Wharton, R. (2000), ‘Hymenopteran Parasitoids on Fruit Infesting Tephritidae (Diptera) in Latin America and the Southern United States: Diversity, Distribution, Taxonomic Status and their Use in Fruit Fly Biological Control’, Integrated Pest Management Reviews,5,81Á107. Ovruski, S.M., Schliserman, P., and Aluja, M. (2004), ‘Indigenous Parasitoids (Hymenoptera) Attacking Anastrepha fraterculus and Ceratitis capitata (Diptera: Tephritidae) in Native and Exotic Host Plants in Northwestern Argentina’, Biological Control, 29, 43Á57. 78 M. Aluja et al.

Ovruski, S.M., Wharton, R.A., Schliserman, P., and Aluja, M. (2005), ‘Abundance of Anastrepha fraterculus (Diptera: Tephritidae) and Its Associated Native Parasitoids (Hymenoptera) in ‘‘Feral’’ Guavas Growing in the Endangered Northernmost Yungas Forests of Argentina with an Update on the Taxonomic Status of Opiine Parasitoids Previously Reported in this Country’, Environmental Entomology, 34, 807Á818. Plummer, C., and McPhail, M. (1941), ‘The Yellow Chapote, a Native Host of the Mexican Fruit Fly’, USDA Technical Bulletin, 775, 1Á12. Purcell, M. (1998), ‘Contribution of Biological Control to Integrated Pest Management of Tephritid Fruit Flies in the Tropics and Subtropics’, Integrated Pest Management Reviews,3, 63Á83. Simberloff, D., and Stiling, P. (1996), ‘How Risky is Biological Control?’, Ecology, 1965Á1974. Sivinski, J. (1991), ‘The Influence of Host Fruit Morphology on Parasitization Rates of the Caribbean Fruit Fly, Anastrepha suspensa’, Entomophaga, 36, 447Á454. Sivinski, J., and Aluja, M. (2003), ‘The Evolution of Ovipositor Length in the Parasitic Hymenoptera and the Search for Predictability in Biological Control’, Florida Entomologist, 86, 143Á150. Sivinski, J., and Smittle, B. (1990), ‘Effects of Gamma Radiation on the Development of the Caribbean Fruit Fly (Anastrepha suspensa) and the Subsequent Development of Its Parasite Diachasmimorpha longicaudata’, Entomologia Experimentalis et Applicata, 55, 295Á297. Sivinski, J., Calkins, C., Baranowski, R., Harris, D., Brambila, J., Diaz, J., Burns, R., Holler, T., and Dodson, G. (1996), ‘Suppression of a Caribbean Fruit Fly (Anastrepha suspensa [Loew]) Population through Augmented Releases of the Parasitoid Diachasmimorpha longicaudata (Ashmead) (Hymenoptera: Braconidae)’, Biological Control, 6, 177Á185. Sivinski, J., Aluja, M., Holler, T., and Eitam, A. (1998a), ‘Phenological Comparison of Two Braconid Parasitoids of the Caribbean Fruit Fly (Diptera: Tephritidae)’, Environmental Entomology, 27, 360Á365. Sivinski, J., Aluja, M., and Holler, T. (1999), ‘The Distributions of the Caribbean Fruit Fly, Anastrepha suspensa (Tephritidae) and Its Parasitoids (Hymenoptera: Braconidae) Within the Canopies of Host Trees’, Florida Entomologist, 82, 72Á81. Sivinski, J., Aluja, M., and Lo´pez, M. (1997), ‘Spatial and Temporal Distributions of Parasitoids of the Mexican Anastrepha Species (Diptera: Tephritidae) Within the Canopies of Fruit Trees’, Annals of the Entomological Society of America, 90, 604Á618. Sivinski, J., Vulinec, K., Menezes, E., and Aluja, M. (1998b), ‘The Bionomics of Coptera haywardi (Ogloblin) (Hymenoptera: Diapriidae) and other Pupal Parasitoids of Tephritid Fruit Flies (Diptera)’, Biological Control, 11, 193Á202. Sivinski, J., Pin˜ero, J., and Aluja, M. (2000), ‘The Distributions of Parasitoids (Hymenoptera) of Anastrepha Fruit Flies (Diptera: Tephritidae) Along an Altitudinal Gradient in Veracruz, Mexico’, Biological Control, 18, 258Á269.

Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009 Sivinski, J., Vulinec, K., and Aluja, M. (2001), ‘Ovipositor Length in a Guild of Parasitoids (Hymenoptera: Braconidae) Attacking Anastrepha spp. Fruit Flies (Diptera: Tephritidae) in Southern Mexico’, Annals of the Entomological Society of America, 94, 886Á895. Vargas, R.I., and Ramadan, M.M. (2000), ‘Comparisons of Demographic Parameters: Six Parasitoids (Hymenoptera: Braconidae) and Their Fruit Fly (Diptera: Tephritidae)’,in Area-Wide Control of Fruit Flies and other Insect Pests, ed. K.H. Tan, Penang, Malaysia: Penerbit Universiti Sains Malaysia, pp. 733Á737. Vargas, R.I., Ramadan, M.M., Hussain, T., Mochizuki, N., Bautista, R.C., and Stark, J.D. (2002), ‘Comparative Demography of Six Fruit Fly (Diptera: Tephritidae) Parasitoids (Hymenoptera: Braconidae)’, Biological Control, 25, 30Á40. Wharton, R. (1983), ‘Variation in Opius hirtus Fischer and Discussion of Desmiostoma Foerster (Hymenoptera: Braconidae)’, Proceedings of the Entomological Society of Washington, 85, 327Á330. Wharton, R. (1989), ‘Biological Control of Fruit-Infesting Tephritidae’,inFruit Flies of Economic Importance 87, ed. R. Cavalloro, Rotterdam: A.A. Balkema, pp. 323Á332. Wharton, R., and Gilstrap, F. (1983), ‘Key to and Status of Opiinae Braconid (Hymenoptera) Parasitoids Used in Biological Control of Ceratitis and Dacus s.l. (Diptera: Tephritidae)’, Annals of the Entomological Society of America, 76, 721Á742. Biocontrol Science and Technology 79

Wharton, R.A., and Marsh, P.M. (1978), ‘New-World Opiinae (Hymenoptera: Braconidae) Parasitic on Tephritidae (Diptera)’, Journal of the Washington Academy Science, 68, 147Á 167. Wharton, R.A., Gilstrap, F.E., Rhode, R.H., Fischel, M.M., and Hart, W.G. (1981), ‘Hymenopterous Egg-Pupal and Larval-Pupal Parasitoids of Ceratitis capitata and Anastrepha spp. (Dip.: Tephritidae) in Costa Rica’, Entomophaga, 26, 285Á290. Wharton, R.A., Ovruski, S.M., and Gilstrap, F.E. (1998), ‘Neotropical Eucoilidae (Cynipoi- dea) Associated with Fruit-infesting Tephritidae, with New Records from Argentina, Bolivia and Costa Rica’, Journal of Hymenoptera Research, 7, 102Á115. Wong, T.T.Y., and Ramadan, M.M. (1992), ‘Mass Rearing Biology of Larval Parasitoids (Hymenoptera: Braconidae, Opiinae) of Tephritid Flies (Diptera: Tephritidae) in Hawaii’, in Advances in Insect Rearing for Research and Pest Management, eds. T.E. Anderson and N.C. Leppla, Boulder, CO: Westview Press, pp. 405Á426.

Appendix1. Composition of the artificial diet used to rear A. ludens larvae under laboratory conditions.

Amount Ingredients

100 g Dried Yeast (Type B-Torula), Lake States, Rhinelander, WI, USA 100 g Natural wheat germ, Nutrisa SA de CV, Me´xico DF 100 g Refined sugar 150 g Sugar cane bagass (from local sugar refinery) OR Corn cob fractions, Mt. Pulaski, Products, Inc. 8 g Sodium benzoate, Baker (J.T. Baker SA de CV, Xalostoc, Edo. de Me´xico) 2 mL Hydrochloric acid, Baker (J.T. Baker SA de CV, Xalostoc, Edo. de Me´xico) 2 u Viterra Plus capsules, Pfizer (Pfizer SA de CV, Toluca, Edo. de Me´xico) 750 mL Distilled water

These amounts are recommended for seeding of 2 mL of eggs and the production of 2,500 to 3,000 larvae. Downloaded By: [Hendrichs, Jorge] At: 15:46 6 November 2009 Biocontrol Science and Technology, Vol. 19, S1, 2009, 81Á94

The suitability of Anastrepha spp. and Ceratitis capitata larvae as hosts of Diachasmimorpha longicaudata and Diachasmimorpha tryoni: Effects of host age and radiation dose and implications for quality control in mass rearing Jorge Cancinoa*, Lia Ru´zı a, Patricia Lo´peza, and John Sivinskib

aDesarrollo de Me´todos, Campan˜a Nacional Contra Moscas de la Fruta, Tapachula, Chiapas, Me´xico; bCenter for Medical, Agricultural and Veterinary Entomology, USDA-ARS, Gainesville, FL, USA

The emergence of parasitoids from irradiated tephritid host larvae of different species and ages was evaluated. Parasitoid and fly longevity and fecundity resulting from each treatment were also assessed. Doses of 5, 10, 20, 30, 40, 50, 80, 100 and 150 Gy were applied to samples (100 larvae) of 6-, 7-, 8- and 9-day-old Anastrepha spp. larvae (A. ludens (Loew), A. obliqua (Mcquart) and A. serpentina (Wiede- mann)) and 6- and 7-day-old Ceratitis capitata (Wiedemann) larvae. Anastrepha larvae were exposed to Diachasmimorpha longicaudata (Ashmead), and C. capitata larvae to D. tryoni (Cameron). Following larval exposures of 20 Gy, fly emergence was totally suppressed in all larval ages of A. ludens and A. serpentina, while in A. obliqua and C. capitata, total suppression was achieved at 30 Gy. In all species, fly emergence decreased with increasing radiation dosages from 5 to 20 Gy. Emerged fly fertility and longevity also decreased as the radiation increased. On the other hand, parasitoids did not suffer decreases in longevity or fecundity as host radiation dose increased. Larval age at the time of irradiation did not influence emergence, longevity and fecundity of either flies or parasitoids. When the irradiated cohort size was raised to one liter of larvae (about 32,000 Anastrepha or 50,000 C. capitata larvae) a dose of 40 Gy in A. ludens, A. serpentina and A. obliqua totally suppressed fly emergence but permitted D. longicaudata emergence, while for C. capitata larvae, it was necessary to increase the dose to 60 Gy. Quality control tests under mass rearing conditions were applied to D. longicaudata reared

Downloaded By: [Hendrichs, Jorge] At: 15:47 6 November 2009 using irradiated A. ludens larvae. There was no statistical difference between parasitoids derived from irradiated or non-irradiated host for most parameters. Only percent pupation after 72 h differed, along with the consequent differences between the percent emergence and pupal weight. The conclusions drawn from this study lead to a greater flexibility in the use of irradiated hosts in the mass rearing of the fruit fly parasitoids D. longicaudata and D. tryoni. Keywords: Diachasmimorpha longicaudata; Diachasmimorpha tryoni; irradiated host; fruit fly parasitoids; mass-rearing parasitoids; quality control; Anastrepha; Ceratitis

Introduction Radiation is an important and novel technique in the mass rearing of natural enemies. Radiation has led to advances such as host storage, emergence suppression,

*Corresponding author. Email: [email protected]

ISSN 0958-3157 print/ISSN 1360-0478 online # 2009 Taylor & Francis DOI: 10.1080/09583150802379577 http://www.informaworld.com 82 J. Cancino et al.

and ease of host manipulation (Morgan, Smittle, and Patterson 1986; Sivinski and Smittle 1990; Roth, Fincher, and Summerlin 1991). The mass rearing of fruit fly parasitoids has taken an important step forward with the use of radiation. Host irradiation has permitted the suppression of emergence from unparasitized hosts without affecting the fecundity or longevity of the adult parasitoids that emerge (Sivinski and Smittle 1990; Cancino, Ru´ız, Go´mez, and Toledo 2002). The earliest experiments applied radiation to Bactrocera dorsalis (Hendel) pupae that had been previously parasitized by Diachasmimorpha longicaudata (Ashmead) in order to obtain parasitoids and sterile flies at the same time. Unfortunately, sterility was found in both species (Ramadan and Wong 1989). Sivinski and Smittle (1990) reported the emergence of D. longicaudata parasitoids and the suppression of Anastrepha suspensa (Loew) emergence when host larvae were irradiated before exposure to parasitoids. As a result, irradiation was incorporated into mass-rearing procedures for various species of fruit fly parasitoids. Some adjustments and applications were proposed by Cancino et al. (2002) using Anastrepha ludens (Loew) as host to D. longicaudata that further simplified the management of large quantities of parasitoids. Host irradiation in the mass rearing of fruit fly parasitoids is currently found useful in many laboratories (Brazil, Florida, Guatemala, Mexico and Peru) (Sivinski et al. 1996; Baeza, Sivinski, Holler, and Aluja 2002; Cancino et al. 2002). However, some effects of radiation on different species remain to be analyzed. There is little known about the effects of the radiation on larval hosts of different ages. This information is vital because the age of a host larva is a determinant of host size and weight, percent parasitism, and emergence of adult parasitoids (Wong and Ramadan 1992; Wong 1993). Other aspects, such as the optimal dose for mass- rearing in the different host species and the effects of dose on parasitoid quality control parameters must also be assessed. In this study, larvae of three species of the genus Anastrepha: A. ludens, (Loew) A. obliqua (Mcquart) and A. serpentina (Wiedemann) were evaluated at different ages as hosts to D. longicaudata. In the same fashion, different larval ages of Ceratitis capitata (Wiedemann) were compared as hosts for Diachasmimorpha tryoni (Cameron). Various radiation doses were then evaluated in each host species with their respective parasitoids. The effects of host irradiation on parameters of quality control in D. longicaudata were also determined. All these appraisals will ultimately Downloaded By: [Hendrichs, Jorge] At: 15:47 6 November 2009 improve the use of radiation in the mass rearing of larval fruit fly parasitoids.

Materials and methods Parasitoids and flies evaluated were taken from their respective colonies maintained in Moscamed-Moscafrut Program ubicated in Metapa de Dom´ınguez, Chiapas, Mexico. Adult parasitoids of D. longicaudata and D. tryoni were obtained from the Moscafrut Plant and the Biological Control Department. These strains have been maintained under laboratory conditions for over 300 generations. Larvae of A. ludens, A. obliqua and A. serpentina were obtained from the laboratory colonies of the Rearing and Colonization Department. Ceratitis capitata larvae were taken from the strain kept under mass rearing conditions. The irradiation of larvae was performed in a Gammacell 220 with a Co 60 source of g radiation. The doses were applied with a range of 2.5Á3.0 Gy/min in free oxygen. Exposure times were determined with Fricke dosimetry (IAEA 2001). Biocontrol Science and Technology 83

Irradiating host Anastrepha and C. capitata larvae at different ages and radiation effects on emergence of parasitoids Anastrepha ludens, A. obliqua and A. serpentina larvae aged 6, 7, 8 and 9 days old and C. capitata larvae aged 6 and 7 days old were subjected to irradiation doses of: 5, 10, 20, 30, 40, 50, 80, 100 and 150 Gy. The irradiated larvae of Anastrepha were then exposed to D. longicaudata and C. capitata larvae to D. tryoni as described below. Non-irradiated larvae were used as the control. A sample of 100 larvae per species, age and dose were exposed to 30:15 parasitoids for 2 h. Five- to 10-day-old parasitoids were put into a ‘Hawaii-type’ screen cage (272727 cm) (Wong and Ramadan 1992). The larval exposure unit consisted of a Petri dish cover containing larvae and diet, and then covered with a piece of organza cloth held in place with a circular plastic top. Following exposure a cylindrical plastic container (84 cm) with vermiculite was placed into each treatment and held at 268Cand60Á80% R.H. The data taken consisted of the number of emerged adults (parasitoids and flies), larval, and pupae weight.

Fecundity and longevity in adults derived from irradiated hosts Samples of 10:5 adults from each treatment were evaluated for longevity and fecundity. For each D. longicaudata treatment, 50 A. ludens larvae were exposed to parasitoids that had emerged from A. ludens, A. obliqua and A. serpentina larvae. Fifty C. capitata larvae were exposed to D. tryoni parasitoids emerged from C. capitata larvae. The larval exposure and holding were as described above. Following adult emergence, daily fecundity was calculated as the number of offspring eclosed daily per female (offspring/female/day). Fecundity of females from age 5 to 15 days was evaluated. Dead parasitoids found in each cage were removed and counted daily for 15 days. The fertility of the emerged flies was also evaluated with samples of 10:10. When females reached 12 days of age, an oviposition substrate was provided per cage for 2 h daily. The oviposition unit was a green agar ball (2 cm diameter) covered with parafilm paper (Boller 1968). The eggs oviposited in the unit were collected with a Downloaded By: [Hendrichs, Jorge] At: 15:47 6 November 2009 scalpel. A sample of 100 of the collected eggs was then immediately incubated inside a Petri dish with a piece of filter paper saturated with water. After 5 days, the eggs were observed with a stereoscope to count the number hatched. Fertility was calculated as the number of hatched eggs/female/day. Longevity, fecundity and fertility of adults emerged from irradiated larvae were compared with the control treatments of parasitoids and flies emerged from non- irradiated larvae.

Radiation dose and host emergence under mass rearing conditions In these experiments, the radiation doses that suppress emergence of unparasitized flies were analyzed. Again, larvae of A. ludens, A. obliqua and A. serpentina (8 days old) were exposed to D. longicaudata and C. capitata larvae (7 days old) were exposed to D. tryoni.ForA. ludens and A. serpentina larvae, doses of 20, 30, 40 and 50 Gy were used. In A. obliqua and C. capitata larvae, 30, 40, 50 and 60 Gy were 84 J. Cancino et al.

applied. In each treatment, 1 L of larvae (approximately 32,000 Anastrepha spp. larvae and 50,000 C. capitata larvae) were irradiated. A sample of 200 larvae from each liter was then exposed to parasitism. Exposure periods lasted 2 h and the subsequent procedures were similar to those used in mass rearing (Cancino 2000). After exposure, larvae were removed from their diet and placed in trays (84 cm) with vermiculite. Before adult emergence, 14 days following pupation, three samples of 100 pupae per treatment were taken and put into a cylindrical plastic container (84 cm). The number of emerged parasitoids and flies was counted.

Quality parameters in the mass rearing of D. longicaudata with irradiated A. ludens larvae Ten lots of mass produced D. longicaudata from irradiated and non-irradiated larvae were evaluated by applying various quality control tests. The parameters were: Percent mortality and pupation of the host at 72 h. Three samples of 100 larvae per lot were taken 72 h after exposure to parasitoids. In each sample, the numbers of pupae, live and dead larvae were counted and the data were used to calculate the percent pupation and mortality. Emergence and sex-ratio. Three samples of 100 pupae were taken per lot and put into cylindrical plastic containers (84 cm). Following emergence, the numbers of parasitoids and flies and their sex-ratio were obtained. Percentage of adults capable of flight. Three samples of 100 pupae per lot were individually introduced into black PVC tubes (1010 cm). The inside walls were covered with talcum powder. Each tube was put into a cage (0.511 m) fitted with a light source. Following emergence, the parasitoids walking inside the tube were considered ‘non-fliers’ and those outside the tube ‘fliers’. Longevity with and without food. 30:15 newly emerged adult parasitoids per lot were held in screen cage and were either provided with honey and water or deprived of both. Daily mortality inside the cages was registered from day one. The test was carried out until the 15th day when honey and water were present and only until the 7th day when no food or water were provided. Downloaded By: [Hendrichs, Jorge] At: 15:47 6 November 2009 Fecundity. One sample per lot of 30:15 was placed into a screen cage. When the females were 5 days old and continuing until they were 15 days old, daily host exposures were carried out with a Petri dish-type oviposition unit containing 50 A. ludens larvae. Following exposure, larvae were maintained in a container with vermiculite for 15 days at 268Cand60Á80% R.H. The number of offspring per day was related to the number of live females. Olfatometric measurements. Thirty groups of three parasitoid females per lot were introduced into a ‘Y’ glass tube olfactometer with arms 21 cm long and 3 cm in diameter, and separated by an angle of 858. Air flow (400 ml/min) was provided by a pressurized tank. The air flow in each arm bore volatiles from one of two sources: mango fruit infested with A. ludens larvae and uninfested mango. A positive response was defined as a walk of at least 10 cm into an arm within 5 min. Onset of oviposition activity. Samples of 10 female (5 days old) parasitoids per lot were introduced into a screen cage with an oviposition unit containing 50 A. ludens Biocontrol Science and Technology 85

larvae. Over three consecutive days, for a period of 4 h, the number of females posing and ovipositing on the artificial unit was observed and recorded hourly. Longevity and fecundity under field conditions. A sample of 100:50 parasitoids per lot were placed into a cage (110.5 m). Over 15 days (from the first to the 15th day of age), the daily mortality of both females and males was recorded, and for a period of 10 days (from the 5th to the 15th day) a fresh oviposition unit with 50 A. ludens larvae was placed inside daily. A leafy branch was added to the cage to simulate environmental conditions. This test was carried out at 25Á308C and 70Á 90% H.R. The fecundity data were calculated from offspring emergence per day and its relationship to the number of live females. Longevity data were obtained from the proportion of live parasitoids per day.

Data analysis Twenty replicates were performed for the evaluations of the effects of radiation on different aged larvae and their subsequent suitability as hosts. The data were obtained by applying a bifactorial design (factors: age of larvae and doses) and analyzed by two-way ANOVA. Data with zero value were compared using Bonferroni’s test. Fecundity of the progeny was also analyzed with two-way ANOVA and Tukey’s multiple range test was used to distinguish means. In the comparison of various radiation doses, 10 replicates were carried out and the results were analyzed using ANOVA and Tukey’s multiple range test. The means for the quality control parameters were obtained from 10 lots of D. longicaudata, and a Student’s t-test was applied to these parameters for statistical analysis. Prior to statistical analysis, data were checked for ANOVA assumptions and transformed, if needed, by log(x1), arcsine and BoxÁcox transformation.

Results Irradiating host Anastrepha and C. capitata larvae at different ages and radiation effects on the emergence of parasitoids

Downloaded By: [Hendrichs, Jorge] At: 15:47 6 November 2009 The percentage emergence and sex-ratio of adult D. longicaudata and D. tryoni parasitoids emerged respectively from Anastrepha and C. capitata larvae are shown in Table 1. The parasitoid emergence in A. ludens and C. capitata decreased with the age of larvae (two-way ANOVA A. ludens:df3, F26.98, PB0.000; C. capitata: df1, F49.42, PB0.000). This inverse relationship was not observed in the other species. However, there were significant differences between particular ages in A. obliqua and A. serpentina (two-way ANOVA A. obliqua:df3, F4.56, P0.004; A. serpentina:df3, F2.70, P0.045). Parasitoid sex ratios were not different statistically in A. obliqua and C. capitata (two-way ANOVA A. obliqua:df3, F 0.40, P0.755; C. capitata:df1, F3.30, P0.070). In A. ludens and A. serpentine, there were significant differences (two-way ANOVA A. ludens:df3, F 6.22, P0.000; A. serpentina:df3, F13.09, PB0.000), but these had no clear relationship with larval age. There were no interactions between irradiation doses and larval age in any host species. 86 J. Cancino et al.

Table 1. Mean (9 S.E.) of emergence and sex-ratio of parasitoids reared on Anastrepha spp. and C. capitata irradiated larvae at different ages.

Age of larva (days) n Parasitoid emergence (%) Sex-ratio (female/males)

A. ludens 6 158 67.5191.10 a 2.1790.12 ab 7 168 60.9091.07 b 2.4490.11 a 8 158 57.3991.11 bc 1.7690.11 b 9 175 57.0991.05 c 2.0390.11 ab A. obliqua 6 93 38.3391.06 a 1.5190.07 a 7 100 36.2891.01 ab 1.4290.07 a 8 99 32.7291.02 b 1.4890.07 a 9 96 36.9191.04 a 1.4190.06 a A. serpentina 6 102 50.1091.29 ab 5.1290.07 c 7 118 48.9291.20 ab 6.0790.64 bc 8 111 48.4491.25 b 7.9790.70 a 9 88 52.6691.41 a 7.2790.76 ab C. capitata 6 206 36.3990.79 a 7.0990.41 a 7 183 28.3390.84 b 5.9790.42 a

Means followed by different letters into the same column indicate a significant difference. Data was analyzed through two-way ANOVA followed by Tuckey Multiple Range test (80.05)

Irradiating host Anastrepha and C. capitata larvae at different ages and radiation’s effects on the emergence of flies Only in C. capitata was the fly emergence between ages statistically different. Averages of 13.1790.99 and 8.6891.05 flies emerged from 6- and 7-day-old larvae, respectively (df1, F6.31, P0.013). The fly emergence from irradiated larvae of Anastrepha spp. were not different between ages (A. ludens:df3, F1.69, P0.169; A. serpentina:df3, F0.29, P0.833; A. obliqua:df3, F2.07, P0.107). In Downloaded By: [Hendrichs, Jorge] At: 15:47 6 November 2009

Table 2. Mean (9 S.E.) of fly emergence from unparasitized hosts of different species of Anastrepha and C. capitata, exposed to various radiation doses.

Irradiation A. ludens A. obliqua A. serpentina C. capitata

Dose (Gy) Fly emergence

0 6.8890.72 a 25.4592.04 a 3.2191.03 a 16.2491.55 a 5 7.590.72 a 25.6992.07 a 4.6191.17 a 21.2591.55 a 10 1.2390.72 b 8.492.04 b 0.1491.19 b 18.3291.61 a 20 0 c 0.5392.04 c 0 c 0.1291.50 b 30 0 c 0 d 0 c 0.1091.54 b 40 to 150 0 c 0 d 0 c 0 c

Means followed by different letters into the same column indicate a significant difference. Data was analized through a two-way ANOVA followed by Tuckey Multiple Range test (80.05). Biocontrol Science and Technology 87

general, Anastrepha fly emergence began to be reduced at 10 Gy. In all species, emergence at 10 Gy was significantly less than at 5 Gy. Fewer flies emerged at each progressively higher dose (A. ludens:df2, F57.67, PB0.000; A. obliqua:df3, F80.93, PB0.000; A. serpentina:df2, F18.04, PB0.000). At 30 Gy, fly emergence was totally suppressed in the three species of Anastrepha. This complete suppression was obtained above 40 Gy in C. capitata. There were statistical differences between doses in C. capitata (df4, F220.39, P0.000) (Table 2).

Longevity, fecundity and fertility in adults derived from irradiated hosts Neither the longevity (Table 3) nor fecundity of parasitoids (Figure 1 and 2) were affected consistently by host irradiation dose, although there were some significant differences amoung host ages. Fertility in flies emerged from unparasitized irradiated larvae was very sensitive to dosage (Table 4). In Anastrepha, fertility decreased significantly at 5 Gy (A. ludens:df7, F5.95, PB0.0001; A. serpentina:df7, 78, F4.37, P0.0004; A. obliqua:df7, 78, F22.56; PB0.0001). A few flies emerged at 10 Gy but they showed abnormalities in their body (wings, legs, antennae, etc.) and it was not possible to evaluate their fertility. A clear reduction of fertility was obtained in A. ludens and A. obliqua; this was variable with A. serpentina because the oviposition unit (green ball of agar) was not an efficient substrate for this species. The larvae of C. capitata were more resistant to radiation at 10 Gy than were those of Anastrepha spp. But beyond this level, fertility was notably reduced (C. capitata:df5, 61, F13.68, PB0.0001). For practical reasons, the number of eggs oviposited by the offspring was

Table 3. Results of statistical analysis of longevity of D. longicaudata and D. tryoni adults emerged from different aged larvae of Anastrepha spp. and C. capitata exposed to different irradiation doses.

Age of larva (days) d.f. x2 P

A. ludens 6 9 4.513 0.875 7 9 23.33 0.005

Downloaded By: [Hendrichs, Jorge] At: 15:47 6 November 2009 8 9 8.52 0.482 9 9 19.97 0.018 A. obliqua 6 9 13.74 0.132 7 9 10.04 0.348 8 9 29.82 0.000 9 9 49.37 B0.0001 A. serpentina 6 9 23.78 0.005 7 9 25.53 0.002 8 9 35.79 B0.000 9 9 33.25 0.000 C. capitata 6 9 19.58 0.021 7 9 16.49 0.057 88 J. Cancino et al.

larvae age (days): 6 7 8 9 A. ludens 100 80 60 40 20 0 0 5 10 20 30 40 50 80 100 150

100 A. obliqua 80 60 40 20 0 0 5 10 20 30 40 50 80 100 150 A. serpentina 100 80 60

Longevity (%) Longevity 40 20 0 0 5 10 20 30 40 50 80 100 150

C. capitata 100 80 60 40 20 0 0 5 10 20 30 40 50 80 100 150 Doses (Gy) Figure 1. Percent adults surviving to the 15th day of Diachasmimorpha longicaudata and

Downloaded By: [Hendrichs, Jorge] At: 15:47 6 November 2009 D. tryoni emerged from different aged larvae of Anastrepha spp. and Ceratitis capitata exposed to different irradiation doses.

not recorded. Nonetheless, we observed that fewer fly eggs were oviposited by adults which had been subjected to higher doses of radiation.

Adjusting dose to avoid host emergence under mass rearing conditions The emergence of adult flies from 1 L of Anastrepha spp. and C. capitata larvae was totally suppressed at higher doses (Table 5) (A. ludens:df2,95, F529.22, PB 0.0001; A. obliqua:df1, 23, F359.47, PB0.0001; A. serpentina:df1,21, F 452.55, PB0.0001; C. capitata:df2,79, F219.53, PB0.0001). In Anastrepha larvae, doses above 30 Gy were necessary to suppress fly emergence and complete suppression of C. capitata was obtained at 50 Gy. Again, there were no negative side- effects of higher doses on parasitoid emergence or sex ratio. Biocontrol Science and Technology 89

20 A. ludens larvae age (days): 6 7 8 9 15 10 5 0 0 5 10 20 30 40 50 80 100 150

20 A. obliqua 15 10 5 0 0 5 10 20 30 40 50 80 100 150

20 A. serpentina 15

Offspring/female/day 10 5 0 0 5 10 20 30 40 50 80 100 150

10 C. capitata

5

0 0 5 10 20 30 40 50 80 100 150 Doses (Gy)

Figure 2. Means (9SE) of fecundity (offspring/female/day) of Diachasmimorpha long- icaudata and D. tryoni emerged from different aged larvae of Anastrepha spp. and Ceratitis

Downloaded By: [Hendrichs, Jorge] At: 15:47 6 November 2009 capitata exposed to different irradiation doses.

Quality parameters in the mass rearing of D. longicaudata parasitoids with irradiated hosts Only the percentage of host pupation at 72 h, pupal weight, and the percentage of D. longicaudata parasitoid emergence were statistically different between the parasitoids that developed in 45 Gy irradiated versus non-irradiated A. ludens larvae (percentage of host pupation at 72 h: t2.35, P0.03; pupae weight: t2.72, P0.008, and the percentage of parasitoid emergence: t2.48, P0.015). No significant difference was found for the other parameters evaluated (Table 6).

Discussion These results indicated broad tolerances in the use of irradiation for the rearing of fruit fly parasitoids. There were no effects of radiation dosage on host-suitability nor 90 J. Cancino et al.

Table 4. Means (9 S.E.) of fertility (percent egg hatch) of flies that emerged from irradiated larvae.

Percent egg hatch

Host Dose age (Gy) (days) A. ludens A. obliqua A. serpentina C. capitata

0 6 85.2793.07 ab 73.3195.97 ab 5.5893.63 b 73.0499.71 a 7 89.2792.66 a 78.7592.24 ab 34.8696.46 a 90.43 91.52 a 8 83.2793.31 ab 85.5792.51 a 9.1496.05 b 9 80.692.92 ab 92.6291.20 a 7.1593.05 ab 5 6 58.895.71 c 42.83910.26 c 090 b 74.7699.26 a 7 75.4693.71 abc 12.6395.86 d 11.6394 ab 82.8295.72 a 8 77.693.75 abc 51.7596.80 bc 090b 96895.95 bc 28.9497.27 cd 7.1093. ab 10 6 11.8594.55 b 7 28.61912.21b

Means followed by different letters within each column are significantly different. Data was analyzed through a ANOVA followed by Tuckey Multiple Range test (80.05).

did the age of the various larvae substantially interact with dosage. The major difference was found between Anastrepha species and C. capitata, i.e. a higher dose was required to suppress adult-host emergence in the latter, and this may be related with the larval size. During these evaluations, the mean larval weights for all ages of A. ludens and A. serpentina were above 21 mg. In A. obliqua, the minimum weight was 17.4 mg, while in C. capitata, weight never exceeded 13 mg. The effects of radiation appear related to the size of the receiving body (Balock, Burditt, and Christenson 1963; Bustos, Enkerlin, Toledo, Reyes, and Casimiro 1992). Fertility of emerged flies was affected at a lower dose, 5 Gy, in the larger Anastrepha spp. Fertility was maintained in the smaller C. capitata up to 10 Gy. Similar results have been published in diverse evaluations of fruit fly larvae irradiated during post- harvest treatments (Arthur and Wiendl 1996; Hallman and Worley 1999; Toledo, Bustos, and Liedo 2001). The studies performed by Bustos et al. (1992) provided important support for the irradiation of larval hosts prior to exposition to D. Downloaded By: [Hendrichs, Jorge] At: 15:47 6 November 2009 longicaudata parasitoids. Parasitoids which emerged from irradiated host larvae demonstrated adequate levels of longevity and fecundity. The results obtained in these evaluations demonstrate that the use of an irradiated host does not have any negative repercussions on parasitoid development. The irradiation affects fly pupal develop- ment and it is independent of parasitoid physiology (Nation, Smittle, Milne, and Dykistra 1995). The availability of irradiated larvae in Anastrepha spp. and C. capitata could be extended to other fruit fly larval parasitoids particularly larval- prepupal parasitoids of the braconid subfamily Opinae (Cancino, Ru´ız, Sivinski, Ga´lvez, and Aluja, 2009). The radiation doses that were effective for suppression of fly emergence in small lots were not effective, however, when large lots of mass reared larvae were exposed. The radiation doses used for 1 L of larvae (about 32,000 larvae in Anastrepha spp. and 50,000 in C. capitata) had to be raised to 40 Gy for A. serpentina, A. ludens and A. obliqua, and 60 Gy were necessary for C. capitata. In addition to species/size Biocontrol Science and Technology 91

Table 5. Means (9 S.E.) of fly and parasitoid emergence and sex-ratio of D. longicaudata and D. tryoni reared on Anastrepha spp. and C. capitata from 1 L of larvae irradiated at different doses.

Irradiation Fly emergence Parasitoid emergence Sex-ratio Dose (Gy) (%) (%) (female/males)

A. ludens 0 31.1391.37 a 65.4791.39 a 1.1390.03 a 20 5.3990.46 b 67.6491.33 a 1.1690.02 a 30 0.6690.21 c 67.8291.32 a 1.2090.01 a 40 0 d 68.3691.51 a 1.1690.01 a 50 0 d 68.5991.18 a 1.1790.02 a A. obliqua 0 30.9393.71 a 20.9391.77 a 1.1290.12 a 30 0.3690.28 b 20.1892.23 a 1.5590.31 a 40 0 b 22.3392.70 a 1.0390.31 a 50 0 b 20.2091.67 a 1.8190.35 a 60 0 b 21.6293.03 a 2.1890.61 a A. serpentina 0 19.0894.15 a 34.5093.66 a 1.9490.29 a 20 0.0890.08 b 41.5893.68 a 1.6890.26 a 30 0 b 41.7593.57 a 1.4790.09 a 40 0 b 40.0093.91 a 1.4090.12 a 50 0 b 43.1793.88 a 1.6190.11 a C. capitata 0 23.1591.62 a 27.5291.61 a 1.6690.20 a 30 5.2391.21 b 25.8791.73 a 1.9690.27 a 40 1.6090.30 c 25.4391.69 a 2.2190.33 a 50 0 d 26.9392.07 a 2.4090.38 a 60 0 d 24.3191.93 a 2.5890.36 a

Means followed by different letters in the same column indicate statistical difference. Data was analyzed through a ANOVA followed by Tuckey Multiple Range test (80.05).

Downloaded By: [Hendrichs, Jorge] At: 15:47 6 November 2009 affects, other factors, such as the physical condition of the larvae after they have been taken off their diet, the number and volume of larvae, and the status of the irradiator, influence the required radiation dose. Prior evaluations to determine the optimal doses for large quantities of larvae have been carried out. For example, in Mexico, in the mass rearing of D. longicaudata at the Moscafrut Plant in Metapa de Dom´ınguez, Chiapas, 45 Gy are routinely applied to 10 million A. ludens larvae daily with the objective of suppressing adult fly emergence (Cancino et al. 2002). In the mass production of D. tryoni using C. capitata larvae as hosts, 70 Gy were necessary to avoid the adult emergence of the non-parasitized flies. Several mass release field studies in Mexico have reported a high efficiency of parasitoids mass reared in irradiated hosts (Montoya et al. 2000). The longevity and fecundity data obtained from parasitoids emerged from irradiated larvae in this study confirm that these attributes do not suffer any reduction. Moreover, the analysis of quality control parameters of parasitoids reared with irradiated and non- irradiated larvae did not show any significant difference. 92 J. Cancino et al.

Table 6. Means (9 S.E.) of quality control parameters of D. longicaudata reared on irradiated to 45 Gy and non-irradiated larvae of A. ludens.

Parameter Irradiated host Non irradiated host

QUALITY OF THE PROCESS Host mortality at 72 h (%) 1.1790.20 a 1.2590.22 a Host pupation at 72 h (%) 97.0390.28 b 98.3190.24 a Pupae weight (mg) 11.6190.12 b 12.0690.12 a QUALITY OF THE PRODUCT Emergence of flies from non-exposed larvae (%) 0 b 90.1691.15 a Emergence of flies from exposed larvae (%) 0 b 4.4490.60 a Parasitoid emergence (%) 65.9491.63 a 59.6991.91 b Sex-ratio of parasitoids (female/male) 3.9590.30 a 3.8690.25 a Parasitoid fliers (%) 88.3291.18 a 88.7990.90 a Longevity and fecundity with food at 20th day (percent of alive adults) Females 69.7095.97 a 66.6795.02 a Males 66.6793.68 a 60.0094.87 a Offspring/female/day 4.2690.13 a 4.3190.12 a Longevity without food at 7th day (percent of alive adults) Female 21.5298.18 a 36.3799.65 a Male 15.0495.98 a 21.6297.07 a Behavior test Female with positive response to infested mango 61.1197.35 a 51.3996.24 a Search and oviposition activity of female Female posing 40.0092.67 a 39.3392.91 a Female ovipositing 20.0092.30 a 20.6792.75 a Evaluation under field conditions Longevity at female at 20th day (percent of alive 53.6797.82 a 5096.18 a females) Longevity of male at 20th day (percent of alive male) 4596.13 a 3797.35 a Offspring/female/day 4.6190.38 a 4.4490.25 a

Means followed by different letters by row implicate significant differences. Data was analyzed through a ANOVA followed by Tuckey Multiple Range test (80.05). Downloaded By: [Hendrichs, Jorge] At: 15:47 6 November 2009 Only the percentage of pupation at 72 h was higher in the non-irradiated larval groups. This suggests that irradiated larvae were slower in their transformation to pupae, perhaps due to damage of some component(s), structures, glands, hormones, etc., that are crucial to the pupation process. This effect was mentioned by Zde´rek (1985), who included irradiation as an important abiotic factor in affecting the pupation time. Fortunately, this delay in the pupation of irradiated larvae is not a problem in the mass rearing process of these parasitoids. The proper temperature and the artificial vermiculite medium promote a high percentage of pupation 72 h after the exposure. In general, the use of irradiation in larval hosts of fruit fly parasitoids does not cause negative effects to the mass rearing process. The use of irradiation in the mass rearing process of fruit fly parasitoids is clearly positive and even indispensable to the application of augmentative parasitoid releases as part of a fruit fly integrated pest management program. Biocontrol Science and Technology 93

Acknowledgements We thank Paula Hipo´lito and Floriberto Lo´pez for their technical help. We would also like to thank Yeudiel Go´mez and the irradiation staff at the Medfly Plant. We appreciate the comments given by Francisco D´ıaz-Fleisher and Pablo Montoya in the first draft of this manuscript. Parasitoids and flies were kindly given by Flor de Ma. Moreno, Julio Dom´ınguez, Eduardo Solis and Emilio Herna´ndez. This work was carried out thanks to the support received under contract No. 10848 with the IAEA in cooperation with the Medfly Program Á SAGARPA.

References Arthur, V., and Wiendl, F.M. (1996), ‘Desinfeswa¯odemawa¯s atacadas por Anastrepha fraterculus (Wied.) (Diptera: Tephritidae) atraves das radiawo¯es gamma do Cobalto-60’, Annais da Sociedade Entomologica do Brasil, 25, 157Á159. Baeza, L.G., Sivinski, J., Holler, T., and Aluja, M. (2002), ‘The Effects of Chilling on the Fecundity and Life Span of Mass-Reared Parasitoids (Hymenoptera:Braconidae) of the Mediterranean Fruit Fly, Ceratitis capitata (Wiedemann) (Diptera: Tephritidae)’, Biocontrol Science and Technology, 12, 205Á215. Balock, J.W., Burditt Jr, A.K., and Christenson, L.D. (1963), ‘Effects of Gamma Radiation on Various Stages of Three Fruit Fly Species’, Journal of Economic Entomology, 56, 42Á46. Boller, E.F. (1968), ‘An Artificial Oviposition Device for the European Cherry Fruit Fly, Rhagoletis cerasi’, Journal of Economic Entomology, 61, 850Á852. Bustos, R.E. ,Enkerlin, W. ,Toledo, J. ,Reyes, J. and Casimiro, A. (1992), ‘Irradiation of Mangoes as a Quarantine Treatment’,inUse of Irradiation as a Quarantine Treatment of Food and Agricultural Commodities, IAEA, Vienna, Austria, pp. 77Á90. Cancino, J. (2000), Procedimientos y fundamentos de la cr´ıa masiva de Diachasmimorpha longicaudata (Ashmead) parasitoide de moscas de la fruta. Memorias del XIII curso internacional sobre moscas de la fruta, Metapa de Dom´ınguez, Me´xico, pp. 427Á35. Cancino, J., Ru´ız, L., Go´mez, Y., and Toledo, J. (2002), ‘Irradiacio´n de larvas de Anastrepha ludens (Loew) (Diptera:Tephritidae) para inhibir la emergencia de moscas en la cr´ıa del parasitoide Diachasmimorpha longicaudata (Ashmead) (Hymenoptera: Braconidae)’, Folia Entomolo´gica Mexicana, 41 (2), 195Á208. Cancino, J., Ru´ız, L., Sivinski, J., Ga´lvez, F.O., and Aluja, M. (2009), ‘Rearing of Five Hymenopterous Larval-prepupal (Braconidae, Figitidae) and Three Pupal (Diapriidae, Chalcidoidea, Eurytomidae) Native Parasitoids of the Genus Anastrepha (Diptera: Tephritidae) on Irradiated A. ludens Larvae and Pupae’, Biocontrol Science and Technology (this issue). Hallman, J.G., and Worley, W.J. (1999), ‘Gamma Radiation Doses to Prevent Adult Emergence from Immature of Mexican and West Indian Flies (Diptera:Tephritidae). Downloaded By: [Hendrichs, Jorge] At: 15:47 6 November 2009 Subtropical Agricultural Research Center. USDA-ARS Weslaco’, Journal of Economic Entomology, 92, 967Á973. IAEA (2001), Gafchromic† Dosimetry System for SIT, Standard Operation Procedure. Division of Nuclear Tecniques in Food and Agriculture, Vienna, Austria, p. 42. Montoya, P., Liedo, P., Benrey, B., Cancino, J., Barrera, J.F., Sivinski, J., and Aluja, M. (2000), ‘Biological Control of Anastrepha spp. (Diptera: Tephritidae) in Mango Orchards through Augmentative Releases of Diachasmimorpha longicaudata (Ashmead) (Hymenoptera: Braconidae)’, Biological Control, 18, 216Á224. Morgan, P.B., Smittle, B.J., and Patterson, R.S. (1986), ‘Use of Irradiated Pupae to Mass Culture of the Microhymenopterous Pupal Parasitoid Spalangia endius Walker (Hyme- noptera: Pteromalidae). I. Musca domestica L. (Diptera: Muscidae)’, Journal of Entomo- logical Science, 21, 222Á226. Nation, J.L., Smittle, B.J., Milne, K., and Dykistra, T.M. (1995), ‘Influence of Irradiation on Development of Caribbean Fruit Fly (Diptera: Tephritidae) Larvae’, Annals of Enomolo- gical Society of America, 88 (3), 348Á352. Ramadan, M.M., and Wong, T.T.Y. (1989), ‘Effect of Gamma Radiation on Biosteres longicaudatus (Ashmead) (Hymenoptera: Braconidae), a Larval Parasitoid of Dacus dorsalis Hendel (Diptera: Tephritidae), Proceedings’, Hawaiian Entomological Society, 29, 111Á113. 94 J. Cancino et al.

Roth, J.P., Fincher, G.T., and Summerlin, J.W. (1991), ‘Suitability of Irradiated or Freeze- Killed Horn Fly (Diptera: Muscidae) Pupae as Hosts for Hymenoptera Parasitoids’, Journal of Economic Entomology, 84, 94Á98. Sivinski, J., and Smittle, B. (1990), ‘Effect of Gamma Radiation on the Development of the Caribbean Fruit Fly Anastrepha suspensa, and the Subsequent Development of its Parasite Diachasmimorpha longicaudata (Ashmead)’, Entomologia Experimentalis et Applicata, 55, 295Á297. Sivinski, J., Calkins, C., Baranowski, R., Harris, D., Brambila, J., Diaz, J., Burns, R., Holler, T., and Dodson, G. (1996), ‘Suppression of a Caribbean Fruit Fly (Anastrepha suspensa (Loew)) Population through Releases of the Parasitoid Diachasmimorpha longicaudata (Ashmead) (Hymenoptera: Braconidae)’, Biological Control, 6, 177Á185. Toledo, J., Bustos, M.E., and Liedo, P. (2001), ‘Irradiacio´n de naranja infestada por Anastrepha ludens (Loew) (Diptera: Tephritidae) como tratamiento cuarentenario’, Folia Entomolo´gica Mexicana, 40 (3), 283Á295. Wong, T.T.Y. (1993), ‘Mass-Rearing of Larval Fruit Fly Parasitoids in Hawaii’,inFruit Flies: Biology and Masagement, eds. M. Aluja and P. Liedo, New York: Springer-Verlag, Inc, pp. 257Á260. Wong, T., and Ramadan, M. (1992), ‘Mass Rearing Biology of Larval Parasitoids (Hymenoptera: Braconidae) of Tephritid Fruit Flies in Hawaii’,inAdvances in Insect Rearing for Research and Pest Management, eds. T. Anderson and N. Leppla, Boulder, CO: Westview Press, pp. 405Á476. Zda´rek, J. (1985), ‘Regulation of Pupation in Flies’,inComprehensive Insect Physiology Biochemistry and Pharmacology, eds. G.G. Kerkut and L.I. Gilbert, Oxford, Great Britain: Pergamon Press, pp. 301Á331. Downloaded By: [Hendrichs, Jorge] At: 15:47 6 November 2009 Biocontrol Science and Technology, Vol. 19, S1, 2009, 95Á109

RESEARCH ARTICLE Evaluation of sequential exposure of irradiated hosts to maximize the mass rearing of fruit fly parasitoids J. Cancinoa*, L. Ru´zı a, J. Hendrichsb, and K. Bloemc

aDesarrollo de Me´todos, Campan˜a Nacional contra Moscas de la Fruta, Central Poniente 14, 30700 Tapachula, Chiapas, Me´xico; bJoint FAO/IAEA Programme, PO Box 100, A-1400 Vienna, Austria; cCenter for Plant Health Science & Technology (CPHST), USDA-APHIS-PPQ, 1730 Varsity Drive, Suite 300, Raleigh, NC 27606, USA

A series of evaluations were carried out to assess the feasibility of sequentially exposing tephritid hosts to a primary and a secondary parasitoid to ascertain if the larvae not parasitized by the primary parasitoid could be attacked by the secondary parasitoid in the pupal stage, thus optimizing the mass production of two or more species of parasitoids. Larvae or pupae of Anastrepha ludens (Loew) were exposed either to no parasitoids, the larval parasitoid Diachasmimorpha longicaudata (Ashmead) (primary parasitoid), the pupal parasitoids Coptera haywardi (Oglobin), Dirhinus sp., and Eurytoma sivinskii (Gates & Grissell) (secondary parasitoids), or sequentially to combinations of both the larval and pupal parasitoids. As part of all evaluations, host larvae were either irradiated or unirradiated. Typically, host larvae are irradiated under parasitoid mass rearing to avoid fly emergence from unparasitized hosts. Results show that a second host exposure did not increase parasitoid production mainly due to a high incidence of mortality caused by multiparasitism. With the exception of pupae exposed to C. haywardi obtained from irradiated larvae previously exposed to D. long- icaudata, multiparasitism was around 50%. This resulted in a reduction in emergence of both parasitoids. To some extent, pupal parasitoids discriminated among pupae, preferring to oviposit in pupae that were not superparasitized previously by D. longicaudata. Notably, pupae resulting from irradiated larvae were not appropriate for the development of C. haywardi. In contrast, in the cases of Dirhinus sp. and E. sivinskii, adult parasitoids did emerge from pupae resulting from irradiated larvae, although emergence was significantly lower than when Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009 pupae from unirradiated larvae were used. Our findings offer critical insights for fine-tuning the possible use of sequential host exposure to maximize parasitoid mass production. Keywords: Tephritidae; Hymenoptera; irradiation; parasitoid mass rearing; multiparasitism; superparasitism; sequential exposure

Introduction In the mass rearing of parasitoids, it is nearly impossible to approximate 100% parasitism. There are a number of factors that prevent complete utilization of hosts, including defensive adaptations of the host (Godfray 1994; Blumberg 1997; Kraaijeveld, Hutcheson, Limentani, and Godfray 2001), parasitoid-induced mor- tality through host-feeding and superparasitism, and technical difficulties in

*Corresponding author. Email: [email protected]

ISSN 0958-3157 print/ISSN 1360-0478 online # 2009 Taylor & Francis DOI: 10.1080/09583150902764140 http://www.informaworld.com 96 J. Cancino et al.

exposing all the members of a host cohort to parasitoids during a limited period of developmental time. In the mass rearing of native parasitoids of Anastrepha fruit fly (Diptera: Tephritidae) percent parasitism is commonly well below 50%. Only in long- colonized species such as Diachasmimorpha longicaudata (Ashmead) that have been mass reared for more than 300 generations, are parasitism levels of 80% routinely reached (Biological Control Department, National Program against Fruit Flies, SAGARPA, Mexico, unpublished rearing records). Because not all hosts are attacked, the presence of fertile flies in parasitoid cohorts destined for augmentative release represents a problem in operational programs. The use of irradiated hosts, which preclude fly development but allow that of the parasitoid, has been employed as an efficient measure to obtain pure cohorts (Sivinski and Smittle 1990; Cancino, Ru´ız, Go´mez, and Toledo 2002). But despite the fact that using irradiated hosts solves the problem of fertile flies, loss of expensive host material remains as one of the major problems in parasitoid mass rearing protocols. There are two means to gain a benefit from these unparasitized hosts. The first is to apply a low radiation dose to allow both the emergence of sterile flies and parasitoids of good quality (Greany and Carpenter 2000). The other is to employ a second parasitoid that can utilize the unparasitized larvae or pupae during a second exposure (Sivinski, Vulinec, Menezes, and Aluja 1998). The first option can have the disadvantage that, at suboptimal radiation doses, flies often emerge from irradiated larvae with physical deformations in wings, antennae or feet (Bustos, Enkerlin, Toledo, Reyes, and Casimiro 1992). Furthermore, such flies exhibit low sexual competitiveness (Rull, Brunel, and Mendez 2005). The second option is to expose hosts that have not been attacked by a primary parasitoid (e.g., an egg or larvalÁ prepupal parasitoid) to a secondary parasitoid (i.e., a pupal parasitoid). For this approach to work optimally, pupal parasitoids should have the capacity to discriminate against already parasitized hosts and so avoid multiparasitism (Menezes et al. 1998; Sivinski et al. 1998). Based on the above, our aim here was to assess parasitism potential of pupal parasitoids (Coptera haywardi [Oglobin], Dirhinus sp., and Eurytoma sivinskii [Gates & Grissell] [Hymenoptera: Diapriidae, Chalcididae and Eurytomidae, respectively]) Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009 when offered pupae that stemmed from Mexican fruit fly (Anastrepha ludens [Loew]) larvae previously exposed to a larvalÁprepupal parasitoid (D. longicaudata). We also wanted to assess the effect of using irradiated larvae on pupal parasitoid performance.

Materials and methods Study site and insects All experiments were carried out in laboratories belonging to the Biological Control Department (Methods Development Unit) of the Moscamed-Moscafrut Program in Metapa, Chiapas, Mexico. A. ludens larvae, as well as larval and pupal parasitoids, were obtained from colonies maintained under laboratory conditions in the various mass rearing facilities of the Moscafrut Plant (details in Cancino 2000; Dom´ınguez, Castellanos, Herna´ndez, and Mart´ınez 2000, Cancino, Ru´ız, Sivinski, Ga´lvez, and Aluja 2009a). Biocontrol Science and Technology 97

Irradiation of A. ludens larvae Eight-day-old larvae handled as described by Cancino, Ru´ız, Lo´pez, and Sivinski (2009b), were irradiated using a Gammacell 220 irradiator (Co60 source) at a rate of 2.5Á3.0 Gy/min. Larvae were exposed until reaching a radiation dose of 40 Gy in a cylindrical plastic container (94.5 cm). Radiation times were determined using a Fricke dosimetry (IAEA 2001).

Experimental design Our study involved the following treatments: (1) irradiated and unirradiated A. ludens larvae exposed to the larvalÁpupal parasitoid D. longicaudata; (2) pupae derived from irradiated and unirradiated larvae exposed to one of three pupal parasitoids, C. haywardi, E. sivinskii and Dirhinus sp.; and (3) irradiated and unirradiated larvae exposed first to the larvalÁpupal parasitoid D. longicaudata and then the resulting pupae exposed to one of three pupal parasitoids, C. haywardi, E. sivinskii and Dirhinus sp. Our design also included irradiated and unirradiated A. ludens larvae not exposed to any parasitoid. In the case of single species exposures, unirradiated and irradiated larvae (pupae) were exposed separately to cohorts of each of the parasitoid species under study over a 10-day period (cohorts of larvae [pupae] replaced daily). We tested three cohorts (each one in a separate cage) per treatment simultaneously. In the case of sequential, multiple-species exposures, we first separately exposed both types of larvae to the larvalÁprepupal parasitoid D. longicaudata (primary parasitoid) and then, once the host larvae had pupated, to one of the following pupal parasitoids (secondary parasitoid): C. haywardi, E. sivinskii and Dirhinus. As was the case with single-species exposures, sequential, multiple-species exposures were replicated three times. We did not test sequential exposures within the pupal parasitoid guild.

Larval and pupal exposure and overall parasitoid handling procedures Exposure of irradiated larvae to D. longicaudata followed within an hour after the irradiation procedure was completed. To expose irradiated and unirradiated larvae Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009 to the parasitism of D. longicaudata (primary parasitoid), we placed 200 larvae with diet in Petri dishes (0.79 cm, depthdiam) tightly covered with a circular piece of organdy cloth (Wong and Ramadan 1992). The 8-day-old larvae were exposed for 2 h to 30 and 15 D. longicaudata adults inside ‘Hawaii type’ cages (272727 cm wooden frame covered with a 1 mm plastic mesh and with a 105 cm elliptical opening at the base) (Wong and Ramadan 1992). As noted before, every day, over a 10-day period, we removed the exposure unit (Petri dish containing 200 larvae) once the 2-h exposure period was over, and repeated the procedure the next day using a new cohort of larvae of the same age. Given that some parasitoids died along the way, every morning we replaced all dead individuals with live ones transferred from cages that had been kept under exactly the same experimental conditions (i.e., parasitoids were also offered larvae every day for 2 h). That way, we assured that each cage had the same number of females over the entire experiment and that the parasitoids used to replace dead ones were of the same age and had been exposed to the exact same experimental conditions up to that point in time. D. longicaudata 98 J. Cancino et al.

individuals were 5-days-old when first exposed to larvae (ended up being 15-days-old when experiment was ended). After exposure to parasitism, larvae were returned to plastic washbowls filled with clean rearing diet (Dom´ınguez et al. 2000) to which we had added corncob grit (Mt. Pulaski Products, IncTM, Mt. Pulaski, Ill) to facilitate handling. Larvae were allowed to feed for 24-h and then removed from the washbowl and washed with tap water to remove all diet residues. After this, they were placed in a cylindrical plastic container (94.5 cm) to which a shallow (1.5 cm) layer of vermiculite had been added to stimulate pupation (Cancino 2000). Two days later, from these containers randomly chosen cohorts of 200 pupae were removed and placed in a Petri dish (1.39 cm, depthdiam) containing a small amount of lightly moistened vermi- culite (not covering pupae) and exposed to 50 and 50 adults of one of the following three species of pupal parasitoids: C. haywardi, Dirhinus sp., and E. sivinskii. All pupal parasitoids were 8-days-old when first exposed to larvae and 18-days-old when the experiment ended. Inside the Petri dish we placed a piece of cardboard (11 cm2) to simulate leaf litter and to also simulate a dark, subterranean environment. Experimental procedure was exactly the same as with D. longicaudata (i.e., daily exposures of pupae over a 10-day period with daily replacement of dead individuals until the experiment was over). After a 24-h exposure period, we removed all pupal parasitoids and maintained the pupae in plastic containers (94.5 cm) at ca. 268C(928C) and 60Á80% relative humidity for an additional 15Á35 days to allow for life cycle completion (species vary with respect to emergence time). Unexposed larvae, unirradiated and irradiated, were handled exactly as described above, with the exception that they did not suffer parasitism or otherwise had any contact with parasitoids.

Determination of parasitism levels Puparia from each species/treatment combination were randomly divided into lots of 100 pupae each and placed in plastic containers with a small amount of lightly moistened vermiculite. On the fourth day after exposure to the pupal parasitoids (secondary exposure), we removed a sample of 10 puparia from each container for every replicate in each corresponding treatment. These puparia were used to Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009 determine if super- or multiparasitism had occurred. First, the puparia were observed under a stereomicroscope (Stemi SV6, Carl Zeiss, Me´xico D.F., Mexico) to determine presence and number of oviposition scars. Then, pupae were dissected over the following four consecutive days to assess immature-stage development (Ramadan and Beardsley 1992; Kazim´ırova and Vallo 1999). In some containers, no puparia were removed until flies and parasitoids started to emerge. Flies invariably emerged before any of the pupal parasitoids (independent of species). In the case of D. longicaudata, the majority of adults emerged before the flies, but after day three, flies and parasitoids started to emerge simultaneously. Adult parasitoid emergence was recorded daily starting on the 15th day follow- ing host-exposure. Development time of the parasitoid species was variable: D. longicaudata and E. sivinskii began to emerge at 15 days after exposure to larvae (pupae in the case of E. sivinskii), while in the cases of C. haywardi and Dirhinus sp. it took up to 35 days before any adult emerged from the puparia. Based on parasitoid and fly emergence we calculated percent parasitism. Biocontrol Science and Technology 99

Statistical analyses Analysis of adult emergence data was carried out using a model of repeated measures with a correlation structure of either compound, HuynthÁFeldt, or unstructured symmetry (Brown and Pressat 1999). In all cases, a bi-factorial design was applied, using as factors double or simple exposure, and larval host condition (irradiated or not). Comparison of means was carried out through orthogonal contrasts. For the analysis, SAS systems for Windows release 8.02 TS level 02M0 was used. The levels of multiparasitism per parasitoid species with irradiated and unirradiated larvae (pupae) were analyzed by one-way ANOVA. Data on scar numbers in relation to the number of immature stages inside pupae were submitted to a regression analysis. Alpha in all cases was 0.05.

Results Results involving the combination of D. longicaudata and C. haywardi are presented in Table 1. In the case of D. longicaudata, there was significant difference in parasitoid emergence between the hosts that were exposed to both parasitoids and those that were exposed to only one (df1, 96.812; F55.05; PB0.0001). The emergence of D. longicaudata decreased when pupae stemming from exposed larvae were utilized as a host for C. haywardi. In both cases, the emergence of D. longicaudata derived from irradiated hosts was significantly higher (df1, 95.2, F7.22, P0.0085). Importantly, there was no emergence of C. haywardi adults if females had parasitized pupae that stemmed from irradiated larvae. In both parasitoid species, percent emergence decreased when the host was exposed twice. In C. haywardi, this difference was highly significant (df1, 17.9, F102.88, PB0.0001). Different factors such as natural mortality, superparasitism, and multiple parasitism were responsible for lower parasitoid emergence relative to the number of exposed hosts. Similar results were obtained in subsequent evaluations. As in the previous case, the emergence of D. longicaudata was significantly lower when Dirhinus sp. was used as the second parasitoid (df1, 33.4, F50.78, PB0.001) (Table 2). Based on the statistical analysis, these treatments were independent and there was no interaction; analyzed separately there was no Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009 significant difference between average D. longicaudata emergence from irradiated and unirradiated larvae in both treatments (df1, 33.5, F2.24, P0.1440). In the case of Dirhinus sp., emergence was statistically similar for pupae resulting from irradiated and unirradiated larvae that had been previously exposed to D. longicaudata. However, the emergence of Dirhinus sp. from pupae not exposed as larvae to D. longicaudata was significantly different between irradiated and unirradiated larvae (df1, 35.8, F183.09, PB0.0001). When the pupae exposed to Dirhinus sp. had not been previously exposed to D. longicaudata in the larval stage, then emergence from unirradiated hosts was much higher. Average Dirhinus sp. emergence from pupae exposed and those not exposed previously as larvae to D. longicaudata was significantly different (df1, 35.8, F6.62, P0.0144) (Table 2). In trials where E. sivinskii was the secondary parasitoid, results were quite similar to those obtained with Dirhinus sp. (Table 3). As was the case in the previous experiment, there was no interaction between treatments. Accordingly, the mean percent emergence of D. longicaudata was not statistically different between 100

Table 1. Parasitoid and fly emergence (mean percentage9SE) in different treatments with one or two exposures of irradiated or unirradiated A. ludens host larvae to D. longicaudata and/or C. haywardi.

Mean proportion of adult emergence

First exposure Seconnd exposure Treatments D. longicaudata C. haywardi A. ludens

D. longicaudata C. haywardi Irradiated larvae 48.4891.53 Aa 0 0 Cancino J. Unirradiated larvae 43.4492.49 Ab 21.0092.77 a 20.4592.26 a D. longicaudata No exposure Irradiated larvae 60.9691.85 Ba 0 0

Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009 6 November 15:48 At: Jorge] [Hendrichs, By: Downloaded Unirradiated larvae 56.8891.54 Bb 0 28.6391.68 b

No exposure C. haywardi Irradiated larvae 0 0 0 al et Unirradiated larvae 0 47.4593.25 b 38.4593.47 c

No exposure No exposure Irradiated larvae 0 0 0 . Unirradiated larvae 0 0 93.0091.74 d

Means followed by different letters within columns indicate statistical differences. Parasitoid emergence was analyzed using a model of repeated measures with a compound symmetry correlation structure; fly emergence was analyzed with a HuynthÁFeldt structure. Comparison of means was carried out through orthogonal contrasts (a0.05). Table 2. Parasitoid and fly emergence (average percentage9SE) in different treatments with one or two exposures of irradiated or unirradiated A. ludens host larvae to D. longicaudata and/or Dirhinus sp. icnrlSineadTechnology and Science Biocontrol Mean proportion of adult emergence

First exposure Second exposure Treatments D. longicaudata Dirhinus sp. A. ludens

D. longicaudata Dirhinus sp. Irradiated larvae 39.7293.57 Aa 11.3591.44 Aa 0 Unirradiated larvae 34.9692.39 Aa 11.2591.29 Aa 1.3590.82 a

Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009 6 November 15:48 At: Jorge] [Hendrichs, By: Downloaded D. longicaudata No exposure Irradiated larvae 62.5093.75 Bb 0 0 Unirradiated larvae 60.9292.80 Bb 0 6.4291.15 b No exposure Dirhinus sp. Irradiated larvae 0 1.5390.81 Bb 0 Unirradiated larvae 0 48.7693.15 Bc 24.3092.98 c No exposure No exposure Irradiated larvae 0 0 0 Unirradiated larvae 0 0 84.5792.38 d

Averages followed by different letters within columns indicate statistical differences. Parasitoid and fly emergence was analyzed using a model of repeated measures with a compound symmetry correlation structure. Comparison of means was carried out through orthogonal contrasts (a0.05). 101 102

Table 3. Parasitoid and fly emergence (average percentage9SE) in different treatments with one or two exposures of irradiated or unirradiated A. ludens host larvae to D. longicaudata and/or E. sivinskii.

Mean proportion of adult emergence

First exposure Second exposure Treatments D. longicaudata E. sivinskii A. ludens .Cancino J.

D. longicaudata E. sivinskii Irradiated larvae 46.8092.23 Aa 5.0490.67 Aa 0 Unirradiated larvae 44.4092.60 Aa 9.6590.95 Ab 7.8692.06 a Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009 6 November 15:48 At: Jorge] [Hendrichs, By: Downloaded D. longicaudata No exposure Irradiated larvae 61.4792.69 Ba 0 0

Unirradiated larvae 59.3392.24 Ba 0 9.4692.00 a al et No exposure E. sivinskii Irradiated larvae 0 2.1490.58 Bc 0 Unirradiated larvae 0 47.1792.93 Bd 33.7493.85 b . No exposure No exposure Irradiated larvae 0 0 0 Unirradiated larvae 0 0 86.9291.43 c

Averages followed by different letters within columns indicate statistical differences. D. longicaudata emergence was analyzed using a model of repeated measures with a compound symmetry correlation structure. E. sivinskii and A. ludens emergence were analyzed with a non-structured model. Comparison of means was carried out through orthogonal contrasts (a0.05). Biocontrol Science and Technology 103

irradiated and unirradiated larvae analyzed by treatment (df1, 36.4, F0.80, P37.80). But as was the case with the other two pupal parasitoids, emergence of D. longicaudata decreased significantly when the host was subsequently exposed to parasitism by E. sivinskii (df1, 30.3; F78.72, PB0.0001). The emergence of E. sivinskii from pupae exposed as larvae to D. longicaudata was significantly lower than that obtained from pupae not exposed as larvae (df1, 39.5, F74.67, PB0.0001) (Table 3). Also, the emergence of E. sivinskii from pupae resulting from irradiated and unirradiated larvae was significantly different (df1, 39.5, F498.5, PB0.0001). The highest values of emergence of E. sivinskii were obtained with pupae resulting from unirradiated larvae that were not exposed to D. longicaudata. As the decreases in individual and summed parasitoid emergences with double exposures may be due to mortality caused by multiparasitism, we performed scar counts and dissections. When dissected, over 50% of sequentially exposed puparia (unirradiated) revealed multiparasitism (Figure 1). No statistical differences in multiparasitism were found in the D. longicaudata/Dirhinus sp. combination between irradiated and unirradiated hosts (df19, F0.1256, P0.7272). Similarly, there were no statistically significant differences in multiparasitism between irradiated and unirradiated hosts exposed to the combination D. longicaudata/E. sivinskii (df19, F0.80, P0.3829). However, when C. haywardi was employed as the secondary parasitoid, significantly more multiparasitism occurred in unirradiated hosts (df19, F14.2258, P0.0014). Oviposition scars resulting from the penetration of D. longicaudata’s ovipositor through the larval cuticle or by that of a pupal parasitoid which pierces the puparium directly, facilitate estimates of superparasitism. The relationship between the number of scars and the number of immature stages of D. longicaudata and C. haywardi per puparium is shown in Figure 2. Note that this relationship is stronger in D. longicaudata (i.e., r2 values were higher for both irradiated and unirradiated larvae). The relationship between the number of scars and parasitoid immature stages was also higher in D. longicaudata when the host was subsequently exposed to Dirhinus sp. and E. sivinskii (Figures 3 and 4). Furthermore, multiparasitism and heterospecific parasitism was more frequent in pupae that were not superparasitized

Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009 90 a 80 a Unirradiated a 70 Irradiated a 60 a 50

40

30 b

% Multiparasitised pupae 20

10

0 C. haywardi C. haywardi Dirhinus sp. Dirhinus sp. E. sivinskii E. sivinskii

Figure 1. Average percentage (9SE) of multiparasitized A. ludens pupae after exposure to two different fruit fly parasitoids, first to the larval parasitoid D. longicaudata and second to either the pupal parasitoids C. haywardi, Dirhinus sp., or E. sivinskii. 104 J. Cancino et al.

16 a) D. longicaudata

14 C. haywardi

12

10

8

6 r2 = 0,2446 D. longicaudata

No. of immature stages 4

2 2 r = 0,0073 C. haywardi

0 0 5 10 15 20 25 No. of scars

b) 16

14

12

10

8 r 2 = 0,3263 D. longicaudata

6

No. of immature stages 4 r 2 = 0,0004 C. haywardi 2

0 0 5 10 15 20 25 No. of scars

Figure 2. Relationship between number of pupal scars and the number of immature stages of parasitoids C. haywardi and D. longicaudata. (a) Irradiated larvae, (b) unirradiated larvae. Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009 by D. longicaudata (Figure 5). Finally, superparasitism by pupal parasitoids was higher in Dirhinus sp. and E. sivinskii.

Discussion Our results indicate that the use of a primary (D. longicaudata) and a secondary (either C. haywardi, E. sivinskii or Dirhinus sp.) parasitoid in sequential fashion to optimize their mass rearing is not a viable proposition, at least under the conditions tested and with the parasitoid species combinations used. Our results agree with first attempts to assess sequential parasitism in mass rearing. Menezes et al. (1998) reported similar results with the sequential exposure of irradiated and unirradiated Anastrepha suspensa (Loew) to D. longicaudata and C. haywardi. The results were very similar in terms of the percentages of parasitism reached by each parasitoid with unirradiated larvae. In addition, development of Biocontrol Science and Technology 105

a) 16 D. longicaudata 14 Dirhinus sp. 12 r 2 = 0,1332 D. longicaudata r 2 = 0,0675 Dirhinus sp. 10 8 6 4 No. of immature stages 2 0 0 5 10 15 20 25 30 35 40 No. of scars

b) 16

14

12

10 r2 = 0,1846 D. longicaudata 8

6

4 r2 = 0,0046 Dirhinus sp. No. of immature stages 2

0 0 5 10 15 20 25 30 35 40 No. of scars

Figure 3. Relationship between number of pupal scars and the number of immature stages of parasitoids Dirhinus sp. and D. longicaudata. (a) Irradiated larvae, (b) unirradiated larvae.

Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009 C. haywardi in pupae resulting from irradiated larvae was not viable. Even though sequential exposure of D. longicaudata with Dirhinus sp. and E. sivinskii did allow pupal parasitoid development in pupae resulting from irradiated larvae, the situation was quite similar. The main problem detected was multiparasitism, which probably increased host mortality. The phenomenon of multiparasitism is common in nature (Bautista and Harris 1997; Cusson et al. 2002; Ribeiro, d’Almeida, and Pinto de Mello 2005), and several cases have been reported in fruit flies (Pemberton and Willard 1918; van den Bosch and Haramoto 1953; Palacio, Ibrahim, and Ibrahim 1991; Leyva 1982; Bautista and Harris 1997). In the present study, superparasitism by C. haywardi (an endopar- asitoid) was relatively low, more frequent in Dirhinus sp., and most common in E. sivinskii. This situation was perhaps caused by the paucity of interspecifically recognized markers that allow ovipositing females to discriminate between parasitized and unparasitized hosts (Godfray 1994; Cusson et al. 2002). However, in our studies, oviposition by pupal parasitoids, while present under all conditions, 106 J. Cancino et al.

a) 20 18 D. longicaudata 16 E. sivinskii 14 12 10 2 D. longicaudata 8 r = 0,2501 6

No. of immature stages 4 r2 = 0,0095 E. sivinskii 2 0 0 5 10 15 20 25 30 35 No. of scars

b) 20 18 16 14 r2 = 0,3831 D. longicaudata 12 10 8 6

No. of immature stages 4

2 r2 = 0,1299 E. sivinskii 0 0 5 10 15 20 25 30 35 No. of scars

Figure 4. Relationship between number of pupal scars and the number of immature stages of parasitoids E. sivinskii and D. longicaudata. (a) Irradiated larvae, (b) unirradiated larvae. Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009 was less frequent in pupae that had been previously superparasitized by D. longicaudata (puparia with multiple oviposition scars). This may not be due to discrimination by ovipositing pupal parasitism. Pupae superparasitized by D. longicaudata suffer significant biochemical changes when the number of super- numerary immature stages is high (Lawrence 1988a,b). Such changes may partially explain lower levels of pupal parasitoid survival. Pupae resulting from irradiated larvae were less suitable hosts than those derived from unirradiated larvae for all of the pupal parasitoids examined, particularly the endoparasitic C. haywardi which in our dissections never developed past the second instar. This was perhaps due to biochemical and morphological changes resulting from the irradiation dose and timing selected. We note that when recently irradiated 4-day-old pupae, rather than irradiated larvae that are allowed to pupate, are used as hosts for pupal parasitism, then high yields of all three species of pupal parasitoids are achieved (Cancino et al. 2009a). Biocontrol Science and Technology 107

70

60 No Irradiated Irradiated larvae 50 larvae

40 C. haywardi 30 20 10 0 1>11>1

70

60 Irradiated No Irradiated 5 50 larvae larvae 4 40 3 Dirhinus sp. 30 2 20 1 10 0 1>11>1

70

60 No Irradiated Irradiated 50 larvae larvae

40

30 E. sivinskii 20

10

0 1>11>1 Larvae of D. longicaudata by host

Figure 5. Frequency distribution (shades of grey) of number of immature stages of pupal parasitoids (y-axis) in either D. longicaudata single parasitized (1) or D. longicaudata superparasitized (1) unirradiated or irradiated A. ludens host. Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009

Morgan, Smittle, and Patterson (1986) similarly found that irradiated pupae were suitable hosts for a pteromalid pupal parasitoid of calypterate flies. In the present case, presumably the type of host development parasitoids require occurs sometime between the last day of larval life and the fourth day of pupation. Given that (1) multiparasitism, as tested here, results in fewer summed parasitoids, and (2) that larval irradiation inhibits the development of the candidate pupal parasitoids, the practicality of sequential rearing with irradiated hosts seems questionable. However, there remain some possibilities of application. For example, while parasitism by Dirhinus sp. and E. sivinskii in pupae from irradiated larvae is relatively low, they might still contribute useful numbers if there were a means of separating larvae parasitized by D. longicaudata from those left unattacked. The latter could be ‘recycled’ to pupal parasitoids without the concerns raised by 108 J. Cancino et al.

potential multiparasitism. Different pupation times between parasitized and unparasitized irradiated larvae (J.C. unpublished data), may make such a scheme viable.

Acknowledgements Floriberto Lo´pez offered outstanding technical support throughout the study. We thank Larissa Guille´n and Dar´ıo Garc´ıa-Medel for support during the final stages of manuscript preparation. Mart´ın Aluja contributed significant support and significant ideas to the final presentation of this manuscript. Javier Valle-Mora proposed the use of statistical methods for the analysis of data. This work is the result of research funded by the International Atomic Energy Agency (IAEA) under contract No. 10848 and the Mexican Campan˜a Nacional contra Moscas de la Fruta (SAGARPA-IICA).

References Bautista, R., and Harris, E.J. (1997), ‘Effects of Multiparasitism on the Parasitization Behavior and Progeny Development of Oriental Fruit Fly Parasitoids (Hymenoptera: Braconidae)’, Journal of Economic Entomology, 90, 757Á764. Blumberg, D. (1997), ‘Parasitoid Encapsulation as a Defense Mechanism in the Coccoidea (Homoptera) and its Importance in Biological Control’, Biological Control, 8, 225Á236. Brown, H., and Pressat, R. (1999), Applied Mixed Models in Medicine, London, UK: John Wiley & Sons. Bustos, R.E., Enkerlin, W., Toledo, J., Reyes, J., and Casimiro, A. (1992), ‘Irradiation of Mangoes as a Quarantine Treatment’,inUse of Irradiation as a Quarantine Treatment of Food and Agricultural Commodities, Vienna: IAEA, pp. 77Á90. Cancino, J. (2000), ‘Procedimientos y fundamentos de la cr´ıa masiva de Diachasmimorpha longicaudata (Ashmead) parasitoide de moscas de la fruta’, Memorias del XIII Curso Internacional sobre Moscas de la Fruta. Programa Moscamed, DGSV-SAGARPA. Metapa de Dom´ınguez, Chis., Me´xico, pp. 427Á435. Cancino, J., Ru´ız, L., Go´mez, Y., and Toledo, J. (2002), ‘Irradiacio´n de larvas de Anastrepha ludens (Loew) (Diptera: Tephritidae) para inhibir la emergencia en la cr´ıa del parasitoide Diachasmimorpha longicaudata (Ashmead) (Hymenoptera: Braconidae)’, Folia Entomolo´- gica Mexicana, 41, 195Á208. Cancino, J., Ru´ız, L., Sivinski, J., Ga´lvez, F.O., and Aluja, M. (2009a), ‘Rearing of Five Hymenopterous LarvalÁprepupal (Braconidae, Figitidae) and Three Pupal (Diapriidae, Chalcidoidea, Eurytomidae) Native Parasitoids of the Genus Anastrepha (Diptera: Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009 Tephritidae) on Irradiated A. ludens Larvae and Pupae’, Biocontrol Science and Technology, this volume. Cancino, J., Ru´ız, L., Lo´pez, P., and Sivinski, J. (2009b), ‘The Suitability of Anastrepha spp. and Ceratitis capitata (Diptera: Tephritidae) Larvae as Hosts of Diachasmimorpha longicaudata and Diachasmimorpha tryoni (Hymenoptera: Braconidae): Effects of Host Age and Radiation Dose and Implications for Quality Control in Mass Rearing’, Biocontrol Science and Technology, (this issue). Cusson, M., Laforge, M., Re´gnie`re, J., Be´liveu, C., Trudel, D., Thireau, J.C., Bellemare, G., Keirstead, N., and Stolz, D. (2002), ‘Multiparasitism of Choristoneura fumiferana by the Ichneumonid Tranosema rostrale and the Tachinid Actia interrupta: Occurrence in the Field and Outcome of Competition under Laboratory Conditions’, Entomologia Experimentalis et Applicata, 102, 125Á133. Dom´ınguez, J., Castellanos, D., Herna´ndez, E., and Mart´ınez, E. (2000), ‘Me´todos de cr´ıa masiva de moscas de la fruta’, Memorias del XIII Curso Internacional sobre Moscas de la Fruta. Programa Moscamed DGSV-SAGARPA. Metapa de Dom´ınguez, Chiapas, Me´xico, pp. 399Á414. Godfray, H.C.J. (1994), Parasitoids, Behavioral and Evolutionary Ecology, Princeton, NJ: Princeton University Press. Biocontrol Science and Technology 109

Greany, P., and Carpenter, J. (2000), ‘Use of Nuclear Techniques in Biological Control’,in Proceedings: Area-Wide Control of Fruit Flies and Other Insect Pests, eds. Keng-Hong Tan, Penerbit Universiti Sains Malaysia, Pulau Pinang, Malaysia, pp. 221Á227. IAEA (2001) Gafchromic† Dosimetry System for SIT, Standard Operation Procedure, Joint FAO/IAEA Division of Nuclear Techniques in Food and Agriculture, pp. 42. Kazim´ırova, M., and Vallo, V. (1999), ‘Larval Morphology and Development of Coptera occidentalis’, Biocontrol, 44, 263Á280. Kraaijeveld, A.R., Hutcheson, K.A., Limentani, E.C., and Godfray, H.C.J. (2001), ‘Costs of Counterdenfenses to Host Resistance in a Parasitoid of Drosophila’, Evolution, 55, 1815Á1821. Lawrence, P. (1988a), ‘Ecdysteroid Titres and Integument Changes in Superparasitized Puparia of Anastrepha suspensa (Diptera: Tephritidae)’, Journal of Insect Physiology, 34, 603Á608. Lawrence, P. (1988b), ‘Superparasitism of the Caribbean Fruit Fly Anastrepha suspensa (Diptera: Tephritidae), by Biosteres longicaudatus (Hymenoptera: Braconidae): Implica- tions for Host Regulation’, Annals of the Entomological Society of America, 81, 233Á239. Leyva, J.L. (1982), ‘Efecto del parasitismo simple y mu´ltiple sobre la mortalidad de Anastrepha ludens (Loew)’, Tesis de Maestria en Ciencias. Colegio de Postgraduados, Chapingo, Me´xico, 89 p. Menezes, E., Sivinski, J., Holler, T., Aluja, M., Jero´nimo, F., and Ramirez, E. (1998), ‘Development of Coptera haywardi (Hymenoptera: Diapriidae) in Irradiated and Unirradiated Pupae of the Caribbean Fruit Fly and the Mediterranean Fruit Fly (Diptera: Tephritidae)’, Florida Entomologist, 81, 567Á570. Morgan, P., Smittle, B., and Patterson, R. (1986), ‘Use of Irradiated Pupae to Mass Culture the Microhymenopterous Pupal Parasitoid Spalangia endius Walker (Hymenoptera: Pteromalidae) I. Musca domestica L. (Diptera: Muscidae)’, Journal of Entomological Science, 21, 222Á227. Palacio, I.P., Ibrahim, A.G., and Ibrahim, R. (1991), ‘Interspecific Competition Among Opiine Parasitoids of the Oriental Fruit Fly, Bactrocera dorsalis (Hendel)’, Philippine Entomologist, 8, 1087Á1097. Pemberton, C.E., and Willard, H.F. (1918), ‘Interrelations of Fruit Fly Parasites in Hawaii’, Journal of Agricultural Research, 12, 285Á296. Ramadan, M.M., and Beardsley, J. (1992), ‘Supernumerary Molts in First Instar of Diachasmimorpha tryoni (Cameron) (Hymenoptera: Braconidae: Opiinae)’, Proceedings of the Hawaiian Entomological Society, 31, 235Á237. Ribeiro, A., d’Almeida, J.M., and Pinto de Mello, R. (2005), ‘Ocurrence of Multiparasitism in Third Instar Larvae and Pupae of Chrysomya megacephala (Fabricius) in Field Conditions’, Revista Brasileira de Entomologia, 85, 1Á16. Rull, J., Brunel, O., and Mendez, M.E. (2005), ‘Mass rearing history negatively affects mating

Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009 success of male Anastrepha ludens (Diptera: Tephritidae) reared for sterile insect technique programs’, Journal of Economic Entomology, 98, 1510Á1516. Sivinski, J., and Smittle, B. (1990), ‘Effect of Gamma Radiation on the Development of the Caribbean Fruit Fly Anastrepha suspensa and the Subsequent Development of its Parasite Diachasmimorpha longicaudata (Ashmead)’, Entomologia Experimentalis et Applicata, 55, 295Á297. Sivinski, J., Vulinec, K., Menezes, E., and Aluja, M. (1998), ‘The Bionomics of Coptera haywardi (Oglobin) (Hymenoptera: Diapriidae) and other Pupal Parasitoids of Teprihitid Fruit Flies’, Biological Control, 11, 193Á202. van den Bosch, R., and Haramoto, F.H. (1953), ‘Competition Among Parasites of the Oriental Fruit Fly’, Proceedings of the Hawaiian Entomological Society, 15, 201Á206. Wong, T.T.Y., and Ramadan, M.M. (1992), ‘Mass Rearing of Larval Parasitoids (Hyme- noptera: Braconidae: Opiinae) of Tephritid Flies (Diptera: Tephritidae) in Hawaii’,in Advances in Insect Rearing for Research and Pest Management, eds. T.E. Anderson and N.C. Leppla, Westwiev Press Inc., Oxford & IBH Publishing Co. Pvt. Ltd., pp. 405Á426. Biocontrol Science and Technology, Vol. 19, S1, 2009, 111Á125

RESEARCH ARTICLE Assessment of infective behaviour and reproductive potential over successive generations of entomopathogenic nematodes, Steinernema glaseri (Rhabditida: Steinernematidae), reared within radiosterilized host larvae, towards Spodoptera litura (Lepidoptera: Noctuidae) Rakesh K. Seth* and Tapan K. Barik

Department of Zoology, University of Delhi, Delhi 110 007, India

The infective behaviour of entomopathogenic nematodes (EPNs), Steinernema glaseri (Steiner), reared within radiosterilized host larvae of the tropical pest, Spodoptera litura (Fabricius), was ascertained towards the unirradiated same host (S. litura) for successive generations and compared with the infectivity of controls (EPNs derived from unirradiated host larvae). A primary goal was to establish a safe mode of transport and dispersal of EPNs without concern that uninfected, reproductively competent hosts would be inadvertently released. Based on prior studies on radiation-mediated effects, two gamma doses (40 and 70 Gy), were used for radiosterilization of last instar S. litura larvae. Tests were performed using the following parameters: Regimen I (Control) with normal infective juveniles (N-IJs) vs. normal (N) hosts; Regimen II with N-IJs vs. Irradiated hosts; Regimen III with F1 IJs (harvested from Regimen II) vs. N-hosts; and Regimen IV with F2 IJs (harvested from Regimen III) vs. N-hosts. The infective performance of F1 IJs was affected more at 70 Gy than at 40 Gy, but the effect was not great enough to nullify the infective efficiency of IJs emerged from irradiated hosts; thus, these IJs could be effectively utilized in pest biocontrol. Furthermore, the infective performance of F2 IJs was almost equivalent to that of the controls, especially at 40 Gy. Hosts radiosterilized at 70 Gy could be considered safer than those exposed to 40 Gy for inundative release of EPNs as biocontrol agents. Hosts radiosterilized at 40 Gy or 70 Gy could be conveniently used, with greater efficiency at 40 Gy, for inoculative release of EPNs for long- term pest management. Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009 Keywords: entomogenous nematodes; host-irradiation; pest management; biocontrol agents; killing efficiency; Spodoptera litura; Steinernema glaseri

Introduction Spodoptera litura (Fabricius) (Lepidoptera: Noctuidae), a common cutworm of polyphagous nature, is an economically serious pest in the Indian subcontinent (Lefroy 1908; Moussa, Zaher, and Kotby 1960; Thobbi 1961; Chari and Patel 1983), and it is considered one of the major threats to the present day intensive agriculture and changing crop patterns worldwide. Increasing environmental hazards from chemical pesticides and development of insecticide resistance in S. litura (Ramakrishnan, Saxena, and Dhingra 1984; Armes, Wightman, Jadhav, and Ranga Rao, 1997) have prompted the development of ecologically sound alternative

*Corresponding author. Email: [email protected]

First Published Online 27 May 2009 ISSN 0958-3157 print/ISSN 1360-0478 online # 2009 Taylor & Francis DOI: 10.1080/09583150902981603 http://www.informaworld.com 112 R.K. Seth and T.K. Barik

methods to control this pest. Among the eco-friendly tactics being developed for management of Spodoptera spp., biological control and other biorational approaches are paramount (Narayanan and Gopalakrishnan 1987; Seth and Sehgal 1993; Barclay and Judd 1995; Armes et al. 1997). One such ecologically compatible approach would be to release entomopatho- genic nematodes (EPNs) as potential augmentative biocontrol agents to control the pest, S. litura. EPNs often have a marked ability to find and kill their hosts rapidly. EPNs exhibit favourable characteristics including high reproductive rate, virulence, and safety for non-target organisms (Kaya 1986; Jansson 1993; Ehlers and Peters 1995; Boemare, Laumond, and Mauleon 1996; Ehlers and Hokkanen 1996). The potential of nematodes as biocontrol agents and their biology, ecology and behaviour have been studied extensively (Poinar 1983; Kaya 1985; Gaugler and Kaya 1990; Smart 1995; Grewal, Ehlers, and Shapiro-Ilan 2005). Steinernematids and heterorhabditids have been reported to infect a number of species of insects in several orders (Poinar 1975; Hom 1994). Third stage ‘dauer’ juveniles (J3) are the infective, non-feeding stage. In soil, these infective juveniles (IJs) can locate hosts with varying degrees of efficiency. Steinernematid and heterorhabditid nematodes are symbiotically associated with specific bacterial species in the genera Xenorhabdus (Thomas and Poinar 1979) and Photorhabdus (Boemare, Akhurst, and Mourant 1993), respectively. Once the nematodes have penetrated the host insect, they release their bacteria, whose multiplication produces proteolytic enzymes that kill the host within 24Á48 h (Poinar 1986). Nematodes spend their free-living stage in soil and thus soil insect pests and/or soil inhabiting stages of the foliage feeding insects can be the ideal targets. The residual effect of nematode treatment was reported to be greater than that of standard chemical pesticide (Bari and Kaya 1984). The feasibility of using entomopathogenic nematodes for the control of some noctuids was demonstrated using Steinernema spp. and Heterorhabditis spp. against Pseudaletia unipuncta Haworth (Morris 1985; Morris, Converse, and Harding 1990; Morris and Converse 1991), S. feltiae (Otio) against P. unipuncta and Spodoptera exigua (Hu¨bner) (Kaya 1985), and S. carpocapsae (Weiser) against S. exigua (Kaya and Hara 1980; Hara and Kaya 1983; Begley 1990), Helicoverpa zea (Boddie), S. littoralis (Boisduval) Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009 (Glazer, Galper, and Sharon 1991), and fennica Tauscher (West and Vrain 1997). Steinernema glaseri (Steiner) has been employed successfully for the manage- ment of target pest species (Bareth, Bhatnagar, and Sharma 2001). Nuclear techniques have been useful in elucidating and promoting the genetic and biological characteristics of insects, and radiation has played a significant role in insect pest management (North 1975; Seth and Sethi 1996). Nuclear techniques can prove useful in biological control in various ways, viz., reproductive sterilization of insect hosts, provision of non-reproductive supplemental hosts for biocontrol agents, safe transport of parasitoids in irradiated hosts, improvement of the suitability of hosts for mass rearing, and sterilization of synthetic media to maintain mass culture (Greany and Carpenter 2000). To enable biocontrol agents to be released along with their hosts into the ecosystem in an environmentally safe manner, factitious insect hosts can be used as well as radiosterilized hosts, including those of insect pest species. For this approach to be successful, it is necessary to verify the parasitization capacity of parasitoids on Biocontrol Science and Technology 113

radiosterilized hosts and their ability to remain infective after being reared within the radiosterilized hosts. Entomopathogenic nematodes may be applied in infected insect cadavers (Creighton and Fassuliotis 1985; Jansson, Lecrone, and Gaugler 1993), and in this approach, nematode-infected cadavers are disseminated and pest suppression is subsequently achieved by the progeny IJs that exit the cadavers. Field application of EPNs in infected hosts may be superior to application in aqueous suspension, in terms of infectivity, dispersal and survival (Shapiro and Glazer 1996; Shapiro and Lewis 1999). EPNs can survive dry or harsh conditions or desiccation for extended periods within host cadaver (Brown and Gaugler 1996; Koppenhofer et al. 1997). Improved persistence within the host cadavers (Perez, Lewis, and Shapiro-Ilan 2003) has been reported as compared to aqueous suspensions wherein EPNs might face osmotic stress. However, EPNs carried within infected hosts are compromised by limitations of storage and application. To an extent, this constraint can be solved by improved formulations (Shapiro-Ilan, Lewis, Behle, and Mcguire 2001). It is reported that EPNs have the ability to seek out and quickly kill hosts within 24Á48 h (Gaugler 1981) and EPN entry into hosts can occur within 6Á10 h (Lewis, Campbell, Griffin, Kaya, and Peters 2006). Therefore, after an adequate time of exposure to IJs, the host can be transported immediately to the field, but this may pose a serious limitation if some viable, non-parasitized host insects escape parasitization. This problem can be overcome by radiosterilization of hosts before host exposure to IJs, and transport to the field before the hosts die and turn into cadavers. In this manner, the IJs from infected hosts can then directly interact with the ecosystem after emergence and seek new hosts (target pests) for progeny propagation. Steinernema glaseri (Steiner) was chosen as a model entomopathogenic species in the present study due to its tropical origin (Grewal, Selvan, and Gaugler 1994) and relatively large size (Stuart, Lewis, and Gaugler 1996). S. glaseri, originally discovered in larvae of the Japanese beetle (Popillia japonica Newman), has been reported to prefer host beetles in the families Chrysomelidae, Curculionidae, Elateridae and Scarabeidae, as well as various moth larvae in the families, Galleriidae, Noctuidae and Pyralidae (Poinar 1979). Prior studies by Kondo and Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009 Ishibashi (1986); Kondo and Ishibashi (1987); Kaya and Koppenhofer (1996); and Park, Yu, Park, Choo, Bae, and Nam (2001), showed that Spodoptera litura can serve as a suitable host for S. glaseri and it has also been reported as one of the preferred hosts for S. glaseri in terms of the EPNs sensory response elicited towards insect emitted attractants (Bilgrami, Kondo, and Yoshiga 2000). Steinernema glaseri has been reported to infect last instar larvae of S. litura, which may be invaded through natural openings and the cuticle. This nematode species has been found to be slowly but steadily attracted toward hosts, followed by invasion, rapid development and the establishment of a high population of EPNs within the host (Kondo and Ishibashii 1986). EPNs invasion efficiency also has been reported to be better in last instar larvae of Spodoptera litura than in the greater wax moth, Galleria mellonella L. (Kondo and Ishibashi 1987). Seth and Barik (2007) showed that the infection rate of IJs of S. glaseri in S. litura (normal as well as radiosterilized) was quite high (79Á 100%) and that this host was suitable for rapid proliferation, growth and development of S. glaseri. 114 R.K. Seth and T.K. Barik

Since EPN infectivity is affected by intrinsic (host specific) and extrinsic factors, and the host specific chemical cues are involved in host finding by EPNs (Kaya and Gaugler 1993), there is a possibility that EPNs might be conditioned in the particular host while developing in vivo. Hence, it was presumed that EPNs, due to their broad host range, would exhibit better attraction and infectivity toward the same host species from which they emerged or most recently encountered rather than showing an innate preference for a particular host species. Therefore, the present study assessed the infective potential and proliferation of S. glaseri, reared within radiosterilized hosts (S. litura), and the infective performance was evaluated up to two successive generations on the normal (non- irradiated) same host, S. litura. Comparisons were made of the relative suitability of hosts exposed to 40 and 70 Gy of radiation and the infectivity of IJs deriving from irradiated hosts vs. those reared in non-irradiated hosts. The intent of these studies was to develop a protocol that would facilitate inoculative and inundative augmentative biological control programmes against S. litura.

Materials and methods Maintenance of host insects Spodoptera litura was mass reared on a semi-synthetic diet (Seth and Sharma 2001) at a temperature of 27.0918C, 7595% relative humidity and a regimen of 12 h light (06:00Á18:00) and 12 h dark in an insectary for the experimental investiga- tions. G. mellonella (often an acceptable host for entomogenous nematodes) was mass-reared on a semi-synthetic diet (Woodring and Kaya 1988). The culture of this moth was maintained at 29Á308C. Last instar larvae were used for parasitization by EPNs.

Maintenance of entomopathogenic nematodes An isolate of an entomogenous nematode, Steinernema glaseri procured from Biosys, USA, was maintained on G. mellonella and S. litura. Optimal environ-

Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009 mental conditions of 25918C, 7595% relative humidity and a 12 h L:12 h D photoperiod regimen were maintained in the rearing facility. Steinernema glaseri was selected for the present study due to its known persistent viability and virulence in a tropical atmosphere (Grewal et al. 1994), for assessing its potential to infect S. litura. EPN survival and viability were persistently monitored and any batch with reduced infective response and survival less than 80Á85% was not used for further bioassays. Viability criterion in the EPN population was taken as 80% or more survival with responsiveness, as suggested by Epsky and Capinera (1994). The experimental studies of EPNs were conducted using last instar (L6) S. litura larvae as hosts, because its positive geotactic behaviour for seeking a pupation site would make it most likely to be encountered by soil inhabiting EPNs. At Â258C the infective juveniles could complete the entire cycle in about 7Á9 days, after entry into the host. Biocontrol Science and Technology 115

In vivo rearing of EPNs on factitious host, Galleria mellonella Steinernema glaseri was maintained on the factitious insect host, G. mellonella, by the method of Woodring and Kaya (1988). Galleria mellonella was chosen as host for stock culture of nematodes since it appears to be a universal host to steinernematid and heterorhabditid nematodes. Multiplication of the nematodes was done by infecting 10Á15 surface-sterilized last instar G. mellonella larvae with a known number of IJs, dispersed on a Whatman #1 filter paper bed made inside a 9 cm diameter inoculation chamber (a pair of sterilized Petri dishes). Surface sterilization of the G. mellonella larvae was done with 1% formalin, followed by three washes with 0.1% formalin and treatment with sterilized distilled water. The infected larvae became flaccid at a later stage due to EPN infection. If heavily contaminated by microbes other than bacterial associates of the nematodes, the host larvae turned black and exhibited a putrid odour. To collect IJs, White Traps (White 1927) were prepared by draping Whatman #1 filter paper (9 cm diameter) over a platform (an inverted 5 cm diameter Petri dish) kept in a 9 cm diameter Petri dish. About 10 mL of 0.1% formalin was added to the large base Petri dish, and the filter paper covering the platform was in contact with the formalin solution. Surface sterilization of infected hosts with formalin was done as above, and the infected hosts were placed on the rim of the inverted Petri dish. After 8Á10 days, IJs began to emerge and migrated through the filter paper into the formalin solution. These IJs were harvested daily until their emergence was reduced considerably or ceased. After collecting the dauers (infective juveniles) from the host cadaver in White traps, they were rinsed 2Á3 times with 0.1% formalin solution. The viable IJs were allowed to pass through Whatman #42 filter paper. These IJs were stored in 0.1% formalin solution in sterilized distilled water at 25.08C for 3Á4 weeks; whereas a proportion of the IJs population was stored at 6Á88C in the refrigerator and a viable culture of EPNs could be revived from this up to 26 weeks, thereafter. IJ viability, to the level of 90%, was better retained at 258C up to the first two generations, although the freshly harvested IJs were taken for the experimental studies.

In vivo rearing of EPNs on Spodoptera litura

Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009 Similar procedures (as stated above) were adopted for rearing EPNs on S. litura, but since S. litura can be cannibalistic, individual larvae were exposed to EPNs in Petri dish chambers (5 cm diameter), rather than by mass exposure when G. mellonella larvae were used as hosts.

Irradiation of insects The irradiation facility at the Institute of Nuclear Medicine and Allied Sciences (INMAS), Ministry of Defence, Delhi, was used for irradiation of 1-day-old sixth instar (L6) S. litura larvae. Sixth instar host larvae were considered as a proper stage for irradiation due to their large size (resulting in a large harvest potential of IJs from infected hosts), and high water and nutrient content (responsible for maintaining viability of EPNs) (Seth and Barik 2007). A 60Cobalt source, emitting radiation at a 116 R.K. Seth and T.K. Barik

dose rate of 95Á110 Gy/min was employed. Radiation doses of 40 and 70 Gy were evaluated for the efficacy studies of EPNs developed within irradiated hosts.

Bioassay for assessing infective behaviour of EPNs of Steinernema glaseri developed within radiosterilized hosts The infective potential and proliferation of EPNs reared in radiosterilized hosts were assessed for two consecutive generations and compared with controls. The doses selected for radiosterilization, 40 and 70 Gy, were based upon the earlier work of Seth and Barik (2007). Various regimens evaluated for infective behaviour of EPNs were as follows: Regimen I (Control): Normal (N) IJs vs. Normal (N) host; Regimen

II: N-IJs vs. Treated (T) host; Regimen III: F1 IJs vs. N-host, (where F1 IJs were derived from Regimen II, hereafter termed as ‘F1 IJs’); Regimen IV: F2 IJs vs. N-host (where F2 IJs were derived from Regimen III, hereafter termed as ‘F2 IJs’). Regimens IIÁIV were evaluated for both the doses. Individual 1Á2-day-old L6 larvae of average weight ranging from 520 to 590 mg were placed in a Petri dish (5015 mm) lined with 2 times folded filter paper (Whatman #1). The freshly harvested IJs in all experimental regimens were taken from the culture maintained on S. litura, and transferred to each Perti dish by distributing them evenly onto the filter paper base to expose them to host larvae at the rate of 25 IJs released in 1 mL water solution per individual host (as described by Koppenhofer, Kaya, Shanmugam, and Wood 1995). Incubation was performed at a constant temperature of 258C.

Parasitization behaviour and development of EPNs Host killing efficiency of EPNs was assessed based on the time taken for an insect host to become moribund and die. These responses were recorded at 4Á6-h intervals. Morbidity is an initial behavioural response due to toxins released into the host’s haemolymph by symbiotic bacteria derived from IJs. Morbidity of infected larva represented a sluggish nature, a reduced response upon being probed, and delayed resumption of a normal posture and slight torsion in the body when turned upside Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009 down. For assessment of percent parasitization, a cohort of 45Á50 larvae bioassayed (by individual exposures) constituted each replicate. After insect death, the parasitized larvae were allowed to incubate until the next generation of IJs developed and began emerging from the host cadaver. After 7Á8 days, IJs were seen wriggling at the outer surface of the insect cadaver. At this stage, the cadavers were removed from the inoculative Petri dishes and transferred to sterilized ‘harvesting dishes’ (90 mm diameter) (also known as White trap dishes). The harvest of IJs was done by using White traps. The incubation time of EPNs (i.e., the period required from host-exposure by IJs to emergence of next generation IJs from host cadaver) was determined. Daily emergence of IJs from host cadavers and their total harvest period (duration of emergence of IJs from cadavers) were recorded. The harvest potential was determined as cumulative yield of IJs per host and IJ yield per mg host weight. Biocontrol Science and Technology 117

Analysis of data The experiments were usually replicated 10Á15 times. To assess host morbidity time and host mortality time by EPNs, incubation time (of EPNs within host until next generation IJs emerged) and total harvest potential of IJs, individual host observations were conducted and each replicate represented the mean value of observations on a set of five to seven individuals. These characteristics were studied in 15 replicates, except in the case of mortality inducing time, percent parasitization, and IJ harvest (cumulative as well as per unit fresh host weight), where 10 replicates were conducted. Data was computed for means, standard error and further analysis of variance (ANOVA) using SPSS, version 11.0 (SPSS Statistics, www.spss.com). Percentage data were transformed using arcsine âx before ANOVA. Means were separated at the 5% significance level by least significant difference (LSD) test (Snedecor and Cochran 1989).

Results The infective behaviour and proliferation of Steinernema glaseri EPNs, that were cultured in irradiated (using 40 or 70 Gy) host larvae of S. litura were ascertained for two successive generations on normal host larvae of S. litura, in different regimens as described below, to assess its potential for use in augmentative biological control.

Killing efficiency and parasitization by EPNs The time taken for EPNs to cause host morbidity was slightly delayed due to parent

host irradiation in case of F1 and F2 generation IJs (F2.89; df6, 98; P50.05) (Table 1). For instance, time to morbidity was recorded as maximum (27.9 h) for

Regimen III (F1 IJs vs. N-host) at 70 Gy, whereas it was 23.4 h in the controls. Further, EPNs caused host mortality in 46 h in the controls (Regimen I) and the mortality time was found to be slightly extended by host irradiation (F6.49; df6, 63; PB0.01) (Table 1). Hence, the time taken by EPNs to induce host morbidity and

Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009 host mortality was influenced by host irradiation, with greater impact at 70 Gy (than at 40 Gy), although the effect was minimal in the case of host-morbidity. The effect

of host irradiation on mortality inducing time was greater in Regimen III (F1 IJs vs. N-host) than in Regimen IV (F2 IJs vs. N-host) at each of the gamma doses tested. Time to mortality of host larvae exhibited by IJs in Regimen IV (F2 IJs vs. Normal host) was 64 h at 70 Gy (39% more than the controls), but at 40 Gy it was 53 h (statistically not different as compared to controls; PB0.05). Host parasitization by EPNs was adversely influenced by host irradiation (F5.15; df 6, 63; PB0.01) as compared to the controls (91.4%) (Table 1). The parasitization response of EPNs was not affected on radiosterilized hosts in Regimen

II (N-IJs vs. T-host). Further, the parasitization response by F1 EPNs (Regimen III) was determined to be less than that by F2 EPNs (Regimen IV) at 40 Gy as well as 70 Gy, but the percentage parasitization by F2 EPNs was almost equivalent to the response of normal EPNs on treated hosts at 40 or 70 Gy, although it was slightly less than the controls. 118

Table 1. Infective behaviour and parasitization of the entomopathogenic nematodes, Steinernema glaseri, reared in irradiated Spodoptera litura larvae.

Host1 irradiation Time required for Time required for dose (Gy) Regimen Nature of parasite2 Nature of host morbidity (h) mortality (h) % Parasitization

0 Gy I Normal IJs (Control) Normal host 23.4a91.1 46.1a92.3 91.4a93.8 (non-irradiated) Barik T.K. and Seth R.K. 40 Gy II Normal IJs Irradiated host (40 Gy) 24.8ab91.2 54.2b92.9 87.1ab93.9 III F1 IJs from treated host Normal Host 25.7ab91.1 59.4bcd93.2 75.1bc93.6 (40 Gy) IV F2 IJs from treated host Normal Host 24.6ab91.2 53.2ab92.5 88.7a92.9 Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009 6 November 15:48 At: Jorge] [Hendrichs, By: Downloaded 40 Gy) 70 Gy II Normal IJs Irradiated host (70 Gy) 24.9ab91.7 55.2bc93.5 83.5ab94.1 III F1 IJs from treated host Normal Host 27.9b91.3 67.3d93.3 69.7c93.4 (70 Gy) IV F2 IJs from treated host Normal Host 26.6ab91.3 64.1cd93.2 86.9ab94.3 (70 Gy)

1L6 of S. litura irradiated as 0Á1-days-old, and exposed at 1Á2-days-old, to EPNs at a dose rate of 25 IJs/host for bioassay. 2 IJs-infective juveniles of entomopathogenic nematode (EPNs), F1 IJs: IJs harvested from EPNs infecting radiosterilized host in Regimen II (Normal-IJs vs. Treated-host), F2 IJs: IJs harvested from F1 EPNs infecting Normal-host in Regimen III (F1 IJs vs. Normal-host). Percentage data were transformed (arcsine) before ANOVA, but data in table are back transformations. Means9SE followed by the same letter in a column are not significantly different at P50.05 level (ANOVA followed by LSD post-test); n10 (except n15 in case of ‘Time required for morbidity’). Table 2. Development and reproduction of the entomopathogenic nematode, Steinernema glaseri, reared in irradiated Spodoptera litura larvae.

Harvest (yield)

Host1 3 irradiation Incubation time Technology and Science Biocontrol dose (Gy) Regimen Nature of parasite2 Nature of host (h) IJs per host IJs per mg host Period (days)

0 Gy I Normal IJs (Control) Normal host 183.6a96.6 20119ab9948 34.1a91.6 12.2a90.41 (non-irradiated) 40 Gy II Normal IJs Irradiated host (40 Gy) 208.2bc98.5 17569bc9879 31.7ab91.5 11.5abc90.56 III F1 IJs from treated host Normal Host 201.1abc98.7 16224cd9726 28.6bc91.4 10.9abc90.54

Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009 6 November 15:48 At: Jorge] [Hendrichs, By: Downloaded (40 Gy) IV F2 IJs from treated host Normal Host 182.9a98.2 21182a91059 33.9a92.8 11.8ab90.71 (40 Gy) 70 Gy II Normal IJs Irradiated host (70 Gy) 221.5c99.0 16278cd9828 27.7bc91.3 10.2bc90.51 III F1 IJs from treated host Normal Host 214.1bc99.8 14779d9738 25.8c91.2 09.8c90.34 (70 Gy) IV F2 IJs from treated host Normal Host 196.2ab97.8 17602bc9880 30.8ab91.6 11.2abc90.66 (70 Gy)

1L6 of S. litura irradiated as 0Á1-days-old, and exposed as 1Á2-days-old to EPNs at a dose rate of 25 IJs/host for bioassay. 2 IJs-infective juveniles of entomopathogenic nematode (EPNs), F1 IJs: IJs harvested from EPNs infecting radiosterilized host in Regimen II (Normal-IJs vs. Treated- host), F2 IJs: IJs harvested from F1 EPNs infecting Normal -host in Regimen III (F1 IJs vs. Normal -host). 3Time required from inoculation of IJs to emergence of next generation IJs from infected host. Percentage data were transformed (arcsine) before ANOVA, but data in table are back transformations. Means9SE followed by the same letter in a column are not significantly different at P50.05 level (ANOVA followed by LSD post-test); n10 (except n15 in case of ‘Incubation period’ and ‘Harvesting period’ of IJs). 119 120 R.K. Seth and T.K. Barik

Development of EPNs in vivo and IJ harvest Host irradiation had a small prolongation effect on the incubation period of infecting EPNs (F6.81; df6, 98; PB0.01) (Table 2). The incubation period of infecting N-IJs on radiosterilized hosts (Regimen II) was extended more than that

of F1 IJs and F2 IJs, with more impact at 70 Gy, and the incubation period of infecting F1 IJs was more than that exhibited by F2 IJs due to host irradiation (at 40 Gy as well as 70 Gy). Further, the incubation period of F2 IJs, recorded as 182.9 and 196.2 h from hosts irradiated at 40 and 70 Gy, respectively, was similar to that in the controls (183.6 h). The proliferation of EPNs was assessed as cumulative IJ harvest per host and IJ harvest per mg fresh weight of host. The cumulative IJ harvest per individual host was affected by host irradiation (F7.77; df6, 63; PB0.01) (Table 2). The IJ harvest per host in the controls (Regimen 1) was ca. 20.1103 IJs and it decreased at 40 Gy to 17.5103 IJs and 16.2103 IJs in Regimens II and III, representing 12.6 and 19.3% reductions with respect to controls, respectively. At 70 Gy, it was decreased by 19% in Regimen II and by 26.5% in Regimen III. Further, in terms of IJ harvest per mg host weight, host irradiation again was found to affect the IJs emergence from infected hosts (F3.94; df6, 63; PB0.05), but this impact was less than that on cumulative IJ harvest per host. The IJ yield in Regimen III (F1 IJs vs. N-host) was less than the IJ yield in Regimen II (N-IJs vs. T-host) and in the controls, with a significant difference (PB0.05) with respect to the latter. The influence of host irradiation on the IJ harvest in Regimen III was more apparent at 70 Gy, indicating a dose dependent

debilitating effect. The IJ harvest in Regimen IV (F2 IJs vs. N-host) was more than that in Regimen II (N-IJs vs. T-host) at 40 Gy as well as 70 Gy. IJ harvest from host

cadavers was found to be affected by 9Á12% in Regimen IV (F2 IJs vs. N-host) at 70 Gy, but the degree of IJ emergence in this Regimen was almost equivalent to the controls at 40 Gy. Similarly, the harvest period of developing IJs from the host cadaver also was found to be slightly influenced by host irradiation (F5.59; df6,98; PB0.05) (Table 2). The effect of host irradiation was more apparent in the harvest period exhibited by infecting F1 IJs (Regimen III) at 70 Gy as compared to

Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009 the non-irradiated controls. The harvest period exhibited by infecting F2 IJs (Regimen IV) was almost equivalent to that of controls (Regimen I) at both the doses (40 and 70 Gy).

Discussion The favourable performance of Steinernema glaseri EPNs reared in irradiated hosts supported the feasibility of using irradiated Spodoptera litura larvae for mass rearing of this entomopathogenic nematode based upon several parameters, including time to host morbidity, time until mortality, percent parasitization, incubation time and IJ harvest (yield). In the present study, S. litura was considered for host irradiation to enable in vivo transport of EPNs, as presumably the EPNs derived from this host would be better acclimatized to act as more effective (in terms of parasitization) and elicit better orientation towards same host, S. litura, as a pest in the field (Seth and Barik 2007). Biocontrol Science and Technology 121

EPNs mass-reared in irradiated hosts (i.e., F1 progeny of EPNs that infected irradiated hosts for mass-rearing), showed little decline in performance as compared with those IJs reared in non-irradiated L6 S. litura larvae, and the parasitization performance of ensuing generation IJs of F1 EPNs, (i.e., F2 IJs that developed from the infection of F1 IJs) was better than that of F1 IJs and not drastically affected with respect to the controls (N-IJs vs. N-host). It also was noticed that the impact of host irradiation on IJ harvest per unit host weight was relatively less distinct as compared to cumulative emergence of IJs per host, which indicated that host irradiation did not critically affect the host’s nutritional quality for the developing EPNs. Use of 70 Gy for host radiosterilization would be preferable to use of 40 Gy for inundative releases because reproductive sterilization is more assured, but 40 Gy should be acceptable for inoculative releases in view of the parastization performance of these EPNs mass-reared in irradiated hosts (at 40 Gy) and its next progeny (i.e., F1 and F2 IJs), where F2 IJ parasitization performance was similar to that of controls. The level of reproductive inhibition of S. litura required for use in inoculative releases would allow for an occasional viable S. litura adult to be released. Even so, the chances of releasing reproductively competent S. litura adults using only 40 Gy would be minimal, as shown by Seth and Barik (2007), who found that 40 Gy induced 80Á91% sterility and a 28Á47% reduction in mating success with respect to control, along with reduced adult emergence (53.9%), and a pronounced (61.2%) incidence of malformations. However, using 40 Gy for host radiosterilization would better preserve the infective potential of EPNs in subsequent generations and this would be helpful in inoculative release programs. Applications of nematodes via radiosterilized infected insect hosts may have great potential for adoption in developing countries (especially tropical) because it is simple and requires no special equipment or water for application, and this would be cost effective with no need for formulation. Altogether, these findings lend support to the use of irradiated S. litura larvae for mass production of Steinernema glaseri to enable their distribution in the field without fear that reproductively competent, non- parasitized S. litura larvae might be inadvertently released. As biocontrol agents derived from these radiosterilized hosts would interact with normal existing populations of insect pests prevailing in that ecosystem exposed to other control measures, it would be desirable to study the compatibility of these biocontrol agents Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009 with other control approaches being used. Therefore, such studies on compatibility of integrated tactics involving EPNs are in progress.

Acknowledgements Financial assistance by the International Atomic Energy Agency, Vienna is gratefully acknowledged for supporting this research work under Research Contract No. IAEA/IND- 10847/RB in a Coordinated Research Project. Thanks are due to Zubeda and Mahtab Zarin for the technical support.

References Armes, N.J., Wightman, J.A., Jadhav, D.R., and Ranga Rao, G. V. (1997), ‘Status of Insecticide Resistance in Spodoptera litura in Andhra Pradesh, India’, Pesticide Science, 50, 240Á248. Barclay, H.J., and Judd, G.J.R. (1995), ‘Models for Mating Disruption by Means of Pheromone for Insect Pest-control’, Research in Population Ecology, 37, 239Á247. 122 R.K. Seth and T.K. Barik

Bareth, S.S., Bhatnagar, A., and Sharma, G. (2001), ‘Compatibility of Entomogenous Nematode, Steinernema glaseri (Steiner) with Some Insecticides Used for White Grub Management’, Pest Management and Economic Zoology, 9, 123Á128. Bari, M.A., and Kaya, H.K. (1984), ‘Evaluation of the Entomogenous Nematode Neoaplec- tana carpocapsae (Steinernema feltiae) Weiser (Rhabditida: Steinernematidae) and the Bacterium Berliner var. kurstaki for Suppression of the Artichoke Plume Moth (Lepidoptera: Pterophoridae)’, Journal of Economic Entomology, 77, 225Á229. Begley, J.W. (1990), ‘Efficacy Against Insects in Habitats Other Than Soil’,inEntomopatho- genic Nematodes in Biological Control, eds. R. Gaugler and H.K. Kaya, Boca Raton, FL: CRC Press, pp. 215Á231. Bilgrami, A.L., Kondo, E., and Yoshiga, T. (2000), ‘Experimental Models for Testing Attraction and Preferential Behaviour of Steinernema glaseri to Several Insects’, Japanese Journal of Nematology, 30, 35Á46. Boemare, N.E.R., Akhurst, R.J., and Mourant, R.G. (1993), ‘DNA Relatedness between Xenorhabdus spp. (Entobacteriaceae), Symbiotic Bacteria of Entomopathogenic Nema- todes, and a Proposal to Transfer Xenorhabdus luminescens to a New Genus, Photorhabdus gen.Nov.’, International Journal of Systematic Bacteriology, 43, 249Á255. Boemare, N.E., Laumond, C., and Mauleon, H. (1996), ‘The Entomopathogenic Nematode Bacterium Complex: Biology, Life Cycle and Vertebrate Safety’, Biocontrol Science and Technology, 6, 333Á346. Brown, I.M., and Gaugler, R. (1996), ‘Cold Tolerance of Steinernematid and Heterorhabditid Nematodes’, Journal of Thermal Biology, 21, 121Á155. Chari, M.S., and Patel, N.G. (1983), ‘Cotton Leafworm Spodoptera litura (Fabr.): Its Biology and Integrated Control Measures’, Cotton Development, 13, 7Á8. Creighton, C.S., and Fassuliotis, G. (1985), ‘Heterorhabditis sp. (Nematoda: Heterorhabditi- dae): A Nematode Parasite Isolated from the Banded Cucumber Beetle, Diabrotica balteata’, Journal of Nematology, 17, 150Á153. Ehlers, R.U., and Hokkanen, H.M.T. (1996), ‘Insect Biocontrol with Non-endemic Entomopathogenic Nematodes (Steinernema and Heterorhabditis spp.): OECD and COST Workshop on Scientific and Regulation Policy Issue’, Biocontrol Science and Technology, 6, 295Á302. Ehlers, R.U., and Peters, V. (1995), ‘Entomopathogenic Nematodes in Biological Control: Feasibility, Perspectives and Possible Risks’,inBiological Control: Benefits and Risks, eds. H.M.T. Hokkanen and J.M. Lynch, Cambridge: University Press, pp. 119Á136. Epsky, N.D., and Capinera, J.L. (1994), ‘Invasion Efficacy as a Measure of Entomogenous Nematode Steinernema carpocapsae (Rhabditida: Steinernematidae)’, Journal of Economic Entomology, 87, 366Á370. Gaugler, R. (1981), ‘Biological Control Potential of Neoaplectanid Nematodes’, Journal of Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009 Nematology, 13, 241Á249. Gaugler, R., and Kaya, H.K. (1990), Entomopathogenic Nematodes in Biological Control of Insects, Boca Raton, FL: CRC Press, p. 365. Glazer, I., Galper, S., and Sharon, E. (1991), ‘Virulence of the Nematode (Steinernematids and Heterorhabditids)-Bacteria (Xenorhabdus spp.) Complex to the Egyptian Cotton Leafworm Spodoptera littoralis (Lepidoptera: Noctuidae)’, Journal of Invertebrate Pathology, 57, 94Á 100. Greany, P.D., and Carpenter, J.E. (2000), ‘Use of Nuclear Techniques in Biological Control’,in Area-Wide Control of Fruit Flies and Other Insect Pests, Joint Proceedings of the International Conference in Area-Wide Control of Insect Pests, May 28ÁJune 2, 1998, and the Fifth International Symposium on Fruit Flies of Economic Importance, June 1Á5, 1998, ed. K.H. Tan, Penang: Penerbit Universititi Sains Malaysia, pp. 221Á227. Grewal, P.S., Selvan, S.A., and Gaugler, R. (1994), ‘Thermal Adaptation of Entomopatho- genic Nematodes Á Niche Breadth for Infection, Establishment and Reproduction’, Journal of Thermal Biology, 19, 245Á253. Grewal, P.S., Ehlers, R.U., and Shapiro-Ilan, D.J. (2005), Nematodes as Biocontrol Agents, Wallingford, UK: CABI Publishing. Biocontrol Science and Technology 123

Hara, A.H., and Kaya, H.K. (1983), ‘Susceptibility of Spodoptera exigua Pupae from Different Pupation Sites to Nematode Neoaplectana carpocapsae’, Journal of Invertebrate Pathology, 42, 418Á420. Hom, A. (1994), ‘Current Status of Entomopathogenic Nematodes’, IPM Practitioner, 16, 1Á12. Jansson, R.K. (1993), ‘Introduction of Exotic Entomopathogenic Nematodes (Rhabditida; Heterorhabditidae and Steinernematidae) for Biological Control of Insects: Potential and Problems’, Florida Entomologist, 76, 82Á91. Jansson, R.K., Lecrone, S.H., and Gaugler, R. (1993), ‘Field Efficacy and Persistence of Entomopathogenic Nematodes (Rhabditida: Steinernematidae, Heterorhabditidae) for Control of Sweet-Potato Weevil (Coleoptera: Apionidae) in Southern Florida’, Journal of Economic Entomology, 86, 1055Á1063. Kaya, H.K. (1985), ‘Entomogenous Nematodes for Insect Control in IPM Systems’,in Biological Control in Agricultural IPM Systems, eds. M.A. Hoy and D.C. Herzog, New York: Academic Press, pp. 283Á302. Kaya, H.K. (1986), ‘Steinernema feltiae: Use Against Foliage Feeding Insects and Effects on Non-target insects’,inFundamental and Applied Aspects of Invertebrate Pathology, eds. R.A. Samson, J.M. Vlak and D. Peters, Proceedings of the Fourth International Colloquium on Invertebrate Pathology: Wageningen, p. 268. Kaya, H.K., and Gaugler, R. (1993), ‘Entomopathogenetic Nematodes’, Annual Review of Entomology, 38, 181Á206. Kaya, H.K., and Hara, A.H. (1980), ‘Differential Susceptibility of Lepidopterous Pupae to the Nematode Neoaplectana carpocapsae’, Journal of Invertebrate Pathology, 36, 389Á393. Kaya, H.K., and Koppenhofer, A.M (1996), ‘Effects of Microbial and Other Antagonistic Organism and Competition on Entomopathogenic Nematodes’, Biocontrol Science and Technology, 6, 357Á371. Kondo, E., and Ishibashi, N. (1986), ‘Infectivity and Propagation of Entomogenous Nematodes, Steinernema spp., on the Common Cutworm, Spodoptera litura (Lepidoptera: Noctuidae)’, Applied Entomology and Zoology, 21, 95Á108. Kondo, E., and Ishibashi, N. (1987), ‘Comparative Infectivity and Development of the Entomogenous Nematodes, Steinernema spp., on the Lepidopterous Insect Larvae, Spodoptera litura (Noctuidae) and Galleria mellonella (Galleridae)’, Japanese Journal of Nematology, 17, 35Á41. Koppenhofer, A.M., Baur, M.E., Stock, S.P., Choo, H.Y., Chinnarsri, B., and Kaya, H.K. (1997), ‘Survival of Entomopathogenic Nematodes within Host Cadavers in Dry Soil’, Applied Soil Ecology, 6, 231Á240. Koppenhofer, A.M., Kaya, H.K., Shanmugam, S., and Wood, G.L. (1995), ‘Interspecific Competition between Steinernematid Nematodes within an Insect Host’, Journal of Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009 Invertebrate Pathology, 66, 99Á103. Lefroy, H.M. (1908), ‘The Tobacco , Prodenia littoralis’, Memoirs of Department of Agriculture, India, Entomology Series,2,79Á93. Lewis, E.E., Campbell, J., Griffin, C, Kaya, H., and Peters, A. (2006), ‘Behavioural Ecology of Entomopathogenic Nematodes’, Biological Control, 38, 66Á79. Morris, O.N. (1985), ‘Species of Agricultural Insect Pests to the Entomogenous Nematodes Steinernema feltiae and Heterorhabditis bacteriophora’, Canadian Entomologist, 11, 401Á 407. Morris, O.N., and Converse, V. (1991), ‘Effectiveness of Steinernematid and Heterorhabditid Nematodes Against Noctuid, Pyralid, and Geometrid Species in Soil’, Canadian Entomol- ogist, 123, 55Á61. Morris, O.N., Converse, V., and Harding, J. (1990), ‘Virulence of Entomopathogenic Nematode-bacteria Complexes for Larvae of Noctuids, a Geometrid and a Pyralid’, Canadian Entomologist, 122, 309Á320. Moussa, M.A., Zaher, M.A., and Kotby, V. (1960), ‘Abundance of Cotton Leafworm, Prodenia litura (F.) in Relation to Host Plants. I. Host Plants and their Effect on Biology’, Bulletin of Society of Entomology, Egypte, 44, 241Á251. 124 R.K. Seth and T.K. Barik

Narayanan, K., and Gopalakrishnan, C. (1987), ‘Effect of Entomogenous Nematode, Steinernema feltiae (Rhabditida: Steinernematidae) to the Pupa, Pre-pupa and Adult of Spodoptera litura (Noctuidae: Lepidoptera)’, Indian Journal of Nematology, 17, 273Á276. North, D.T. (1975), ‘Inherited Sterility in Lepidoptera’, Annual Review of Entomology, 20, 167Á182. Park, S.H., Yu, Y.S., Park, J.S., Choo, H.Y., Bae, S.D., and Nam, M.H. (2001), ‘Biological Control of Tobacco Cutworm, Spodoptera litura (Fabricius) with Entomopathogenic Nematodes’, Biotechnology and Bioprocess Engineering, 6, 139Á143. Perez, E.E., Lewis, E.E., and Shapiro-Ilan, D.I. (2003), ‘Impact of the Host Cadaver on Survival and Infectivity of Entomopathogenic Nematodes (Rhabditida: Steinernematidae and Heterorhabditidae) under Desiccating Conditions’, Journal of Invertebrate Pathology, 82, 111Á118. Poinar Jr, G.O. (1975), Entomogenous Nematodes. A Manual and Host List of Insect-Nematode Associations, Leiden Netherlands: E. J. Brill, p. 317. Poinar Jr, G.O. (1979), Nematodes for Biological Control of Insects, Baco Raton, FL: CRC Press, p. 277. Poinar Jr, G.O. (1983), The Natural History of Nematodes, Englewood Cliffs: Prentice- Hall, p. 143. Poinar Jr, G.O. (1986), ‘Entomophagous Nematodes’,inBiological Plant and Health Protection, ed. H. Franz, Fortschritte der Zoologie, Bd.32. G.Fischer Verlog, Stuttgart, New York, pp. 95Á121. Ramakrishnan, N., Saxena, V.S., and Dhingra, S. (1984), ‘Insecticide Resistance in the Population of Spodoptera litura (Fabr.) in Andhra Pradesh’, Pesticides, 18, 23Á27. Seth, R.K., and Barik, T.K. (2007), ‘Effect of Host Irradiation on Bio-infectivity and Proliferation Capacity of Steinernema glaseri as Entomopathogenic Nematodes on a Serious Tropical Pest, Spodoptera litura’, Journal of Nuclear Agriculture and Biology, 36, 81Á101. Seth, R.K., and Sehgal, S.S. (1993), ‘Partial Sterilizing Radiation Dose-effect on the F1 Progeny of Spodoptera litura (Fabr.): Growth, Bioenergetics and Reproductive Compe- tence’,inProceedings on Management of Insect Pests: Nuclear and Related Molecular and Genetic Techniques, Vienna, 19Á23 October 1992, eds. P. Howard-Kitto, R.F. Kelleher and G.V. Ramesh, International Atomic Energy Agency, Vienna, pp. 427Á440. Seth, R.K., and Sethi, G.R. (1996), ‘Nuclear Techniques as Efficient Methods in Entomology and Pest Management: Potential and Perspectives’, in: Isotopes and Radiations in Agriculture and Environment Research, eds. M.S. Sachdev, P. Sachdev and D.L. Deb, DAE-BRNS Symposium, Oct 1996, BARC, Trombay, pp. 219Á231. Seth, R.K., and Sharma, V.P. (2001), ‘Inherited Sterility by Substerilizing Radiation in Spodoptera litura (Lepidoptera: Noctuidae): Bioefficacy and Potential for Pest Suppression’, Florida Entomologist, 84, 183Á193. Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009 Shapiro, D.I., and Glazer, I. (1996), ‘Comparison of Entomopathogenic Nematode Dispersal from Infected Hosts Versus Aqueous Suspension’, Environmental Entomology, 25, 1455Á1461. Shapiro, D.I., and Lewis, E.E. (1999), ‘Comparison of Entomopathogenic Nematode Infectivity from Infected Hosts Versus Aqueous Suspension’, Environmental Entomology, 28, 907Á911. Shapiro-Ilan, D.I., Lewis, E.E., Behle, R.W., and Mcguire, M.R. (2001), ‘Formulation of Entomopathogenic Nematode-infected Cadavers’, Journal of Invertebrate Pathology, 78, 17Á23. Smart, G.C. (1995), ‘Entomopathogenic Nematodes for the Biological Control of Insects’, Journal of Nematology, 27, 529Á534. Snedecor, G.W., and Cochran, W.G. (1989), Statistical Methods, 8th ed, Ames, IA: Iowa State University Press. Stuart, R.J., Lewis, E.E., and Gaugler, R. (1996), ‘Selection Alters the Pattern of Emergence from the Host Cadaver in the Entomopathogenic Nematode, Steinernema glaseri’, Parasitology, 113, 183Á189. Thobbi, V.V. (1961), ‘Growth Potential of Prodenia litura in Relation to Certain Food Plants in Surat’, Indian Journal of Entomology, 23, 262Á264. Biocontrol Science and Technology 125

Thomas, G.M., and Poinar Jr, G.O. (1979), ‘Xenorhabdus gen. nov. a Genus of Entomopatho- genic Nematophilic Bacteria of the Family Enterobacteriaceae’, International Journal of Systematics in Bacteriology, 29, 352Á360. West, R.J., and Vrain, T.C. (1997), ‘Nematode Control of Black Army Cutworm (Lepidoptera: Noctuidae) under Laboratory and Field Conditions’, Canadian Entomologist, 129, 229Á239. White, G.F. (1927), ‘A Method for Obtaining Infective Nematode Larvae from Cultures’, Science, 66, 302Á303. Woodring, J.L., and Kaya, H.K. (1988), ‘Steinernematid and Heterorhabditid Nematodes: A Handbook of Biology and Techniques’, Southern Cooperative Series Bulletin 331, Arkansas Agricultural Experiment Station, Arkansas: Fayetteville, p. 30. Downloaded By: [Hendrichs, Jorge] At: 15:48 6 November 2009 Biocontrol Science and Technology, Vol. 19, S1, 2009, 127Á138

RESEARCH ARTICLE Suitability of irradiated and cold-stored eggs of Ephestia kuehniella (Pyralidae: Lepidoptera) and Sitotroga cerealella (Gelechidae: Lepidoptera) for stockpiling the egg-parasitoid Trichogramma evanescens (Trichogrammatidae: Hymenoptera) in diapause Aydin S. Tunc¸bilek*, Ulku Canpolat, and Fahriye Sumer

Department of Biology, Erciyes University, Faculty of Arts & Sciences, 38039 Kayseri, Turkey

The use of irradiated and cold-stored host eggs could be one option to facilitate the mass rearing of egg parasitoids to control lepidopteran pests. The effect on Trichogramma evanescens (L.) wasp quality after 3-month storage of host eggs of the Mediterranean flour moth, Ephestia kuehniella Zeller (Lepidoptera: Pyrali- dae) and the Angoumois grain moth, Sitotroga cerealella (Olivier) (Lepidoptera: Gelechidae), that had previously been irradiated with gamma radiation, was investigated. Efficiency of T. evanescens was studied by measuring parasitization, adult and female emergence. There was no significant difference in parasitization and in adult and female T. evanescens emergence between gamma radiation doses and the untreated control for up to 30 days for E. kuehniella eggs and, thereafter they decreased drastically as the storage time increased for up to 60 and 30 days for E. kuehniella and S. cerealella eggs, respectively. No parasitization was observed when the eggs were stored longer and then offered to T. evanescens females. Data obtained from diapaused T. evanescens stored at 38C for 20, 70, 100 and 150 days indicated that pre-storage temperatures affected the induction of diapause. It was possible to induce diapause in developmental stages of T. evanescens by exposing the immature stages (prior to the pre-pupal stage) inside host eggs to 10 and 128C for 30 days. Under these conditions, parasitoids could be stored for a period of 50 days without adverse affects on emergence. Emergence appeared to decrease with an increase in the duration of storage for a period up to 150 days for the eggs of E. kuehniella. Parasitoids failed to enter diapause when pre-storage conditions were 3 and 78C for host eggs of both E. kuehniella and S. Downloaded By: [Hendrichs, Jorge] At: 15:49 6 November 2009 cerealella. The long-term storage of parasitoids in diapause improved the mass rearing potential for lengthened releases of this species. Keywords: Trichogramma evanescens; cold storage; diapause; host egg; gamma radiation; Ephestia kuehniella; Sitotroga cerealella

Introduction Moth species of the genus Ephestia, especially the Mediterranean flour moth, Ephestia kuehniella Zeller, and Angoumois grain moth, Sitotroga cerealella (Olivier), are serious pests in cereal-based food processing facilities, stored maize and other cereals in Turkey (Ministry of Agriculture and Rural Affairs 1995a). Typically, control of these pests is undertaken by regular treatment of infested areas with a pesticide such as malathion, dichlorvos, and methyl bromide (Ministry of Agriculture

*Corresponding author. Email: [email protected]

ISSN 0958-3157 print/ISSN 1360-0478 online # 2009 Taylor & Francis DOI: 10.1080/09583150902985588 http://www.informaworld.com 128 A.S. Tunc¸bilek et al.

and Rural Affairs 1995b). The disadvantages of using chemicals, together with the effect of fumigants on the ozone layer and the development of pest resistance, have all contributed to the urgency of the current search for grain protection systems that target the pest species. Biological control is an often-underutilized component of integrated pest management of stored grains (Brower, Smith, Vail, and Flinn 1996; Scho¨ller, Prozell, Al-Kirshi, and Reichmuth 1997). Recent legislation in the USA has allowed for augmentative releases of beneficial insects in stored products. Parasitoids from the genus Trichogramma are of interest for control of pyralid moths in flourmills because they attack the egg stage and kill the pest before it reaches the larval stage and starts to produce webbing (Hansen and Jensen 2002). Eggs of E. kuehniella and S. cerealella are suitable hosts for Trichogramma evanescens (L.) (Hansen and Jensen 2002; Pitcher, Hoffmann, Gardner, Wright, and Kuhar 2002), a parasitoid that could be effectively used for suppression of stored products moths. However, if all fertile host eggs that are deployed are not parasitized, the eggs that hatch will serve to increase the number of moth larvae present. This would be unacceptable. One way to avoid this problem is to kill the moth egg embryos by gamma radiation before they are deployed in food storage facilities. Therefore, before placement in food storage facilities, the host eggs are irradiated with gamma radiation to prevent development. In our previous study (Tunc¸bilek, Canpolat, and Ayvaz 2009), we showed that the irradiated E. kuehniella and S. cerealella eggs presented to female T. evanescens in choice experiments were equally acceptable and suitable for parasitoid development. The objective of a mass rearing programme is ‘to produce the maximum quantity of quality-assured individuals by predetermined dates at a minimal cost’ (King 1993). An important aspect in mass production and deployment of biological control agents is development of storage techniques to provide flexibility and efficiency of their use, and to make standardized stocks available for use in research (Ravensberg 1992; Greenberg, Nordlund, and King 1996; Leopold 1998). Many insects and mites are able to overwinter at sub-zero temperatures in a supercooled state (Lee 1991). Few studies have been conducted to examine storage temperatures below 08Cto facilitate mass production and/or release. Short-term storage has been used in the range of 3Á158C to accumulate organisms for shipment to a consumer, to synchronize a specific stage for peak release or to maintain them in a quiescent Downloaded By: [Hendrichs, Jorge] At: 15:49 6 November 2009 state during shipment to prevent damage or loss. Long-term storage would benefit a programme and be cost effective in terms of maintaining the low temperatures and conducting the necessary pre-programme treatments, and where it is desired to shut down mass-rearing during the off-season (Leopold 1998). Currently, cold storage for most species that are used in biocontrol, Sterile Insect Technique (SIT), or an Integrated Pest Management (IPM) programme involves chilling at temperatures above 08C. The two main cold storage techniques that have been used in mass rearing of Trichogramma spp. are either with and/or without previous diapause induction (Greenberg et al. 1996). Immature stages of several Trichogramma species can enter diapause or become quiescent within host eggs, allowing them to tolerate long periods of subfreezing temperatures (Smith 1996). Such an approach has been attempted for a number of Trichogramma species by exposing the parasitoids to a range of cold storage conditions (Laing and Corrigan 1995; Garcı´a, Wajnberg, Pizzol, and Oliveira 2002; Pitcher et al. 2002). To date, very little data are available on the diapause requirements of T. evanescens and on storage Biocontrol Science and Technology 129

techniques for irradiated host eggs. Therefore the objectives of this study were (1) to evaluate the effects of gamma radiation and storage temperature on two types of host eggs and on their parasitization, and (2) to determine the role of these factors on the induction of diapause in T. evanescens, as a way of facilitating extended storage.

Materials and methods Host rearing Strains of E. kuehniella and S. cerealella were obtained from the Department of Plant Protection, Faculty of Agriculture of Ankara University and Adana Plant Protection Research Institute, respectively. E. kuehniella was reared on a mixture consisting of one kg wheat flour, 5% yeast and 30 g wheat germs (Marec, Kollarova, and Pavelka 1999). S. cerealella was reared on wheat grain. Throughout the rearing, cultures were kept in a rearing room at 27918Cand7095% RH and under a light regime of 14 h light followed by 10 h darkness (14 h L:10 h D). To obtain eggs for the tests, large numbers of 1Á2-day-old adults of E. kuehniella and S. cerealella were collected from stock cultures and placed in plastic jars with screen bottoms. Eggs that fell through the screen were collected the following days and sifted to remove insect parts and frass, and placed in a Petri dish. The eggs removed daily were exposed to parasitoids in glass tubes for 24 h.

Rearing of Trichogramma evanescens The T. evanescens strain used in this experiment was obtained from Adana Plant Protection Research Institute. It originated from Ostrinia nubilalis (Lep: Pyralidae) eggs collected in southern Turkey in 1999. In the laboratory, T. evanescens was mass- reared on E. kuehniella eggs for several generations. Throughout the rearing, cultures were kept in the rearing room at 24918C and 7095% RH, and under a light regime of 14 h L:10 h D. Parasitoid cultures were started from a single female on E. kuehniella eggs and maintained in glass rearing vials (27.5 cm). Downloaded By: [Hendrichs, Jorge] At: 15:49 6 November 2009 Preparation of egg cards For each experiment, a series of egg selection tests were performed with ‘egg cards’. These were prepared using certain numbers of moth eggs as described by Brower (1982). Strips of lightweight cardboard (22.5 or 2.54 cm) were glued with gum Arabica, and the gum was allowed to dry for at least 1 h. Eggs were then counted and equal numbers (150910) were sprinkled on these cards. When exposed to parasitoids during the experiments, the parasitoid to host ratio was high in these tests to ensure that most acceptable eggs would be parasitized.

Storage experiments Large numbers of 1-day-old E. kuehniella and S. cerealella eggs were placed in glass Petri dishes and irradiated in a calibrated 60Co irradiator (Therathronics 780C) with a source strength of ca. 3811 Ci and at a dose rate of ca. 1 Gy/min. The dose rate was 130 A.S. Tunc¸bilek et al.

verified with Fricke dosimetry, the best known chemical radiation dosimeter, which relies on oxidation of ferrous ions into ferric ions in an irradiated ferrous sulphate solution (Andreo, Seuntjens, and Podgorsak 2005). The eggs were exposed to gamma radiation doses of 0, 50, 100, 150 or 200 Gy to prevent development. After irradiation exposure, these eggs were placed at 48C and 7095% RH in the dark for a period of 30, 60 or 90 days. Following each period of storage at 48C, eggs were transferred to normal room temperature. Eggs glued to lightweight cardboard cards as described above were placed in tubes along with a single female T. evanescens. A single female per tube was obtained by capturing a 24-h-old female from a group of females scattered on a piece of white paper: the glass tube was placed over a medium size female and the female was allowed to walk up the vial towards the light (replicated 10 times). All females had no previous contact with host eggs, were fed with honey and allowed to mate in the tubes. The lid of the tube was covered tightly with plastic hardware cloth to prevent the wasp from leaving the tube. After 24 h, the wasps were discarded from the tubes and eggs incubated at controlled conditions. The parasitization was assessed by counting the number of black eggs after 5 days of development. Emergence was determined by counting emergence holes on black eggs. Longevity of wasps was measured every 24 h (from the time of emergence until death) by keeping the adults individually in a 15cm glass tube with food. The sex ratio was determined by examining fully developed dead adults under a microscope. Parasitization, adult emergence, and sex ratio were scored. Females that died during the experiment were excluded from evaluation.

Diapause experiments Our studies on diapause concentrated on immature stages of T. evanescens, especially pre-pupal and pupal stages. To induce diapause in immature stages of the parasitoid, less than 24-h-old eggs were presented to the wasps on the egg cards after irradiation (200 Gy). Each egg card (150910 eggs) was exposed to five mated T. evanescens females (B24-h-old) inside a glass rearing tube (18180 mm) with a drop of honey

Downloaded By: [Hendrichs, Jorge] At: 15:49 6 November 2009 solution to provide parasitoids with a carbohydrate source. After 24 h exposure to parasitoids at 2490.58C, 7095% RH and a photoperiod of 14 h L:10 h D, egg cards were assigned to each of the four different pre-storage temperatures (3, 7, 10 and 128C) for 30 days. After pre-storage, these egg cards were placed in the dark at 38C (Garc´ıa et al. 2002) for a period of 20, 70, 100 or 150 days for storage. Finally, eggs were incubated at normal room temperature in the rearing room mentioned above. Twenty-five egg cards (150 parasitized host eggs in each) were randomly assigned to each period of exposure at the respective pre-storage temperatures. Subsequently, from each period of storage at 38C, 5 egg cards (i.e., replicates) were transferred to the rearing room until adult emergence. This was carried out at steeply increasing deacclimation temperatures (7, 10, 12 and 208C) in different environmental chambers, holding the egg cards for 24 h progressively in each of these chambers (Table 1). Females that died during the experiment were excluded from the experiment. Parasitization, total adult emergence, and female emergence were recorded. Biocontrol Science and Technology 131 Table 1. Experimental design for inducing diapause in T. evanescens regarding pre-storage, storage and deacclimation temperatures and durations.

Pre-storage Storage Deacclimation Rearing temperature temperature and time temperature (8C) temperature (8C) for 30 days (8C) (days) (24 h each) (8C)

320 7 3 70 7 10 12 20 24.5 10 100 12 150

Statistical analysis Data collected from storage and diapause experiments were analyzed using a two- factor analysis of variance, with storage time and dose of radiation for the storage experiment, and pre-storage temperature and storage time for the diapause experiments as sources of variation (ANOVA) (SPSS 1999). All data were transformed to square root before statistical analysis was performed. When significant differences occurred, Tukey-HSD was used for separation of means.

Results Storage experiments We used the number of parasitized eggs as an estimation of fecundity because actual oviposition was difficult to measure. Typically, only one parasitoid emerges from each E. kuehniella or S. cerealella egg. There was no significant difference between control and stored eggs for up to 30 days for parasitization, adult emergence, and female emergence of T. evanescens (Figure 1 and Table 2) for E. kuehniella eggs. Nevertheless, there was a significant difference for adult emergence and female emergence (87 and 84%, respectively) of the parasitoid stored for 60 days (F1358.165, df3, 180, PB0.001; F159.178, df3, 180, PB0.001, respec- tively). On the other hand, data from S. cerealella eggs stored for 30 days were Downloaded By: [Hendrichs, Jorge] At: 15:49 6 November 2009 significantly lower when compared with the equivalent data obtained from stored eggs of E. kuehniella. The parasitization and emergence of the wasp from stored eggs of S. cerealella was only 63 and 54% after 30 days, drastically lower than the control (Figure 1 and Table 3) (F977.540, df2, 135, PB0.001; F606.598, df2, 135, PB0.001). No wasp emergence was recorded after 90 and 60 days of storage for E. kuehniella and S. cerealella eggs, respectively. When the complete results were evaluated, while parasitization, adult emergence and female emergence of T. evanescens were not significantly affected by irradiation doses for host eggs of E. kuehniella (Figure 1), the same parameters for S. cerealella were significantly decreased by irradiation (F3.925, df4, 135, P0.005; F3.832, df4, 135, P0.006). The difference in female emergence was not significant (Tables 2 and 3). On the other hand, parasitization, adult emergence and female emergence of T. evanescens were not influenced by gamma radiation for both host eggs, but they were influenced by storage time. Mean parasitization and adult emergence decreased 132 A.S. Tunc¸bilek et al.

(a) Parasitized egg 40 E. kuehniella S. cerealella 0 Gy 35 50 Gy 30 100 Gy 25 150 Gy 200 Gy 20 Mean 15 10 5 0 0 30 60 90 0 30 60 90 Storage (Day)

(b) Adult emergence 0 Gy 40 50 Gy E. kuehniella S. cerealella 100 Gy 35 150 Gy 30 200 Gy 25 20

Mean 15 10 5 0 0 3060900 306090 Storage (Day)

(c) Female emergence 0 Gy 50 Gy 40 100 Gy E. kuehniella S. cerealella 150 Gy 30 200 Gy

20 Mean 10

0 0 3060900 306090 Storage (Day)

Figure 1. Mean numbers of percent parasitization, adult emergence and female emergence ( SD) of T. evanescens reared on eggs of E. kuehniella and S. cerealella after storage at 48C Downloaded By: [Hendrichs, Jorge] At: 15:49 6 November 2009 9 for 0, 30, 60 or 90 days. The eggs were irradiated at 0, 50, 100, 150 or 200 Gy before storage.

significantly with the length of storage time. There was interaction between gamma radiation doses and storage time for eggs of E. kuehniella and S. cerealella (Tables 2 and 3). Parasitization, adult emergence and female emergence of T. evanescens reared on the irradiated eggs of E. kuehniella were significantly higher than on similarly treated eggs of S. cerealella.

Diapause experiments Data obtained from diapaused T. evanescens in relation to pre-storage and storage temperatures for host eggs of E. kuehniella and S. cerealella are presented in Figure 2. Results indicated that pre-storage temperatures affected the induction of diapause. The parasitization of T. evanescens was significantly higher in the eggs of E. kuehniella Biocontrol Science and Technology 133

Table 2. Two-way ANOVA results comparing the effects of each storage time and irradiation dose treatments for parasitization, adult emergence and female emergence of T. evanescens from eggs of E. kuehniella.

Mean Mean square square Treatments Parameters N df (treatment) (error) F Significance

Storage time Parasitization 200 3,180 209.51 0.11 1983.46 B0.001 Adult 200 3,180 187.42 0.14 1358.17 B0.001 emergence à emergence 200 3,180 140.46 0.88 159.18 B0.001 Radiation dose Parasitization 200 4,180 0.19 0.11 1.81 0.129 Adult 200 4,180 0.22 0.14 1.60 0.176 emergence à emergence 200 4,180 0.51 0.88 0.58 0.678 Storage time Parasitization 12,180 0.24 0.11 2.23 0.012 Radiation dose Adult 12,180 0.31 0.14 2.27 0.010 emergence à emergence 12,180 1.37 0.88 1.55 0.109

pre-stored at 10 and 128C and eggs of S. cerealella pre-stored at 128C than at 7 and 108C, respectively. These values were drastically reduced in the eggs stored at 38Cas pre-storage temperature. Our results showed that it was possible to induce diapause in developmental stages of T. evanescens by exposing the immature stages (prior to the pre-pupal stage) to 10 and 128C for 30 days. The two-way ANOVA analysis showed that parasitization, adult emergence and female emergence of T. evanescens were significantly affected by the duration of

Table 3. Two-way ANOVA results comparing the effects of each storage time and irradiation dose treatments for parasitization, adult emergence and female emergence of T. evanescens from eggs of S. cerealella.

Mean Downloaded By: [Hendrichs, Jorge] At: 15:49 6 November 2009 Mean square square Treatments Parameters N df (treatment) (error) F Significance

Storage time Parasitization 150 2,135 205.05 0.21 977.54 B0.001 Adult 150 2,135 132.71 0.22 606.60 B0.001 emergence à emergence 150 2,135 97.30 0.43 228.23 B0.001 Radiation dose Parasitization 150 4,135 0.82 0.21 3.93 0.005 Adult 150 4,135 0.84 0.22 3.83 0.006 emergence à emergence 150 4,135 0.30 0.43 0.71 0.588 Storage time Parasitization 8,135 0.64 0.21 3.05 0.003 Radiation dose Adult 8,135 0.74 0.22 3.36 0.002 emergence à emergence 8,135 1.79 0.43 4.20 B0.001 134 A.S. Tunc¸bilek et al.

Parasitized egg (a) 200 E. kuehniella S. cerealella 3ºC 180 7ºC 10ºC 160 12ºC 140 120 100

Mean 80 60 40 20 0 Pre 20 70 100 150 Pre 20 70 100 150 Storage (Day)

(b) Adult emergence 100 E. kuehniella S. cerealella 3ºC 80 7ºC 10ºC 60 12ºC

Mean 40

20

0 Pre 20 70 100 150 30 20 70 100 150 Storage (Day)

(c) Female emergence 80 E. kuehniella S. cerealella 3ºC 70 7ºC 60 10ºC 50 12ºC 40

Mean 30 20 10 0 Pre 20 70 100 150 Pre 20 70 100 150 Storage (Day)

Figure 2. Mean numbers of percent parasitization, adult emergence and female emergence Downloaded By: [Hendrichs, Jorge] At: 15:49 6 November 2009 (9SD) of T. evanescens reared on irradiated eggs of E. kuehniella and S. cerealella after 30 days of exposure to: 3, 7, 10 or 128C, followed by storage at 38C for 0, 20, 70, 100 or 150 days.

storage at 38C for both host eggs (F4.453, df4, 80, P0.003; F61.016, df4, 80, PB 0.001; F54.738, df4, 80, PB0.001 for E. kuehniella; F38.219, df4, 80, PB0.001; F31.317, df4, 80, PB0.001; F31.899, df4, 80, PB0.001 for S. cerealella) (Tables 4 and 5). Interestingly, parasitization for both host eggs was significantly higher in each period of storage compared to the control regardless of irradiaton doses, pre-storage and storage duration. Wasp emergence from E. kuehniella eggs was high and remained the same up to day 20 of storage after 30 days of pre-storage. This was higher than that of S. cerealella eggs. Thereafter, it drastically decreased in the eggs stored for 70 days and continued decreasing to 150 days when it reached its lowest level for E. kuehniella eggs. Biocontrol Science and Technology 135

Table 4. Two-way ANOVA results inducing diapause effect of each storage time and temperature treatments for parasitization, adult emergence and female emergence of T. evanescens from eggs of E. kuehniella.

Mean Mean square square Treatments Parameters N df (treatment) (error) F Significance

Storage time Parasitization 150 4,80 9.94 2.23 4.45 0.003 Adult 150 4,80 71.91 1.18 61.02 B0.001 emergence à emergence 150 4,80 60.67 1.11 54.74 B0.001 Temperature Parasitization 150 3,80 385.49 2.23 172.67 B0.001 Adult 150 3,80 69.28 1.18 58.79 B0.001 emergence à emergence 150 3,80 58.28 1.11 52.58 B0.001 Storage time Parasitization 112,80 21.51 2.23 9.64 B0.001 Temperature Adult 112,80 3.28 1.18 2.78 0.003 emergence à emergence 112,80 2.94 1.11 2.65 0.005

The two-way ANOVA showed that parasitization, adult emergence and female emergence of T. evanescens were significantly affected by the pre-storage temperature for both host eggs (F172.674, df3, 80, PB0.001; F58.790, df3, 80, PB 0.001; F52.582, df3, 80, PB0.001 for E. kuehniella; F156.292, df3, 80, PB0.001; F22.429, df3, 80, PB0.001; F18.179, df3, 80, PB0.001 for S. cerealella) (Tables 4 and 5). These values at the pre-storage temperatures of 3 and 78C decreased considerably and the parasitoids failed to enter diapause for both host

Table 5. Two-way ANOVA results inducing diapause effect of each storage time and temperature treatments for parasitization, adult emergence and female emergence of T. evanescens from eggs of S. cerealella.

Mean Downloaded By: [Hendrichs, Jorge] At: 15:49 6 November 2009 Mean square square Treatments Parameters N df (treatment) (error) F Significance

Storage time Parasitization 100 4,80 52.70 1.38 38.22 B0.001 Adult 100 4,80 9.81 0.31 31.32 B0.001 emergence à emergence 100 4,80 5.56 0.17 31.90 B0.001 Temperature Parasitization 100 3,80 215.49 1.38 156.29 B0.001 Adult 100 3,80 7.03 0.31 22.43 B0.001 emergence à emergence 100 3,80 3.17 0.17 18.18 B0.001 Storage time Parasitization 12,80 15.41 1.38 11.18 B0.001 Temperature Adult 12,80 2.25 0.31 7.17 B0.001 emergence à emergence 12,80 1.21 0.17 6.96 B0.001 136 A.S. Tunc¸bilek et al.

eggs. In contrast, the development of parasitoids that were held at 10 and 128C for 30 days was arrested in pre-pupal stage, indicating that wasps entered diapause, tolerating storage at 38C (Figure 2).

Discussion In this study, the host eggs of both E. kuehniella and S. cerealella were irradiated with gamma radiation and then stored at 48C for up to 90 days. When T. evanescens parasitoids were reared on E. kuehniella, there were no significant differences in terms of parasitization and adult emergence between control and stored eggs for up to 30 days, but there was a difference in the number of adult emergence and female emergence after 60 days. The adult and female emergence from the treatments for 30 and 60 days was still comparable to our stock culture (87 and 84%, for adult and female emergence, respectively) and quality control parameters of IOBC (2002). These values obtained from S. cerealella eggs were less than for eggs of E. kuehniella. A maximum storage of 4 weeks is reported for irradiated Ephestia eggs held at 28C and 90% RH (Bigler 1994). Parasitized S. cerealella eggs stored at 9 and 128C allowed comparatively good emergence of T. ostrinia after storage of 4 weeks (Pitcher et al. 2002). These results are similar to those of Iacob and Iacob (1972) who found that 9Á128C was an acceptable range to store T. evanescens. They also reported that storage for more than 6 weeks caused emergence to decline to levels that would not be commercially acceptable. There are also several reports of reductions in fitness traits of insects, including Trichogramma, after cold storage (Frei and Bigler 1993; Laing and Corrigan 1995). Contrary to these findings, our results showed that adult and female emergence were still acceptable after 60 days. These were much higher than the findings of Bigler (1994) and Pitcher et al. (2002). Therefore, the irradiated eggs, especially E. kuehniella eggs, can be stored more effectively for the parasitoids to be used after the low production season. The access production of the eggs of E. kuehniella and S. cerealella can be stored at 48C after irradiation at a dose of 200 Gy without any loss in number and quality of subsequent parasitoid production. Data obtained from diapaused T. evanescens indicated that pre-storage tempera- tures affected the induction of diapause. It was possible to induce diapause in immature stages of T. evanescens by exposing them within their hosts to 10 and 128C Downloaded By: [Hendrichs, Jorge] At: 15:49 6 November 2009 for 30 days. Under these conditions, the parasitoid could be stored at 38Cfor a period of 50 days, without adverse effects on emergence. The emergence of T. evanescens appeared to decrease with longer duration of storage of E. kuehniella eggs towards 150 days. While diapause induction resulted in an increase in parasitization after exposure to storage at the low temperature, emergence of the wasp decreased after 70 days of diapause induction. This may be interpereted in two ways: one likely explanation is that parasitized eggs may also have developed to the pupal stage in cold storage, but could not to reach the adult stage. The other possible explanation is that parasitized eggs cannot be identified soon after parasitization (before the pre-pupal stage), and that is why it could be not determined how many eggs were actually parasitized by the female wasp at the time of the first day of storage. Among insects, including Trichogramma, diapause generally occurs in a specific stage (Zaslavki and Uvarova 1990). Thus, as indicated by Leopold (1998), there is a need to determine the most cold tolerant stage of parasitoids before cold storage. Biocontrol Science and Technology 137

Garc´ıa et al. (2002) reported that a mimimum conditioning period (30 days) and a precise temperature regime (108C) during a particular immature stage (prior to pre- pupal stage) are necessary to induce diapause in the pre-pupae of T. cordubensis (Garc´ıa et al. 2002). Knowledge from the parasitoid emergence and fitness portions of the experiment leads us to conclude that host eggs stored at a low temperature and cold-stored T. evanescens in diapause within irradiated host eggs facilitate the mass rearing process by allowing stockpiling of parasitoids for future releases without any loss in parasitoid production. Further research on the effects of fluctuating temperatures for inducing diapause is needed to improve the storage of T. evanescens.

Acknowledgements The authors would like to thank the International Atomic Energy Agency, Vienna, Austria for the support through Research Contract No: IAEA/TUR-10782. We thank Dr. Gernot Hoch for comments on earlier drafts of the manuscript and Mrs S. O¨ ztemiz for supplying the Trichogramma evanescens used in the experiments, and the Department of Radiation Oncology for allowing the use of the Co60 irradiator.

References Andreo, P., Seuntjens, J.P., and Podgorsak, E.B. (2005), ‘Calibration of Photon and Electron Beams’,inRadiation Oncology Physics: A Handbook for Teachers and Students, ed. E.B. Podgorsak, STI/PUB/1196, IAEA, Austria, pp. 301Á354. Bigler, F. (1994), ‘Quality Control in Trichogramma Production’,inBiological Control with Egg Parasitoids, eds. E. Wajnberg and S.A. Hassan, Wallingford, UK: CAB International, pp. 33Á111. Brower, J.H. (1982), ‘Parasitization of Irradiated Eggs from Irradiated Adults of the Indian Meal Moth (Lepidoptera: Pyralidae) by Trichogramma pretiosum (Hymenoptera: Tricho- grammatidae)’, Jounal of Economic Entomology, 75, 939Á944. Brower, J.H., Smith, L., Vail, P.V., and Flinn, P.W. (1996), ‘Biological Control’,inIntegrated Management of Insects in Stored Products, eds. B. Subramanyam and D.W. Hagstrum, New York: Marcel Dekker, pp. 223Á268. Frei, G., and Bigler, F. (1993), ‘Fecundity and Host Acceptance Tests for Quality Control of Trichogramma brassica’, in Seventh Workshop of the IOBC Global Working Group, Quality Control of Mass Reared Arthropod. Rimini, Italy, September 13Á16, pp. 81Á96. Downloaded By: [Hendrichs, Jorge] At: 15:49 6 November 2009 Garc´ıa, P., Wajnberg, E., Pizzol, J., and Oliveira, L. (2002), ‘Diapause in the Egg Parasitoid Trichogramma cordubensis: Role of Temperature’, Journal of Insect Physiology, 48, 349Á355. Greenberg, S.M., Nordlund, D.A., and King, E.G. (1996), ‘Mass Production of Trichogramma spp.: Experiences in the Former Soviet Union, China, The United States and Western Europe’, Biocontrol News Information, 17, 51Á60. Hansen, L.S., and Jensen, K.M.V. (2002), ‘Effect of Temperature on Parasitism and Host- feeding of Trichogramma turkestanica (Hymenoptera: Trichogrammatidae) on Ephestia kuehniella (Lepidoptera: Pyralidae)’, Jounal of Economic Entomology, 95, 50Á56. Iacob, M., and Iacob, N. (1972), ‘Influence of Temperature Variation on the Resistance of Trichogramma evanescens (Westw.) to Storage with a View to Field Release’, Analele Institului de Cercetari pentru Protectia Plantelor, 8, 191Á199. IOBC (2002), IOBC Quality Control Guidlenes for Natural Enemies. http://users.ugent.be/ padclerc/AMRQC/images/guidelines.pdf Laing, J.E., and Corrigan, J.E. (1995), ‘Diapause Induction and Post-diapause Emergence in Trichogramma minutum Rilay (Hymenoptera: Trichogrammatidae): the Role of Host Species, Temperature and Photoperiod’, Canadian Entomologist, 127, 103Á110. Lee, R.E. (1991), ‘Principals of Insect Low Temperature Tolerance’,inInsects at Low Temperature, eds. R.E. Lee and D.L. Denlinger, New York: Chapman & Hall, pp. 17Á46. 138 A.S. Tunc¸bilek et al.

Leopold, R.A. (1998), ‘Cold Storage of Insects for Integrated Pest Management’,in Temperature Sensitivity in Insects and Application in Integrated Pest Management, eds. G.J. Hallman and D.L. Denlinger, Boulder, CO: Westview Press, pp. 235Á267. Marec, F., Kollarova, I., and Pavelka, J. (1999), ‘Radiation-induced Inherited Sterility Combined with a Genetic Sexing System in Ephestia kuehniella (Lepidoptera: Pyralidae)’, Annual Entomology Society of America, 92, 250Á259. Ministry of Agriculture and Rural Affairs (1995a), Technical Instructions for Plant Protection Vol I. Ministry of Agriculture and Rural Affairs, General Directorate of Protection and Control, Ankara, 393 pp. Ministry of Agriculture and Rural Affairs (1995b), Technical Instructions for Plant Protection Vol IV. Ministry of Agriculture and Rural Affairs, General Directorate of Protection and Control, Ankara, 393 pp. Pitcher, S.A., Hoffmann, M.P., Gardner, J., Wright, M.G., and Kuhar, T.P. (2002), ‘Cold Storage of Trichogramma ostrinia Reared on Sitotroga cerealella Eggs’, Biocontrol, 47, 525Á535. Ravensberg, W.J. (1992), ‘Production and Utilization of Natural Enemies in Western European Glasshouse Crops’,inAdvances in Insect Rearing for Research and Pest Management, eds. T.E. Anderson and N.C. Leppla, Boulder, CO: Westview Press, pp. 465Á487. Scho¨ller, M., Prozell, S., Al-Kirshi, A.G., and Reichmuth, C. (1997), ‘Towards Biological Control as a Major Component of Integrated Pest Management in Stored Product Protection’, Journal of Stored Product Research, 33, 81Á97. SPSS (1999), SPSS Version 10.0, Chicago, IL: USA. Tunc¸bilek, A.S., Canpolat, U., and Ayvaz, A. (2009), ‘Effects of Gamma Radiation on Suitability of Stored Cereal Pest Eggs and the Reproductive Capability of the Egg Parasitoid Trichogramma evanescens (Trichogrammatidae: Hymenoptera)’, Biocontrol Science and Technology, this volume. Zaslavki, V.A., and Uvarova, T.Y. (1990), ‘Environmental and Endogenous Control of Diapause in Trichogramma Species’, Entomophaga, 35, 23Á29. Downloaded By: [Hendrichs, Jorge] At: 15:49 6 November 2009 Biocontrol Science and Technology, Vol. 19, S1, 2009, 139Á155

RESEARCH ARTICLE Interaction of entomopathogenic nematodes, Steinernema glaseri (Rhabditida: Steinernematidae), cultured in irradiated hosts, with ‘F1 sterility’: Towards management of a tropical pest, Spodoptera litura (Fabr.) (Lepidoptera: Noctuidae) Rakesh K. Seth*, Tapan K. Barik, and Sonal Chauhan

Department of Zoology, University of Delhi, Delhi Á 110 007, India

Efficacy of the entomopathogenic nematode (EPN), Steinernema glaseri, (Steiner) cultured in radio-sterilized host, was studied vis-a`-vis radiation-induced F1 sterility on a tropical lepidopteran pest, Spodoptera litura (Fabr.). To ensure safe transport of S. glaseri EPNs in vivo, host radio-sterilization was done; and the parasitising performance of S. glaseri infective juveniles (IJs), cultured in irradiated last instar S. litura larvae (with either 40 or 70 Gy of gamma rays) was evaluated on F1 sterile insects (progeny of male moths treated with 100 Gy, 130 Gy). S. glaseri EPNs cultured in radio-sterilized larvae at 40 Gy, had better infective potential than those cultured in sterilized host larvae at 70 Gy. F1 sterile larval hosts (progeny from 100 or 130 Gy treated parents) were equally acceptable to the EPNs cultured in radio- sterilized hosts, although the nematode harvest was reduced on F1 sterile hosts at 130 Gy.Infectivity of IJs derived from F1 sterile host was almost similar on F1 sterile larvae and normal larvae of S. litura, although their parasitisation efficacy on the F1 sterile hosts was relatively less than the controls. The IJs performance was little influenced by irradiation of IJs’ parent host and current host’s irradiation history individually, but both the parameters together did not have any further negative interaction on the performance of IJs. The present results indicate that S. glaseri harvested from F1 sterile larval hosts (progeny from 100 or 130 Gy treated parents) retained a reasonably high degree of infectivity on normal and F1 sterile S. litura hosts (61Á83% of controls). Two highly promising operational modes of integrating S. glaseri EPNs with ‘F1 sterility’ to suppress S. litura populations (initial releases of EPNs to strongly suppress the density of the pest followed by use Downloaded By: [Hendrichs, Jorge] At: 15:50 6 November 2009 of F1 sterility vs. simultaneous use of both techniques) are discussed. Keywords: Spodoptera litura; cutworm; gamma irradiation; entomopathogenic nematodes; Steinernema glaseri; parasitoid; population suppression; inherited sterility; F1 sterility

Introduction Spodoptera litura (Fabricius) (Lepidoptera: Noctuidae) is a common defoliator of a polyphagous nature (recorded on more than 120 host plants), and it has attained an economically serious status in the Indian subcontinent (Lefroy 1908; Moussa, Zaher, and Kotby 1960; Chari and Patel 1983; Higuchi, Yamamoto, and Suzuki 1994). Increasing environmental hazards from the use of chemical pesticides and develop- ment of insecticide-resistance in this pest (Ramakrishnan, Saxena, and Dhingra

*Corresponding author. Email: [email protected]

First Published Online 21 April 2009 ISSN 0958-3157 print/ISSN 1360-0478 online # 2009 Taylor & Francis DOI: 10.1080/09583150902814515 http://www.informaworld.com 140 R.K. Seth et al.

1984; Armes, Wightman, Jadhav, and Ranga Rao 1997) have encouraged entomol- ogists to seek environmentally sound alternative measures to control this pest. One such ecologically compatible pest control strategy is biological control. Increased efforts in recent years have been focused on biological control using entomopatho- genic nematodes in the families, Heterorhabditidae and Steinernematidae. Entomo- pathogenic nematodes (EPNs) of these two families kill their hosts due to the action of their endo-symbiotic bacteria; and they have a broad host range, can be mass- produced using conventional fermentation technology, and are exempted from registration requirements in several countries (Kaya and Gaugler 1993). Therefore, they are commercially available for insect control in nurseries, greenhouses, and turfgrass around the world (Grewal and Georgis 1998). The EPNs mainly attack soil insects in the families Chrysomelidae and Curculionidae. They are also reported to parasitize some insects in the order Lepidoptera (Poinar 1979). Successful applica- tion of Steinernema carpocapsae (Rhabditida: Steinernematidae) against the beet armyworm, Spodoptera exigua was attained in a commercial nursery in Florida (Kaya and Hara 1980; Begley 1990). The residual effect of the nematode treatment lasted longer than that of standard chemical pesticides (Bari and Kaya 1984). The biocontrol potential of S. carpocapsae against the cutworm, S. litura has been suggested (Narayanan and Gopalakrishnan 1987; Choo, Kaya, and Reed 1989; Sezhian, Sivakumar, and Venugopal 1996). The feasibility of using induced ‘F1 sterility’ as a genetic control method has been studied for several species of Lepidoptera (Knipling 1970). Since high doses of gamma irradiation (200Á350 Gy) are required to induce complete sterility in Lepidoptera (North and Holt 1971) and exposure to high radiation levels also adversely affects male mating behaviour and competence, one approach to reduce the negative effects of radio-resistance in Lepidoptera has been the use of inherited or F1 sterility (North 1975; Carpenter, Bloem, and Marec 2005). In view that F1 sterile progeny are produced in the field, the release of partially sterile insects offers greater suppressive potential than the release of fully sterile insects (LaChance 1985) and is more compatible with other pest control mechanisms or strategies (Carpenter 1993). Knipling (1970) explored the theoretical basis for the application of F1 sterility for control of lepidopteran pests by using mathematical models, which was further Downloaded By: [Hendrichs, Jorge] At: 15:50 6 November 2009 supported by later studies (Carpenter 1993; Anisimov 1998). Many studies have shown that F1 sterility can be effectively combined with other biological control tactics such as pheromone disruption (Bloem, Bloem, Carpenter, and Calkins 2001), entomopathogens (Hamm and Carpenter 1997), host plant resistance (Carpenter and Wiseman 1992a,b) and natural enemies (Carpenter, Hidrayani, and Sheehan 1996; Greany and Carpenter 1999). As a result of these studies, F1 sterility is regarded as the best genetic method for combined use with other suppressive measures against Lepidoptera (Carpenter and Bartlett 1999). F1 sterility, as a parabiological (genetic) control measure, has been proposed for the suppression of S. litura (Seth and Sehgal 1993; Seth and Sharma 2001), instead of using the Sterile Insect Technique (SIT) wherein the high (100%) sterilizing dose impairs the mating competitiveness of the moths. A range of 100Á130 Gy was suggested to be employed in order to implement F1 sterility for the suppression of S. litura populations, because under the influence of this range of doses of ionising Biocontrol Science and Technology 141

radiation, the mating competitiveness of S. liturawas not drastically impaired unlike in case of the higher and fully sterilizing radiation doses. Therefore, two doses, viz. 100 and 130 Gy were selected in the present study to ascertain the interaction of EPNs with host insects derived from irradiated male parents. Further, it has been suggested that the EPNs carried within the host (i.e., in vivo) retain better viability than in aqueous releases. For instance, the dispersal abilities of EPNs such as Heterorhabditis bacteriophora Poinar (HP88 strain) and S. carpocap- sae, were reported to be significantly greater when nematodes were applied in cadavers than when they were applied in aqueous suspension. This difference in migration ability was presumed to be due to physiological or behavioural differences between nematodes exiting hosts and those kept in aqueous suspension; and there could also be a difference in fitness and behaviour of nematodes carried in vivo as compared to nematodes used in aqueous application (Shapiro and Glazer 1996; Shapiro and Lewis 1999). Since entomopathogenic nematodes applied in infected hosts (in vivo) may have dispersal advantages and increased efficacy in biological control, it was desirable to understand the efficacy of EPNs carried within their host in relation to other compatible control measures. Steinernema glaseri (Steiner) was selected as model entomopathogenic species in the present study due to its tropical origin and persistent viability in tropical conditions (Grewal, Selvan, and Gaugler 1994) and relatively large size (Stuart, Lewis, and Gaugler 1996). Further, S. glaseri has been reported to parasitize S. litura (Kondo and Ishibashi 1986; Kaya and Koppenhofer 1996), and S. litura has been reported as one of the preferred hosts for S. glaseri in terms of sensory response elicited towards insect emitted attractants (Bilgrami, Kondo, and Yoshiga 2000). A fraction of the individuals of the pest species that are used with the intention that they carry EPNs may escape parasitisation and add to pest population in the ecosystem. Therefore, the radio-sterilization of the insect-hosts was considered necessary to have a risk-free mode of carrying EPNs in vivo, so as to ensure safe transport of EPNs. The present study for assessing the parasitising performance of EPNs transported in vivo was designed to ensure the transport of viable EPNs within irradiated host, and to study the interaction of S. glaseri EPNs, carried within radio-

Downloaded By: [Hendrichs, Jorge] At: 15:50 6 November 2009 sterilized S. litura hosts with F1 sterility (substerilizing radiation induced genetic control method). An attempt was made to evaluate (i) the parasitising performance of S. glaseri EPNs against normal and F1 sterile S. litura hosts, (ii) the infective potential of S. glaseri EPNs cultured in radio-sterilized hosts against normal and F1 sterile S. litura hosts, and (iii) the parasitising performance of S. glaseri EPNs harvested from F1 sterile S. litura hosts.

Materials and methods Maintenance of insect hosts Maintenance of S. litura as a potential host of entomopathogenic nematodes Mass rearing of S. litura was conducted on semi-synthetic diet (Seth and Sharma 2001) at 2791oC, 7595% relative humidity and 12 h L:12 h D regimen in an insectary for experimental investigations. 142 R.K. Seth et al.

Maintenance of Galleria mellonella as a factitious host of S. glaseri EPNs The greater wax moth, Galleria mellonella (L.) was reared on a semi-synthetic diet consisting of 350 mL honey, 350 mL glycerin, 400 g corn starch, 200 g wheat flour, 200 g wheat bran, 200 g milk powder, and 50 g yeast granules. A mixture of honey and glycerin was prepared. Then the dry constituents were mixed and added to honey-glycerin mix. This amount of diet could accommodate about 100 larvae. The culture of this moth was maintained at 3091oC (as described by Woodring and Kaya 1988).

Maintenance of S. glaseri entomopathogenic nematodes A core culture of an entomogenous nematode species, S. glaseri, procured from Biosys, USA, was maintained on the factitious insect-host, the greater wax moth, G. mellonella, larvae at optimum environmental conditions, viz.,2590.5oC, 7595% relative humidity, by the method of Woodring and Kaya (1988). G. mellonella was chosen to serve as host for the stock culture of nematodes as it was found to be quite susceptible to infection and to be an acceptable host. S. glaseri dauers (infective juveniles) stored in 0.1% formalin solution in sterilized distilled water at 6Á88C maintained proper viability for 2Á4 weeks. These EPNs were allowed to acclimatize at ambient temperature (25oC) for 24 h before being applied to host larvae for ongoing in vivo rearing or for experimental studies.

Irradiation of insects In the irradiation facility of the Institute of Nuclear Medicine and Allied Sciences (INMAS), Ministry of Defence, Delhi, S. litura was irradiated with a Cobalt-60 source at the dose rate of 80Á95 Gy/min for the present investigations. A dose range of 40Á70 Gy was used for radio-sterilization of last instar larvae of S. litura.A sublethal dose range of 100Á130 Gy was used to irradiate male moths (0Á1-day-old) to produce partially sterilized males that could be released to mate with normal females and produce F1 sterile progeny. These F1 sterile progeny derived from sub- sterile male parents were used for the present experimental study. Downloaded By: [Hendrichs, Jorge] At: 15:50 6 November 2009

Bioassay of S. glaseri EPNs A bioassay requiring detailed observations was designed to assess the various features of the nematode’s infective behaviour and reproduction.

Inoculation Individual 1Á2-day-old sixth instar S. litura larvae (L6) were each placed into a Petri dish (5015 mm) lined with 2-fold filter paper (Whatman #1). Freshly harvested (up to 2 weeks old) infective juveniles (IJs) in all experimental regimens were transferred into each Petri dish by distributing them evenly onto the filter paper base in order to inoculate at the rate of 25 IJs per individual host. Incubation was performed at 25oC and 7595% relative humidity. Each assay for ascertaining IJs infective behaviour was conducted in an individual mode with respect to host (comprising of single L6 Biocontrol Science and Technology 143

larva inoculated by 25 IJs per assay) due to the cannibalistic behaviour of host larvae.

Parasitisation performance and harvesting of nematodes The time profile for the induction of morbidity and mortality in the exposed S. litura larvae was recorded at 4Á6 h intervals. Morbidity is an initial behavioural response to haemolymph septicemia caused by toxins released by symbiotic bacteria of EPNs, followed by the host’s resultant mortality. Morbidity criteria of infected larva included their minimal response to a probe, sluggish nature, and delayed resumption of the normal posture with slight torsion in the body when turned upside down. After host death, the parasitised (host) larvae were incubated until the next generation of IJs were developed. After 7Á8 days, IJs were seen wriggling at the outer surface of the insect cadaver. The incubation times of EPNs (i.e., from inoculation to emergence of next generation IJs from host cadaver) were recorded. The harvest of IJs was done using the White Trap method (White 1927). For this, the cadavers having proliferating IJs inside were removed from the inoculative Petri dishes and transferred to sterilized harvest dishes (90 mm diameter). The daily profiles of emergence of IJs out of host cadavers and their total emergence (harvest) period were recorded. Their harvest potential (IJs’ yield) was determined in terms of cumulative number of IJs harvested over the total harvest period per host, and the number of IJs harvested per mg fresh weight of host. IJs when dead appeared to have lost weight, because they tended to float as compared to surviving nematodes. Dead IJs remained completely straight. The viability criterion in the S. glaseri EPN population was taken as 80% or more survival with responsiveness, as suggested by Epsky and Capinera (1994). For assessment of host morbidity timing and its mortality timing by EPNs, incubation time of EPNs (within host) and harvest potential of IJs, individual host observations were conducted and each replicate represented the mean value of observations on a set of five to seven individual hosts; whereas for the assessment of percent parasitisation, a cohort of 25 host larvae (evaluated by individual exposures) constituted each replicate. Downloaded By: [Hendrichs, Jorge] At: 15:50 6 November 2009

Interaction of S. glaseri EPNs with radiation induced F1 sterility in S. litura

Bioefficacy of EPNs on F1 sterile insect hosts

Firstly, the bioefficacy of S. glaseri EPNs was assayed as indicated above against F1 sterile S. litura larvae derived from matings in which their male parent had been treated with sub-sterilizing gamma doses (100 or 130 Gy). Bio-efficacy of S. glaseri was assessed by recording the times needed to induce morbidity and mortality, the incubation time, the percent parasitisation, development and proliferation (as computed by the harvest potential) of EPNs.

Bioefficacy of S. glaseri IJs cultured in radio-sterilized insect hosts The bioefficacy of S. glaseri IJs cultured in radio-sterilized insect hosts was evaluated in terms of times taken to morbidity and mortality of host, parasitising efficacy, 144 R.K. Seth et al.

incubation period and IJs’ harvest potential against normal and against F1 sterile S. litura hosts. Two doses, viz., 40 and 70 Gy, were selected for radio-sterilization of S. litura sixth instar larvae in view of viability and EPNs’ parasitising performance on radio-sterilized hosts (Seth and Barik 2007).

Bioefficacy of S. glaseri IJs cultured in F1 sterile insect larvae Lastly, the parasitising performance and harvest potential of IJs (indicating reproductive potential) of the entomogenous nematode cultured in parasitised F1 sterile insects (progeny of irradiated male parents), was ascertained against normal and against F1 sterile S. litura hosts.

Data analysis All the characteristics related to host killing efficiency and parasitization perfor- mance of S. glaseri were studied in 10Á12 replicates. The data were computed for means, standard error and further analysis of variance (ANOVA, SPSS, 11.0). Two- way ANOVAwas used to test the interaction of irradiation history of IJs’ parent host (either as radio-sterilized host or F1 sterile larval progeny of irradiated male parent) and current host’s irradiation background on the performance of IJs. Percentage data were transformed using arcsine âx before ANOVA. Means were separated at the 5% significance level by least significant difference (LSD) test (Snedecor and Cochran 1989).

Results

Bioefficacy of EPNs on F1 sterile insects The bioefficacy of normal EPNs (i.e., cultured in un-irradiated insect host) was ascertained against F1 sterile S. litura larvae produced in matings of untreated females with substerilized male moths, irradiated with 100 and 130 Gy, in order to understand the degree of acceptability and suitability of F1 progeny as potential hosts to entomophilous nematodes (Table 1, Figure 1). ANOVA performed on data Downloaded By: [Hendrichs, Jorge] At: 15:50 6 November 2009 pertaining to bioefficacy of EPNs on F1 sterile insects indicated that there was a significant influence of induced sterility of the host (F1 sterile insects) on the parasitisation efficacy of EPNs, but not on the mortality induction process. The onset of morbidity and mortality induced by normal IJs (i.e., IJs derived from un- treated host) was not different in F1 sterile hosts and un-irradiated controls (i.e., untreated host) (P0.05). The incubation time taken by IJs on F1 sterile hosts was slightly prolonged, being significantly longer in F1 hosts derived from 130 Gy treated male parent (PB0.05). Parasitisation response was reduced in a dose dependent manner on F1 sterile hosts, with a significant impact on F1 larvae derived from fathers that had been treated with 130 Gy (PB0.05). The IJs’ harvest was found to be moderately reduced on F1 hosts, but the effect was significant on F1 hosts at 130 Gy (PB0.05). The number of IJs harvested from F1 sterile host was reduced by about 8Á11% at 100 Gy and 17Á20% at 130 Gy with respect to controls. The harvest period was also slightly affected by the radiation dose applied to the host in the previous generation (PB0.05). Biocontrol Science and Technology 145

Table 1. Infective performance of entomopathogenic nematodes (EPNs), Steinernema glaseri, on F1 sterile Spodoptera litura larvae (progeny of matings of untreated females and males irradiated with sub-sterilizing gamma doses of 100 or 130 Gy).

Time Time Harvest (yield) of IJs required required for for IJs per Nature of Nature of morbidity mortality Incubation IJs per mg body Period EPN host (h) (h) time (h) host wt (days)

Normal IJs F1 host from 22.8a 48.6a 177.4a 20855a 34.9a 11.9a (Control) unirradiated 91.1 91.8 94.9 91022 91.2 90.4 male parent (Control) Normal IJs F1 host from 23.8a 50.9a 182.9ab 18426ab 31.9ab 11.2a 100 Gy 91.2 92.2 95.2 9894 91.1 90.4 irradiated male parent Normal IJs F1 host from 23.1a 49.3a 192.8b 16667b 28.9b 9.6b 130 Gy 90.9 91.5 94.1 91006 91.7 90.5 irradiated male parent

Sixth instar host larvae (1Á2-day-old) were bioassayed, IJs, infective juveniles of EPNs, Means9SE followed by same letter in a column are not significantly different at P50.05 level (ANOVA followed by LSD post-test); n12.

Bioefficacy of EPNs cultured in radio-sterilized host larvae The infective performance of EPNs cultured in radio-sterilized host was studied against normal and F1 sterile host-insects to ensure the viability and virulence of EPNs transported in radio-sterilized host (Table 2, Figure 2). There was no significant interaction between irradiation of the IJs’parent-host (in which the IJs had been produced) and the nature of the current host, i.e., F1 sterile larvae (i.e., having radiation dose applied to its male parent) on parasitisation performance of EPNs (P0.05), except in case of the time to mortality (F 8.9;df4, 99; PB0.01) and the harvest period (F5.04; df4, 99; PB0.01) as indicated by two-way Downloaded By: [Hendrichs, Jorge] At: 15:50 6 November 2009 ANOVA. Further, as per one-way analysis of variance, both the factors, viz., irradiation of IJs’ parent host and nature (quality) of current F1 sterile host (i.e., the degree of sterility induced by irradiated male parent), individually had a significant, though not drastic, impact on time to mortality, percent parasitisation, incubation period and the number of IJs harvested from host cadaver (PB0.05). Infection was induced a little later by IJs that had been cultured in irradiated hosts than cultured in the untreated control, i.e., IJs emerging from normal hosts. Time taken for inducing morbidity was slightly more in case of IJs harvested from 70 Gy irradiated hosts than by IJs from 40 Gy irradiated hosts. The time to mortality induced by IJs that had been cultured in radio-sterilized host was slightly delayed on both normal and F1 sterile hosts, as compared to that induced by control IJs (PB0.05), although the IJs from 40 Gy irradiated hosts killed host faster than the IJs from 70 Gy irradiated hosts. The incubation time taken by IJs from radio-sterilized hosts was prolonged on normal as well as F1 hosts. The effect was a little more apparent when the IJs that 146 R.K. Seth et al.

Figure 1. Parasitisation efficacy of entomopathogenic nematodes (EPNs), Steinernema glaseri, against normal (N) and F1 sterile Spodoptera litura larvae. The latter were the progeny of matings of untreated S. litura females with males irradiated with sub-sterilizing doses (100 or 130 Gy) of gamma radiation. Sixth instar host larvae (1Á2-day-old L6) were bioassayed to ascertain the parasitisation by EPNs (25 IJs/host larva). Means9SE (of the bars) denoted by the same letter are not significantly different at PB0.05 level (calculated using ANOVA followed by LSD post-test); percentage data were arcsine transformed before ANOVA, but data in the figure are back transformations; n12 (a cohort of 25 host larvae, evaluated by individual exposures, constituted each replicate).

had been cultured in 70 Gy irradiated hosts pursued parasitisation, and also it was more apparent when parasitisation occurred towards F1 hosts at 130 Gy (PB0.05). Bioassays showed that the parasitisation efficacy (Figure 2) of IJs derived from radio-sterilized hosts was impaired (PB0.05) by 14Á16% at 40 Gy and by 18Á21% at 70 Gy with respect to controls. F1 sterile hosts from 100 Gy or 130 Gy treated male parents appeared to be almost equally acceptable to IJs from irradiated hosts. The EPNs’ harvest was moderately but significantly reduced when IJs that had been cultured in larvae irradiated with 40 Gy infected normal hosts (PB0.05). The Downloaded By: [Hendrichs, Jorge] At: 15:50 6 November 2009 harvest of EPNs was further reduced in host larvae irradiated with 70 Gy (PB0.05). Moreover, EPNs’ harvest was further negatively affected when IJs that had been cultured in irradiated hosts infected F1 sterile hosts. This resulted in the evident reduction of the harvest in the case of F1 sterile hosts at 130 Gy. Similarly the harvest period was shortened, the effect being more apparent on F1 hosts at 130 Gy (PB 0.05). On the other hand, the harvest period was not significantly shortened when IJs from 40 Gy irradiated hosts parasitised normal or F1 host insects (P0.05).

Bioefficacy of EPNs cultured in F1 sterile host larvae

The infective performance of IJs cultured in parasitised F1 sterile S. litura larvae was ascertained on normal (unirradiated) hosts and on F1 sterile host-progeny of irradiated male parents, so as to understand the persistence of infective viability of IJs harvested from F1 sterile hosts (Table 3, Figure 3). It is worth noting that no evident interaction was noticed between the irradiation background of IJ’s parent Biocontrol Science and Technology 147

Table 2. Infective performance of entomopathogenic nematodes (EPNs, Steinernema glaseri, cultured in radio-sterilized Spodoptera litura larvae (40 Gy, 70 Gy) and applied to normal and F1 sterile S. litura larvae (progeny of matings of untreated females and males irradiated with sub-sterilizing gamma doses of 100 or 130 Gy).

Time Time Harvest (yield) of IJs required required for for IJs per Nature of Nature of morbidity mortality Incubation IJs per mg body Period EPN host (h) (h) time (h) host wt (days) Normal IJs Normal host 22.8a 47.5a 179.3a 21010a 35.1a 11.8a (Control) (Control) 91.2 91.7 94.2 9911 92.2 90.5 IJs from 40 Normal host 23.9ab 68.7c 199.2bc 16986bc 30.2abc 10.1bc Gy treated 91.6 93.4 97.4 9889 91.5 90.5 host IJs from 40 F1 host from 24.2ab 54.7b 188.9abc 17459b 31.1ab 11.1ab Gy treated 100 Gy 91.2 91.1 97.1 9981 91.6 90.5 host irradiated male parent IJs from 40 F1 host from 24.4ab 54.3ab 195.9bc 15270bc 28.7bc 10.1bc Gy treated 130 Gy 91.1 92.6 95.9 9839 91.4 90.4 host irradiated male parent IJs from 70 Normal host 26.7b 69.3c 214.1c 15940bc 27.0bc 9.1c Gy treated 91.2 92.6 97.8 9797 91.3 90.4 host IJs from 70 F1 host from 24.4ab 55.7b 194.5abc 16222bc 29.1bc 10.1bc Gy treated 100 Gy 90.9 93.6 96.2 9785 91.5 90.5 host irradiated male parent IJs from 70 F1 host from 26.2b 58.1b 201.5bc 14413c 26.4c 9.7bc Gy treated 130 Gy 90.8 92.9 99.1 9653 91.3 90.5 host irradiated male parent

Downloaded By: [Hendrichs, Jorge] At: 15:50 6 November 2009 Note: Sixth instar host larvae (1Á2-day-old) were bioassayed, IJs, infective juveniles of EPNs, Means9SE followed by same letter in a column are not significantly different at P50.05 level (two-way ANOVA followed by LSD post-test); n12.

host, i.e., F1 sterile larvae (in which the IJs used had been cultured) and the nature (quality) of the current host (i.e., F1 sterile host) on parasitisation behaviour of EPNs (P0.05) except in case of time to mortality (F 2.94; df4, 81; PB0.05), as reflected by two-way ANOVA. However, one-way analysis of variance reflected an evident impact (but not drastic) of these factors individually, on time to mortality, percent parasitisation, incubation time and the numbers of nematodes harvested from host larva (PB0.05). Induction of morbidity and mortality in normal host larvae caused by IJs cultured in F1 sterile host larvae (progeny of matings of untreated females with males irradiated with 100/130 Gy) occurred at a similar time as in the controls (P0.05), although their infective response was slightly delayed on F1 sterile host larvae (PB0.05). Incubation time taken by IJs that had been cultured 148 R.K. Seth et al.

Figure 2. Parasitisation efficacy of entomopathogenic nematodes (EPNs), Steinernema glaseri, cultured in radio-sterilized Spodoptera litura larvae (40 or 70 Gy) and applied to normal (N) and F1 sterile S. litura larvae (progeny of matings of untreated females and males irradiated with sub-sterilizing doses (100 or 130 Gy) of gamma radiation). The parasitisation bioassay was conducted by applying 25 infective nematode juveniles to each sixth instar S. litura larvae (1Á2-day-old L6). Means9SE (of the bars) denoted by the same letter are not significantly different at PB0.05 level (calculated using ANOVA followed by LSD post-test); percentage data were arcsine transformed before ANOVA, but data in figure are back transformations; n12 (a cohort of 25 host larvae, evaluated by individual exposures, constituted each replicate).

in F1 sterile hosts was not significantly affected on normal host larvae with respect to the control, whereas it was slightly prolonged on F1 sterile host larvae. The effect was especially apparent when IJs that had been cultured in F1 sterile host larvae infected Downloaded By: [Hendrichs, Jorge] At: 15:50 6 November 2009 F1 sterile host larvae at 130 Gy (PB0.05). Parasitisation efficacy (Figure 3) was reduced by 18Á38% when IJs, which had been cultured in F1 sterile host larvae, were allowed to infect normal and F1 sterile insects, as compared to the controls (90.5% efficacy). IJs emerged out of F1 sterile host did not show any differential infective response towards F1 sterile host as compared to normal host larvae (P0.05), except the significantly reduced parasitisation response by IJs from F1 sterile hosts (progeny of 130 Gy treated male) towards F1 sterile host larvae at 130 Gy (55.4%; PB0.05). The IJs’ harvest was reduced by 16Á25% when IJs that had been cultured in F1 sterile hosts (progeny of 100 Gy treated male) parasitised F1 sterile hosts, with respect to controls. Further, a harvest reduction of 18Á31% was recorded in case of infection by IJs that had been cultured in F1 sterile hosts at 130 Gy. The reduction in harvest potential of IJs depended upon the gamma dose administered to male parent of F1 insects (as hosts). The IJs’ harvest potential exercised by the infection of IJs that had been cultured in F1 sterile hosts was reasonably high on normal insect-hosts and nearly similar to Biocontrol Science and Technology 149

Table 3. Infective performance of entomopathogenic nematodes (EPNs), Steinernema glaseri, cultured in F1 sterile Spodoptera litura larvae and applied to normal and F1 sterile S.litura larvae (progeny of matings of untreated females and males irradiated with sub-sterilizing gamma doses of 100 or 130 Gy).

Time Time Harvest (yield) of IJs required required for for IJs per Nature of Nature of morbidity mortality Incubation IJs per mg Period EPN host (h) (h) time (h) host body wt (days) Normal IJs Normal host 22.9a 47.5a 180.1a 21202a 36.7a 11.18a (Control) (Control) 91.2 91.6 95.9 91055 91.8 90.6

IJs from F1 Normal host 23.6ab 46.7a 185.2a 19139b 33.1ab 10.6ab host (derived 90.9 92.2 94.6 9845 91.2 90.5 from 100 Gy treated male) IJs from F1 F1 host from 25.8bc 52.9ab 188.5a 17724b 30.7bcd 9.2bc host (derived 100 Gy 91.1 92.3 96.9 9886 91.2 90.4 from 100 Gy irradiated treated male) male parent IJs from F1 F1 host from 24.3abc 54.8bc 217.7b 17774bc 27.6cd 9.1bc host (derived 130 Gy 91.9 92.7 910.4 9890 91.3 90.4 from 100 Gy irradiated treated male) male parent IJs from F1 Normal host 25.2abc 52.4ab 191.3a 18197b 31.3b 10.1ab host (derived 91.2 91.6 96.9 9711 91.1 90.3 from 130 Gy treated male) IJs from F1 F1 host from 25.9c 62.2cd 215.9b 17343b 25.4d 9.7ab host (derived 100 Gy 91.3 93.1 99.5 9893 91.2 90.4 from 130 Gy irradiated treated male) male parent IJs from F1 F1 host from 27.4c 65.4d 226.2b 15412c 26.6d 8.9c host (derived 130 Gy 91.5 92.5 910.1 9691 91.3 90.2 from 130 Gy irradiated Downloaded By: [Hendrichs, Jorge] At: 15:50 6 November 2009 treated male) male parent

Note: Sixth instar host larvae (1Á2-day-old) were bioassayed, IJs, infective juveniles of EPNs, Means9SE followed by same letter in a column are not significantly different at P50.05 level (two-way ANOVA followed by LSD post-test); n10.

control. The harvest period of IJs was also found to be slightly affected, the effect being especially apparent on F1 hosts derived from 130 Gy treated male (PB0.05).

Discussion The bio-infective performance of normal IJs, in terms of inducing morbidity and mortality, was similar towards F1 sterile host (derived from irradiated male parent) and control S. litura larvae, whereas the parasitisation and the harvest of EPNs were diminished on F1 hosts at 130 Gy. 150 R.K. Seth et al.

Figure 3. Parasitisation efficacy of entomopathogenic nematodes (EPNs), Steinernema glaseri, cultured in F1 sterile Spodoptera litura larvae and applied to normal (N) and F1 sterile S. litura larvae (progeny of matings of untreated females and males irradiated with sub- sterilizing doses (100 or 130 Gy) of gamma radiation). The parasitisation bioassay was conducted by applying 25 infective nematode juveniles to each sixth instar S. litura larvae (1Á 2-day-old L6). Means9SE (of the bars) denoted by the same letter are not significantly different at PB0.05 level (calculated using ANOVA followed by LSD post-test); percentage data were arcsine transformed before ANOVA, but data in figure are back transformations; n10 (a cohort of 25 host larvae, evaluated by individual exposures, constituted each replicate).

The parasitising performance of IJs cultured in 40 Gy treated hosts was better and their viability was greater than that of IJs cultured in 70 Gy treated host larvae. Although 70 Gy was ascertained to be complete (reliable) sterilizing dose for L6, the Downloaded By: [Hendrichs, Jorge] At: 15:50 6 November 2009 overall sterilizing impact 40 Gy (that induced 80Á91% reproductive suppression in L6 in different irradiated crosses) was apparently much more (almost complete) if coupled with 28Á47% reduction in mating success (with respect to control) plus reduced adult emergence (53.9%) with pronounced degree of malformation (61.2%) at this dose. Hence, even 40 Gy dose could also be considered as safe for radio- sterilization of L6, in order to ensure an almost risk free mode of transport of EPNs (by avoiding potential pest release) (Seth and Barik 2007, Seth unpublished). The bio-infective performance of IJs cultured in radio-sterilized hosts exhibited a reasonable degree of parasitisation potency, with better performance from IJs derived 40 Gy irradiated host than that from 70 Gy irradiated hosts towards F1 sterile host progeny of irradiated male parents. Time profile of onset of morbidity caused by these IJs (from irradiated host) was more or less similar in F1 sterile insects and in normal hosts. That indicated that release of the toxins by endosymbiotic bacteria must have occurred in a similar fashion in F1 insects as in the controls. Further, it was interesting to note that F1 hosts probably possessed less immunity or Biocontrol Science and Technology 151

offered less resistance towards the infecting IJs cultured in irradiated hosts; hence mortality induced by these IJs occurred more quickly in F1 sterile insects than in normal hosts. The F1 progeny from the 100 Gy treatment were found to be more acceptable and suitable hosts (than F1 larvae from 130 Gy treated male parent) for S. glaseri EPNs transported in the radio-sterilized host. The bio-infective performance of IJs harvested from F1 sterile host towards normal and F1 sterile insects, however, indicated a great degree of persistent efficacy of these EPNs (emerging out of parasitised F1 sterile host) that would interact with the insect population in the ecosystem. The relatively reduced (though not drastic) performance of IJs, which had been cultured in F1 sterile insects (especially at 130 Gy) might be attributed to probably lower host quality in terms of nutrition and secondary chemicals, which could be correlated with affected nutritional efficiencies and energy budget of S. litura as reported by Seth and Sehgal (1993). Further, it was interesting to note that the IJs cultured in the F1 sterile host (derived from treated males at 100 Gy, 130 Gy) could also retain their infective potential up to 61.2Á82.5%, with 70Á91% harvest potential of IJs (with respect to controls). This reflects that these IJs, if released in inoculative mode, in conjunction with F1 sterility, could be used quite effectively in the field for several generations keeping in view the acceptability and suitability of F1 sterile insects as potential host coupled with tremendous recycling capacity of these EPNs within F1 insects. Inherited sterility induced by irradiated male parent moths has been proposed for the reproductive suppression of lepidopteran pests. Combination with certain ecologically safe strategies like biological control may further improve the control of lepidopteran pests using F1 sterility; hence, the probability of integrating S. glaseri EPNs with this genetic control measure was investigated in the present study on S. litura. It is difficult to find complementary control strategies for synergistic use in conjunction with sterile moth release programmes. Gouge, Lee, Bartlett, and Henneberry (1998) suggested that S. carpocapsae may be an ideal entomopathogenic nematode to be used in conjunction with inherited sterility for the management of the pink bollworm, Pectinophora gossypiella,asS. carpocapsae would more likely infect the mobile native pink bollworm larvae than the sedentary F1 larvae from irradiated parents. S. carpocapsae basically is a passive ambusher, but in our present experiments, we have used the species S. glaseri, which is a cruise forager, and highly Downloaded By: [Hendrichs, Jorge] At: 15:50 6 November 2009 mobile and responsive to long-range host volatiles. S. glaseri is best adapted to parasitize hosts possessing low mobility and residing within the soil profile, as suggested by Gaugler, Campbell, Selvan, and Lewis (1992). We have attempted to study the feasibility of integrating the use of EPNs with F1 sterility and have focused on investigations regarding the interaction with F1 sterile insects of normal EPNs, and EPNs that had been cultured in radio-sterilized insects. F1 insects, derived from 100 Gy irradiated male parent, were acceptable and suitable hosts for EPNs, almost with the same degree as untreated insects (control), whereas F1 insects (from 130 Gy treated parents) were relatively less acceptable hosts for parasitisation by S. glaseri EPNs. These F1 larvae (progeny of 130 Gy treated male) were slightly sluggish as well. Probably the host volatiles and nutritional quality of these treated larvae were adversely affected; and this could be one of the causes of the reduced infectivity potential of S. glaseri. Therefore, simultaneous release of EPNs (in inundative mode) with F1 insects from 100 Gy treated parents might be good proposition unlike with F1 insects from 130 Gy treated parents. Moreover, due 152 R.K. Seth et al.

to reduced mating competitiveness and reduced sperm transfer by F1 progeny insects (from 130 Gy male parent), 130 Gy has been proposed as relatively less preferred than 100 Gy dose to be employed in F1 sterility technique (Seth and Sharma 2001). Hence, release of EPNs along with F1 sterile insects might limit or influence the effectiveness of F1 sterility for pest suppression, depending upon the gamma dose (to be used in F1 sterility) and the timing of EPNs release. Further, since the compatibility of two control measures was confirmed, and the pest population suppression was feasible by both techniques, a management strategy could be devised. In this situation, simultaneous application of both tactics, due to acceptability and suitability of F1 insects as host for EPNs reared in radio-sterilized host, may show an additive effect, because these two methods are not antagonistic to each other. Provided timing and logistics are taken into consideration, synergy may be achieved in response to inoculative release of EPNs along with F1 sterility. As per review of our investigations, the use of ‘genetic pest control method’ (F1 sterility technique) in conjunction with EPNs could be a feasible strategic component in IPM of S. litura, in which operational modality might be either (i) ‘sequential’, i.e., EPNs application preceding the use of F1 sterility so as to reduce the load of release of sub-sterilized moths, or (ii) ‘simultaneous’ for some initial specific phase, because F1 insects (at 100 Gy) would be equally acceptable as normal insects, and pest suppression would be operating against different stages of the life cycle, i.e., against larvae and pupae (through EPNs) and against adults (via F1 sterility). The intermittent (inundative) releases of EPNs could also be effectively pursued, alternately with F1 sterility, so as to keep the pest population below the economic threshold. Further, in a situation where F1 sterility has been successful in suppression of pest population, the inoculative releases of EPNs could be considered for biological pest management, especially as a quarantine measure along with release of partially sterile insects, in view of bio-infective potential of IJs, cultured in F1 sterile insects, observed in the present study. Cautious field simulated studies are warranted to judge the operational approach with respect to pest density and timing, so as to optimally integrate the use of EPNs with F1 sterility. The inundative and inoculative releases of EPNs might be possible and effective in view of acceptability and suitability of normal and irradiated S. litura and their F1 sterile progeny as hosts of EPNs, according to the present investigation, Downloaded By: [Hendrichs, Jorge] At: 15:50 6 November 2009 where the pest was found to be responsive towards both control tactics.

Acknowledgements Financial assistance by the International Atomic Energy Agency, Vienna is gratefully acknowledged for supporting this research work under Research Contract No. IAEA/IND- 10847/R0/RB as part of a Coordinated Research Project. Thanks are due to the technical support provided by Mr Manas K. Dhal.

References Anisimov, A.I. (1998), ‘Computer Simulations of Population Suppression by Irradiated and Mutant Lepidoptera Pests’,inBook of Abstracts FAO/IAEA International Conference on Area-Wide Control of Insect Pests Integrating the Sterile Insect and Related Nuclear and Other Techniques Penang, Malaysia,28MayÁ2 June, 1998. IAEA-CN-71, 223 pp. Armes, N.J., Wightman, J.A., Jadhav, D.R., and Ranga Rao, G.V. (1997), ‘Status of Insecticide Resistance in Spodoptera litura in Andhra Pradesh, India’, Pesticide Science, 50, 240Á248. Biocontrol Science and Technology 153

Bari, M.A., and Kaya, H.K. (1984), ‘Evaluation of the Entomogenous Nematode Neoaplec- tana carpocapsae (Steinernema feltiae) Weiser (Rhabditida: Steinernematidae) and the Bacterium Bacillus thuringiensis Berliner var. kurstaki for Suppression of the Artichoke Plume Moth (Lepidoptera: Pterophoridae)’, Journal of Economic Entomology, 77, 225Á229. Bilgrami, A.L., Kondo, E., and Yoshiga, T. (2000), ‘Experimental Models for Testing Attraction and Preferential Behaviour of Steinernema glaseri to Several Insects’, Japanese Journal of Nematology, 30, 35Á46. Bloem, S., Bloem, K.A., Carpenter, J.E., and Calkins, C.O. (2001), ‘Season-long Releases of Partially Sterile Males for Control of Codling Moth, Cydia pomonella (Lepidoptera: Tortricidae), in Washington Apples’, Environmental Entomology, 30, 763Á769. Carpenter, J.E. (1993), Integration of Inherited Sterility and other Pest Management Strategies for Helicoverpa zea,inProceedings of the International Symposium on Management of Insect Pests: Nuclear and Related Molecular and Genetic Techniques, IAEA/FAO, Vienna, Austria, pp. 363Á370. Carpenter, J.E., and Bartlett, A.C. (1999), Genetic Approaches to Managing Arthropod Pests, in Handbook on Pest Management, ed. J.R. Ruberson. New York: Marcel Dekker, pp. 487Á 519. Carpenter, J.E., and Wiseman, B.R. (1992a), ‘Spodoptera frugiperda (Lepidoptera: Noctuidae) Development and Damage Potential as Affected by Inherited Sterility and Host Plant Resistance’, Environmental Entomology, 21, 57Á60. Carpenter, J.E., and Wiseman, B.R. (1992b), ‘Effects of Inherited Sterility and Insect Resistant Dent-corn Silks on Helicoverpa zea (Lepidoptera: Noctuidae) Development’, Journal of Entomological Science, 27, 413Á420. Carpenter, J.E., Hidrayani, and Sheehan, W. (1996), ‘Compatibility of F1 Sterility and a Parasitoid, Cotesia marginiventris (Hymenoptera: Braconidae), for Managing Spodoptera exigua (Lepidoptera: Noctuidae): Acceptability and Suitability of Hosts’, Florida Entomol- ogist, 79, 289Á295. Carpenter, J.E., Bloem, S., and Marec, F. (2005), Inherited Sterility in Insects, in Sterile Insect Technique. Principles and Practice in Area-wide Integrated Pest Management, eds. V.A. Dyck, J. Hendrichs and A.S. Robinson, Dordrecht, The Netherlands: Springer, pp. 115Á146. Chari, M.S., and Patel, N.G. (1983), ‘Cotton Leafworm Spodoptera litura (Fabr.): Its Biology and Integrated Control Measures’, Cotton Development, 13, 7Á8. Choo, H.Y., Kaya, H.K., and Reed, D.K. (1989), ‘Biological Control of Onion Maggot and Tobacco Cutworm with Insect Parasitic Nematodes, Steinernema feltiae and Heterorhabditis heliothidis’, Korean Journal of Applied Entomology, 27, 185Á189. Epsky, N.D., and Capinera, J.L. (1994), ‘Invasion Efficacy as a Measure of efficacy of Entomogenous Nematode Steinernema carpocapsae (Rhabditida: Steinernematidae)’, Journal of Economic Entomology, 87, 366Á370. Gaugler, R.J., Campbell J.F., Selvan, M., and Lewis E. (1992), ‘Large-scale Inoculative Downloaded By: [Hendrichs, Jorge] At: 15:50 6 November 2009 Releases of the Entomopathogen Steinernema glaseri: Assessment 50 Years Later’, Biological Control, 2, 181Á187. Gouge, D.H., Lee, L.L., Bartlett, A., and Henneberry, T.J. (1998), ‘Pectinophora gossypiella (Lepidoptera: Gelechiidae): Susceptibility of F1 Larvae from Irradiated Parents to Entomopathogenic Nematodes (Rhabditida: Steinernematidae, Heterorhabditidae)’, Journal of Economic Entomology, 91, 869Á874. Greany, P.D., and Carpenter, J.E. (1999), ‘Use of Nuclear Techniques in Biological Control of Insects and Weeds’, Nuclear News, 42, 32Á34. Grewal, P.S., and Georgis, R. (1998), Entomopathogenic Nematodes, in Biopesticides: Use and Delivery, eds. F.R. Hall and J.J. Menn, Totowa, NJ: Humana Press, pp. 271Á299. Grewal, P.S., Selvan, S.A., and Gaugler, R. (1994), ‘Thermal Adaptation of Entomopatho- genic Nematodes Á Niche Breadth for Infection, Establishment and Reproduction’, Journal of Thermal Biology, 19, 245Á253. Hamm, J.J., and Carpenter, J.E. (1997), ‘Compatibility of Nuclear Polyhedrosis Viruses and Inherited Sterility for Control of Corn Earworm and Fall Armyworm (Lepidoptera: Noctuidae)’, Journal of Entomological Science, 32, 48Á153. Higuchi, H., Yamamoto, H., and Suzuki, Y. (1994), ‘Analysis of Damage to Soyabeans Infested by the Common Cutworm Spodoptera litura Fabricius (Lepidoptera:Noctuidae). 154 R.K. Seth et al.

II. Estimation of Leaf Area Damged by Young Larvae Using Spectral Reflectivity’, Japanese Journal of Applied Entomology and Zoology, 38, 297Á300. Kaya, H.K., and Gaugler, R. (1993), ‘Entomopathogenic Nematodes’, Annual Review of Entomology, 38, 181Á206. Kaya, H.K., and Hara, A.H. (1980), ‘Differential Susceptibility of Lepidopterous Pupae to the Nematode Neoaplectana carpocapsae’, Journal of Invertebrate Pathology, 36, 389Á393. Kaya, H.K., and Koppenhofer, A.M. (1996), ‘Effects of Microbial and other Antagonistic Organism and Competition on Entomopathogenic Nematodes’, Biocontrol Science and Technology, 6, 357Á371. Knipling, E.F. (1970), ‘Suppression of Pest Lepidoptera by Releasing Partially Sterile Males: A Theoretical Appraisal’, BioScience, 20, 465Á470. Kondo, E., and Ishibashi, N. (1986), ‘Infectivity and Propagation of Entomogenous Nematodes, Steinernema spp., on the Common Cutworm, Spodoptera litura (Lepidoptera: Noctuidae)’, Applied Entomology and Zoology, 21, 95Á108. LaChance, L.E. (1985), Genetic Methods for the Control of Lepidopteran Species: Status and Potential, USDA, Washington, DC, Agricultural Research Series (ARS), 28, 40 pp. Lefroy, H.M. (1908), ‘The Tobacco Caterpillar, Prodenia littoralis’, Memoirs of the Department of Agriculture in India Entomological Series,2,79Á93. Moussa, M.A., Zaher, M.A., and Kotby, V. (1960), ‘Abundance of Cotton Leafworm, Prodenia litura (F.) in Relation to Host Plants. I. Host Plants and their Effect on Biology’, Bulletin of the Society of Entomology Egypt, 44, 241Á251. Narayanan, K., and Gopalakrishnan, C. (1987), ‘Effect of Entomogenous Nematode, Steinernema feltiae (Rhabditida: Steinernematidae) to the Pre-pupa, Pupa and Adult of Spodoptera litura (Noctuidae: Lepidoptera)’, Indian Journal of Nematology, 17, 273Á276. North, D.T. (1975), ‘Inherited Sterility in Lepidoptera’, Annual Review of Entomology, 20, 167Á182. North, D.T., and Holt, G.G. (1971), Inherited Sterility and Its Use in Population Suppression of Lepidoptera, in Applications of Induced Sterility for Control of Lepidopterous Populations, International Atomic Energy Agency, Vienna. Poinar, Jr, G.O. (1979), Nematodes for Biological Control of Insects, Boca Raton, FL: CRC Press Inc, p. 277. Ramakrishnan, N., Saxena, V.S., and Dhingra, S. (1984), ‘Insecticide Resistance in the Population of Spodoptera litura (Fb.) in Andhra Pradesh’, Pesticides, 18, 23Á27. Seth, R.K., and Barik, T.K. (2007), ‘Effect of Host Irradiation on Bio-infectivity and Proliferation Capacity of Steinernema glaseri as Entomopathogenic Nematodes on a Serious Tropical Pest, Spodoptera litura’, Journal of Nuclear Agriculture and Biology, 36, 81Á 101. Seth, R.K., and Sehgal, S.S. (1993), ‘Partial Sterilizing Radiation Dose-effect on the F1 Progeny of Spodoptera litura (Fabr.): Growth, Bioenergetics and Reproductive Compe- Downloaded By: [Hendrichs, Jorge] At: 15:50 6 November 2009 tence’, pp. 427Á440, in Proceedings, Management of Insect Pests: Nuclear and Related Molecular and Genetic Techniques, eds. P. Howard-Kitto, R.F. Kelleher, and G.V. Ramesh, Vienna, 19Á23 October 1992. International Atomic Energy Agency, Vienna, pp. 669. Seth, R.K., and Sharma, V.P. (2001), ‘Inherited Sterility by Substerilizing Radiation in Spodoptera litura (Lepidoptera: Noctuidae): Bioefficacy and Potential for Pest Suppression’, Florida Entomologist, 84, 183Á193. Sezhian, N., Sivakumar, C.V., and Venugopal, M.S. (1996), ‘Alteration of Effectiveness of Steinernema carpocapsae Weiser (Steinernematidae: Rhabditida) against Spodoptera litura (F.) (Noctuidae: Lepidoptera) Larvae on Sunflower by Addition of an Insect Phagostimu- lant’, Indian Journal of Nematology, 26, 77Á81. Shapiro, D.I., and Glazer, I. (1996), ‘Comparison of Entomopathogenic Nematode Dispersal from Infected Hosts Versus Aqueous Suspension’, Environmental Entomology, 25, 1455Á 1461. Shapiro, D.I., and Lewis, E.E. (1999), ‘Comparison of Entomopathogenic Nematode Infectivity from Infected Hosts Versus Aqueous Suspension’, Environmental Entomology, 28, 907Á911. Snedecor, G.W., and Cochran, W.G. (1989), Statistical Methods, (8th ed), Ames, IA: Iowa State University Press. Biocontrol Science and Technology 155

Stuart, R.J., Lewis, E.E., and Gaugler, R. (1996), ‘Selection Alters the Pattern of Emergence from the Host Cadaver in the Entomopathogenic Nematode, Steinernema glaseri’, Parasitology, 113, 183Á189. White, G.F. (1927), ‘A Method for Obtaining Infective Nematode Larvae from Cultures’, Science, 66, 302Á303. Woodring, J.L., and Kaya, H.K. (1988), ‘Steinernematid and Heterorhabditid Nematodes: A Handbook of Biology and Techniques’, Southern Cooperative Series Bulletin 331, Arkansas Agricultural Experiment Station. Fayetteville, Arkansas, pp. 1Á30. Downloaded By: [Hendrichs, Jorge] At: 15:50 6 November 2009 Biocontrol Science and Technology, Vol. 19, S1, 2009, 157Á165

RESEARCH ARTICLE Parasitism rate and sex ratio of Psyttalia (Opius) concolor (Hymenoptera: Braconidae) reared on irradiated Ceratitis capitata larvae (Diptera: Tephritidae) Bahriye Hepdurgun*, Tevfik Turanli, and Aydin Zu¨mreog˘lu

Plant Protection Research Institute, Genc¸lik caddesi No. 6, 35040 Bornova, I˚zmir, Turkey

Tests were conducted to evaluate use of irradiated Mediterranean fruit fly (medfly), Ceratitis capitata (Wiedemann) (Diptera: Tephritidae), larvae as factitious hosts for mass-rearing the parasitoid Psyttalia concolor (Szepligeti) (Hymenoptera: Braconidae). In order to prevent the release of the alternative host adults, eclosion from unparasitized larvae must be prevented. Exposure of medfly larvae to 40, 50 or 60 Gy of 60Co gamma radiation showed that 60 Gy was the most effective dose in inhibiting adult medfly eclosion. The most suitable host:parasitoid ratio was found to be three larvae per parasitoid regardless of exposure time. When three larvae per parasitoid were exposed for 3 h, there was no significant difference in parasitism rates using irradiated vs. unirradiated host larvae (16.4 vs. 18%, respectively). Irradiation of host larvae also had no significant effect on the sex ratio of resulting parasitoids. Implementation of this practice will improve the efficiency of mass production and release of this biocontrol agent. Keywords: gamma radiation; Ceratitis capitata; Bactrocera oleae; Psyttalia concolor; mass rearing

Introduction The olive fruit fly, Bactrocera oleae (Gmelin) (Diptera: Tepritidae), is one of the most important pests in olive orchards in Turkey. Infested fruits fall and the level of oil acidity may increase by 10Á12% in damaged fruits. Damage levels of 30Á50% are not

Downloaded By: [Hendrichs, Jorge] At: 15:51 6 November 2009 uncommon (Hepdurgun, Koc¸lu, Zu¨mreog˘lu, and Turanli 2004) and can reach 90Á 100% damage in epidemic years (Aysu 1957), especially in early maturing pulpy and oily olive fruit cultivars. Aerial bait-spraying (bait attractantinsecticide), organized by the government (Plant Protection Research Institute, PPRI, of Bornova, Turkey), has been used to achieve area-wide control over 15 million trees in coastal areas of the Aegean Region since 1980 (Pala, Zu¨mreog˘lu, Fidan, and Altin 1997). Although bait-spraying is less harmful to beneficial organisms and the environment than traditional cover spraying (insecticide alone), non-target effects still occur. Accordingly, alternate strategies that further minimize detrimental treatment side-effects, including mass-rearing and release of parasitoids, continue to gain in importance in the control of B. oleae. To date, a substantial amount of B. oleae biological control research has been carried out throughout the Mediterranean basin (Bjelis, Pelicaric, and Masten

*Corresponding author. Email: [email protected]

First Published Online 6 July 2009 ISSN 0958-3157 print/ISSN 1360-0478 online # 2009 Taylor & Francis DOI: 10.1080/09583150903090479 http://www.informaworld.com 158 B. Hepdurgun et al.

2003). Exploration for B. oleae natural enemies led to discovery of a larvalÁpupal endoparasitoid in Tunisia by Marchal (1910), described as Opius concolor (Hymenoptera: Braconidae) by Szepligeti. This species was placed in the subgenus Psyttalia by Fischer (1987) and subsequently elevated to generic rank by Wharton (1987). Most of the countries that suffer from B. oleae damage have conducted studies on the rearing of P. concolor for use in biological control of the pest (Biliotti and Delanoue 1959; Jannone and Binaghi 1959; Monastero 1959; Monastero and Genduso 1962; Fenili and Pegazzano 1965; Brnetic 1973; Avilla and Albajes 1983; Raspi and Loni 1994). Additional studies on the hostÁ parasitoid relationship were conducted by Avilla and Albajes (1984). It was found by Delanoue in 1958 (Arambourg 1983) that C. capitata could serve as a factitious host for rearing P. concolor. Because techniques for mass rearing C. capitata are advanced, it was chosen as a preferred laboratory host for parasitoid mass production. Nuclear techniques can play an important role in augmentative biological control, for example by facilitating mass rearing (Greany and Carpenter 2000). In recent years, irradiating host larvae before exposure to parasitoids has been an important technique in the mass rearing of fruit fly parasitoids to prevent the emergence of adult flies, thus eliminating concerns about releasing or needing to separate host material when conducting parasitoid releases (Sivinski and Smittle 1990). Large scale rearing of the braconid parasitoid Diachasmimorpha longicaudata (Ashmead) has been undertaken in the USA and Mexico using irradiated Anastrepha suspensa (Loew) (Sivinski et al. 1996) and Anastrepha ludens (Cancino, Ruiz, Gomez, and Toledo 2002), respectively as hosts, and the technique has been examined in numerous other locations and with a variety of parasitoid species (e.g., this volume). The studies described below were conducted using C. capitata to mass rear P. concolor as a means to develop an environmentally friendly approach for area- wide control of B. oleae. The radiation dose needed to completely prevent adult eclosion of C. capitata was determined and optimal parasitization exposure time and host:parasitoid ratios were established. The parasitism of irradiated vs. unirradiated mature medfly larvae was compared, and the affect of radiation on parasitoid sex

Downloaded By: [Hendrichs, Jorge] At: 15:51 6 November 2009 ratio determined.

Materials and methods Stock cultures of C. capitata and P. concolor were kept at 2591oC, 6595% relative humidity and a photoperiod 16 h L:8 h D. C. capitata larvae were reared according to the method described by Zu¨mreog˘lu (1979), using an artificial medium. For the mass rearing of P. concolor, rearing procedures outlined in the parasitoid rearing manuals from Hawaii and Guatemala were used (Anonymous 1997, 1998). Mature third instar medfly larvae to be used as hosts in the rearing of P. concolor were collected in trays of water placed at the bottom of cabinets after they exited the rearing trays. The mass reared P. concolor strain in Turkey was originally set-up with wasps derived from the National Agricultural Research Foundation (NA- GREF) of Greece, which had origins from collections in Greece and Italy (Karam et al. 2008). Biocontrol Science and Technology 159

Irradiation dose Irradiation tests were performed to determine the dose of gamma radiation applied to medfly larvae required to prevent adult emergence. Groups of ca. 2500 mature larvae in water in Petri dishes were irradiated at 40, 50 or 60 Gy; ca. 2500 larvae from the same batch were left untreated as controls. Irradiation was performed using a type C-146, 7000 C: Cobalt-60 Theratron irradiator, delivering 107.33 cGy/min. Percent emergence for the irradiated (40, 50 and 60 Gy) and non-irradiated control larvae was determined by placing each group of 2500 larvae in a wooden box containing fine sand and kept in a pupation room at 2091oC and 7095% RH. Pupae were sifted 2 days before the expected emergence date and four separate groups of 400 pupae were randomly selected from each treatment and placed in 946 mL polystrene containers. Cube sugar and moist sponges were placed on the screen tops of the containers to provide food and water for the emerged flies. Adult emergence from the pupae was checked every other day for 10 days and the numbers of deformed pupae, half-emerged flies and fully emerged flies were recorded.

Determination of optimal host:parasitoid ratio and exposure time Tests were conducted to determine the optimal host:parasitoid ratio and exposure duration. For these tests, the following host larvae (L):à parasitoid (P) ratios were evaluated: 1:1, 2:1 and 3:1 and these ratios were tested using 1-, 2-, and 3-h exposure periods. Thus, for each series of tests, nine cages were set-up (3 L:à P ratios  3 exposure times). The tests were carried out under laboratory conditions (2591oC, 6595 RH) using 304030 cm aluminum cages (frame and bottom) covered with fine mesh organdy cloth. Each cage was populated with 50 à and 50 ß parasitoids that had emerged on the same day a given test series was initiated. Medfly larvae were exposed to the parasitoids using ‘sting units’, which were prepared using two interlocking PVC rings approximately 11 cm in diameter, with unirradiated third instar medfly larvae sandwiched between two pieces of fine mesh organdy cloth held

Downloaded By: [Hendrichs, Jorge] At: 15:51 6 November 2009 in place by the rings. Each sting unit contained either 50, 100 or 150 larvae depending on the L:P ratio being tested (1:1, 2:1 or 3:1, respectively) and was placed in a cage for 1Á3 h. This procedure was followed using the same cages of parasitoids for 5 consecutive days. Means for each treatment over the 5 consecutive days were considered as one replicate. The entire procedure was replicated on five separate occasions. Larvae taken from the sting units from each cage after parasitoid exposure were transferred to labeled jars (12 cm height, 15 cm diameter) containing fine sawdust. The jars were ventilated using organdy cloth and kept at 2591oC, 6595RHfor adult emergence. Following completion of adult emergence and death of the individuals, all parasitoids produced in each jar were counted as male or female. The 5-day means of each parameter were subjected to SAS statistical analysis and pooled. The five replicates over time were also subjected to SAS statistical analysis (randomized complete block design) and the means were ranked according to the LSD test. 160 B. Hepdurgun et al.

Parasitoid development on irradiated and unirradiated medfly larvae For this experiment, two cages each were provisioned with 500 pairs of newly emerged parasitoids. Sting units were then prepared containing either ca. 750 unirradiated or 750 irradiated (60 Gy) mature medfly larvae. Two of the same type sting units were introduced into each cage (3L:1P) for a period of 3 h. This procedure was followed twice-a-day for 5 consecutive days. The larvae were taken from each cage/sting unit at the end of each parasitization period and were transferred to labeled plastic trays containing fine sawdust. Pupae were sifted 6 days after pupation, left for 1 day in pupal trays for air circulation, and then placed in Petri dishes labeled according to their parasitism dates and times and whether or not they were from irradiated or unirradiated larvae. From each day’s collection of pupae, 50 pupae were randomly selected from each of the AM and PM exposures (100 total) for dissection to determine ‘expected’ rates of parasitism and another 100 pupae similarly selected were set-up in emergence grids to determined ‘observed’ rates of parasitism. The remaining pupae from each day were pooled (AM & PM exposures) and placed in separate emergence cages, five (one for each day) with pupae from irradiated larvae and five with pupae from unirradiated larvae. Upon emergence, 100 à and 100 ß were randomly selected from each emergence cage and placed in individual parasitation cages. A sting unit was then placed in each cage for 3 h that contained 5 mL larvae (ca. 300), giving a ratio of 3L:1P. This process was repeated once-a-day for 10 consecutive days. Pupae were collected as previously described. Parasitism efficiency comparing F1 adults reared from irradiated and unirradiated larvae was determined by placing 100 pupae from each trial into eclosion grids. Emerged F2 individuals were counted and sexed to determine percent successful parasitism and sex ratio.

Results Our experiments showed that no medfly adults emerged when mature third instar medfly larvae were irradiated at 60 Gy using a 60Co gamma radiation source. The percentages of adult emergence from pupae whose larvae were irradiated at the doses of 50 and 40 Gy were 0.56 and 9.18%, respectively. Percent adult emergence from Downloaded By: [Hendrichs, Jorge] At: 15:51 6 November 2009 pupae of unirradiated larvae averaged 94.75% (Table 1). Tests to determine the optimal host larvae (L) to female parasitoid (P) ratio and duration of larval exposure to parasitoids indicated that the highest numbers of parasitoids were recovered from treatments using the highest numbers of larvae regardless of the exposure time. Percent parasitism for the 3L:1P ratios using 150 larvae to 50 à parasitoids were 40.32, 43.96 and 42.84% for 1-, 2- and 3-h exposures, respectively (Table 2). Percent parasitism using 2L:1P and 1L:1P and a 3-h exposure were 23.60 and 7.48%, respectively. The rate at which irradiated (60 Gy) larvae were parasitized was not significantly different from that of unirradiated larvae. In these experiments, the mean parasitism rate of irradiated larvae based on pupal dissections was 18.4 vs. 19.2% for unirradiated larvae. The mean parasitism rates based on parasitoid emergence from pupae of irradiated and unirradiated larvae were 16.4 and 18.0%, respectively (Table 3). The use of irradiated vs. unirradiated host larvae also did not affect the sex ratio of emerging F1 parasitoids. In both cases the proportion of females was higher Biocontrol Science and Technology 161

Table 1. Comparative effect of various irradiation doses applied to third instar medfly larvae on adult eclosion.

Deformed Half-emerged Adult Treatment Replicate1 pupae (%) pupae (%) eclosion (%)

40 Gy 1 84.50 8.75 6.75 2 82.25 9.00 8.75 3 74.00 13.50 12.25 4 80.50 10.50 9.00 Mean 80.31 10.43 9.18 50 Gy 1 99.50 0.25 0.25 2 99.50 0.50 0.00 3 98.00 0.00 2.00 4 99.75 0.25 0.00 Mean 99.18 0.25 0.56 60 Gy 1 0 0 0 20 0 0 30 0 0 40 0 0 Mean 0 0 0 Non-Irrad. Control 1 2.75 2.00 95.25 2 3.00 1.25 95.75 3 5.50 2.00 92.50 4 2.50 2.00 95.50 Mean 3.43 1.81 94.75

1400 pupae were set-up per replicate.

than males Á 1.0:2.9 male:female using irradiated larvae and 1.0:2.1 male:female using unirradiated larvae. The use of irradiated larvae as hosts did not affect the quality of the F1 adults produced based on their parasitism rates over a 10-day period, as none of the day-

Table 2. Percent parasitism by Psyttalia concolor at various unirradiated host:parasitoid

Downloaded By: [Hendrichs, Jorge] At: 15:51 6 November 2009 ratios and exposure times.

Treatment

No. host No. à L:P Exposure % Parasitism larvae (L) parasitoids (P) ratio time (h) (Mean9SE)1

50 50 1:1 1 13.88 de92.43 100 50 2:1 1 27.52 c97.87 150 50 3:1 1 40.32 ab98.59 50 50 1:1 2 12.20 e93.86 100 50 2:1 2 30.92 bc97.50 150 50 3:1 2 43.96 a99.73 50 50 1:1 3 7.48 e93.28 100 50 2:1 3 23.60 cd95.64 150 50 3:1 3 42.84 a99.53

1Means followed by different letters are significantly different (LSD test, P0.05). 162 B. Hepdurgun et al.

Table 3. Parasitism rates by Psyttalia concolor and sex ratio of emerging F1 progeny using irradiated (60 Gy) and unirradiated medfly larvae.

Expected1 Observed2 Sex ratio of parasitism rate (%) parasitism rate (%) emerging adults (M:F)

Exposure day Irrad. Unirrad. Irrad. Unirrad. Irrad. Unirrad.

1 10 16 6 8 1:5.0 1:1.7 2 10 22 7 18 1:1.3 1:1.0 3 44 22 25 23 1:4.0 1:1.6 4 12 16 28 24 1:2.5 1:1.4 5 16 20 16 17 1:1.7 1:4.7 Mean 18.4 19.2 16.4 18 1:2.9 1:2.1

1Expected percent parasitism was determined by pupal dissections. 2Observed percent parasitism was determined by adult eclosion.

wise comparisons between adults from irradiated vs. unirradiated larvae were significantly different (Table 4). The sex ratios of the F2 adults produced in these trials were also similar. There was a marked decline in total percent parasitism over the 10-day period by F1 parasitoids obtained using both irradiated and unirradiated medfly larvae, decreasing from around 60% on day 1 to around 16% on day 10.

Discussion Our results showed that irradiated medfly larvae could be used to mass rear the olive fruit fly parasitoid P. concolor without having to worry about separating or releasing adult medflies. Our results also indicated that the quality of the parasitoids produced using irradiated and unirrradiated medfly larvae was not significantly different. The

Table 4. Percent parasitism by F1 P. concolor adults produced from either irradiated (60 Gy) or unirradiated medfly larvae.

1 Parasitism rate (%) Sex ratio of emerging F2 adults (M:F) Downloaded By: [Hendrichs, Jorge] At: 15:51 6 November 2009

F1 adults from F1 adults from F1 adults from F1 adults from Exposure day irrad. larvae unirrad. larvae irrad. larvae unirrad. larvae

1 60.0 60.6 1:2.7 1:3.7 2 49.6 47.0 1:3.0 1:3.3 3 42.2 40.0 1:2.0 1:2.9 4 32.4 27.8 1:2.4 1:3.2 5 28.4 28.6 1:3.1 1:5.5 6 20.2 21.4 1:3.4 1:5.7 7 13.2 13.0 1:6.3 1:2.1 8 16.8 18.4 1:3.0 1:3.2 9 22.8 11.8 1:2.5 1:4.9 10 16.2 16.0 1:2.4 1:6.3 Mean 30.18 28.46 1:3.1 1:4.1

1No day-wise comparisons were significantly different after transformation using a chi-square analysis. Biocontrol Science and Technology 163

complete inhibition of adult eclosion from mature medfly larvae irradiated with 60 Gy 60Co gamma radiation, as was found in the studies presented here, is in keeping with the results of Delrio (1994). Likewise, Cancino, Ruiz, Lo´pez, and Sivinski (2009) found that a dose of 60 Gy was needed to completely prevent the eclosion of adult medflies when these larvae were utilized in the rearing of the parasitoids D. longicaudata and D. tryoni (Cameron). Based on these studies, this technique was successfully used to conduct a field trial to control olive fruit fly populations in Turkey through a combination of mass trapping and mass releases of the parasitoid P. concolor reared on irradiated medfly larvae (Hepdurgun, Turanli, and Zu¨mreog˘lu 2009). As noted by Greany and Carpenter (2000), this is a more useful application of gamma radiation than that of Ramadan and Wong (1989), who exposed pupae of the Oriental fruit fly Bactrocera dorsalis (Hendel) to gamma radiation after having already exposed the larvae to parasitization by D. long- icaudata. This resulted in sterility of the adult parasitoids. In the present studies, parasitism rates were relatively low (generally below 50%), with the highest parasitism rates for P. concolor being achieved using three larvae per adult female. In hindsight, the low parasitism rates may have been due to a lack of sufficient host larvae resulting in superparasitism. Lawrence, Greany, Nation, and Baranowski (1978) found that for D. longicaudata, when 15 or more A. suspensa larvae were provided per female, the females discriminated between parasitized and non-parasitized hosts and oviposited preferentially in non-parasitized ones, which resulted in 70% parasitoid progeny survival. However, when there were only six host larvae per female, superparasitism occurred and parasitoid progeny survival was B30%. Ashley and Chambers (1979) also found that the production of progeny by D. longicaudata was affected by host availability, as well as parasitoid density, age, and previous ovipositional experience. In their experiments, maximum rearing efficiency was achieved at a hostÁlarva to parasitoid ratio of 4:1. Future studies may find further increases in rearing efficiency can be achieved for P. concolor using irradiated medfly larvae if higher larvae to parasitoid ratios than 3:1 are used.

Acknowledgements The authors thank Dr. Jorge Hendrichs, Project Coordinator and Head of the Insect Pest

Downloaded By: [Hendrichs, Jorge] At: 15:51 6 November 2009 Control Section, Joint FAO/IAEA, for his valuable support and interest during this project (no. 10783/TUR). Thanks are also due to Felipe Jeronimo and Gustavo Baeza for their kind assistance during the training of the author in rearing techniques at the Moscamed Programme facilities in Guatemala.

References Anonymous (1997), Procedural Manual for Mass Rearing Six Species of Tephritid Fruit Fly Parasitoids. USDA. Honolulu. Hawaii. Anonymous (1998), Mass Rearing Production Manual ‘Fruit fly parasitoids’. Aurora Rearing Facility, USDA/APHIS PPQ Guatemala. Arambourg, Y. (1983), Programme on Integrated and Biological Control. In Progress Report 1979Á1981, Luxembourg (LU): Commission of the European Communities, pp. 103Á110. Ashley, T.R., and Chambers, D.L. (1979), ‘Effects of Parasite Density and Host Availability on Progeny Production by Biosteres (Opius) longicaudatus (Hym.: Braconidae), a Parasite of Anatrepha suspensa (Dip.: Tephritidae)’, Entomophaga, 24, 363Á369. Avilla, J., and Albajes, R. (1983), ‘Prieliminary Studies on Mutual Interference in Opius concolor Szepl. (Hym.: Braconidae)’, Journal of Applied Entomology, 96, 27Á32. 164 B. Hepdurgun et al.

Avilla, J., and Albajes, R. (1984), ‘The Influence of Female Age and Host Size on the Sex Ratio of the Parasitoid Opius concolor’, Entomologia Experimentalis et Applicata, 35, 43Á47. Aysu, R. (1957), ‘Dacus oleae Rossi. Zeytin sine(i’, Tomurcuk, 63, 15Á19. Biliotti, E., and Delanoue, P. (1959), ‘Contribution a l’e´tude biologique d’Opius concolor Sze´pl. (Hym. Braconidae) en e´levage de laboratoire’, Entomophaga,4,7Á14. Bjelis, M., Pelicaric, V., and Masten, T. (2003), ‘Olive Fruit Fly Á Bactrocera oleae Gmelin (Diptera. Tephritidae) in Croatia; Damage in New Milenium and Advanced methods of Control’, 1st European Meeting of the IOBC/WPRS Study Group ‘Integrated Control in Olives’ 2003 May 29Á31; MAICh-Chania. Crete/Greece. Brnetic, D. (1973), ‘Artifical Rearing of the Wasp Opius concolor Szepl. and Possibility of its Employment to Control the Olive Fly (Dacus oleae Gmel.) in Dalmatia’s Olive-groves’, Final Report of the project No. E 30-ENT 8. Institute for Adriatic Agriculture and Karst Reclamation. Split. Yugoslavia, 193 pp. Cancino, J., Ruiz, L., Gomez, Y., and Toledo, J. (2002), ‘Irradiacio´n de larvas deAnastrepha ludens (Loew) (Diptera: Tephritidae) para inhibir la emergencia de moscas en la cr´ıa del parasitoide Diachasmimorpha longicaudata (Ashmead) (Hymenoptera: Braconidae)’, Folia Entomologia Mexico, 41, 195Á208. Cancino, J., Ruiz, L., Lo´pez, P., and Sivinski, J. (2009), ‘The Suitability of Anastrepha spp. and Ceratitis capitata (Wiedeman) (Diptera: Tephritidae) Larvae as Hosts of Diachasmimorpha longicaudata (Ashmead) and Diachasmimorpha tryoni (Cameron) (Hymenoptera: Braconi- dae): Effects of Host Age and Radiation Dose and Implications for Quality Control in Mass Rearing’, Biocontrol Science and Technology (this volume). Delrio, G. (1994), Recent Research on Control Methods against Olive Fly in Italy. FAO. Inter- Regional Cooperative Research Network on olives. Marrequech. Marocco. 5Á7 Oct. 1994. Fenili, G.A., and Pegazzano, F. (1965), ‘Contributo alla conoscanza dei parassiti del Dacus oleae Gmel. Ricerche eseguite in Toscana negli anni 1967 e 1968’’, Redia, LII, 1Á29. Fischer, M. (1987), ‘Hymenoptera: Opiinae III Á athiopische, orientalische, australische und ozeanische’’, Region Das Tierreich, 104, 1Á734. Greany, P.D., and Carpenter, J.E. (2000), ‘Use of Nuclear Techniques in Biological Control’,in Area wide control of fruit flies and other insect pests Penerbit Universiti Sains Malaysia, , ed. K.-H. Tan, Pulau Pinang, pp. 221Á227. Hepdurgun, B., Koc¸lu, T., Zu¨mreog˘lu, A., and Turanli, T. (2004), ‘Preliminary Field Cage Studies on the Effectiveness of Psyttalia concolor Szepl. (Hym.: Braconidae) in Order to Suppress Bactrocera oleae Gmel. (Dip.: Tephritidae) Populations in Go¨kc¸eada Island’, Turkey. 5th International Olive Symposium 27 SeptemberÁ2 October 2004. I˚zmir/Turkey. Hepdurgun, B., Turanli, T., and Zu¨mreog˘lu, A. (2009), ‘Control of the Olive Fruit Fly, Bactrocera oleae (Diptera: Tephritidae) through Mass Trapping and Mass Releases of the Parasitoid Psyttalia concolor (Hymenoptera: Braconidae) Reared on Irradiated Mediterra- nean Fruit Fly’, Biocontrol Science and Technology (this volume). Downloaded By: [Hendrichs, Jorge] At: 15:51 6 November 2009 Jannone, G., and Binaghi, G. (1959), ‘Primi esperimenti di introduzione in Liguria di un endofago della mosca delle olive: Opius concolor Sze´pl. (O. siculus Mon.) (Hymenoptera: Braconidae) della Sicilia’, Bollettino del Laboratorio di Entomologia Agraria Portici, 17, 89Á 123. Karam, N., Guglielmino, C.R., Bertin, S., Gomulski, L.M., Bonomi, A., Baldacchino, F., Simeone, V., and Malacrida, A.R. (2008), ‘RAPD Analysis in the Parasitoid Wasp Psyttalia concolor Reveals Mediterranean Population Structure and Provides SCAR Markers’, Biological Control, 47, 22Á27. Lawrence, P.O., Greany, P.D., Nation, J.L., and Baranowski, R.M. (1978), ‘Oviposition Behavior of Biosteres longicaudatus, a Parasite of the Caribbean Fruit Fly, Anastrepha suspensa’, Annals Entomological Society of America, 71, 253Á256. Marchal, P. (1910), ‘Sur un braconide nouveau. parasite de Dacus oleae’, Bulletin de la Societe Entomologique de France, 13, 243Á244. Monastero, S. (1959), ‘Altra straordinaria cattura di Opius parassiti di Dacus oleae Gmel.. in Sicilia nel 1959’, Bollettino d.Istituto Entomologia e Osservatorio Fitopatologia di Palermo, III, 261Á269. Monastero, S., and Genduso, P. (1962), ‘La lotta biologica contro la mosca delle olive’, Bollettino d.Istituto Entomologia e Osservatorio Fitopatologia di Palermo,V,31Á53. Biocontrol Science and Technology 165

Pala, Y., Zu¨mreog˘lu, A., Fidan, U¨ ., and Altin, M. (1997), ‘Recent Integrated Pest Manage- ment Studies in Olive Orchards in Turkey’, Olivae, 68, 37Á38. Ramadan, M.M., and Wong, T.T.Y. (1989), ‘Effect of Gamma Radiation on Biosteres longicaudatus (Ashmead) (Hymenoptera: Brachonidae). A Larval Parasitoid of Dacus dorsalis Hendel (Diptera: Tephritidae)’, Proceedings of the Hawaiian Entomological Society, 29, 111Á113. Raspi, A., and Loni, A. (1994), ‘Alcune sull’allevamento massale di Opius concolor Sze´pligetti (Hym:Braconidae) e su recenti tentativi d’introduzione delle specie in Toscana e liguria’, Frustula Entomologica, n.s., 30, 135Á145. Sivinski, J.M., and Smittle, B. (1990), ‘Effects of Gamma Radiation on the Development of the Caribbean Fruit Fly (Anastrepha suspensa) and the Subsequent Development of its Parasite Diachasmimorpha longicaudata’, Entomologia Experimentalis et Applicata, 55, 295Á297. Sivinski, J.M., Calkins, C.O., Baranowski, R., Harria, D., Brambila, J., Diaz, J., Burns, R.E., Holler, T., and Dodson, G. (1996), ‘Suppression of a Caribbean Fruit Fly (Anastrepha suspensa (Loew) Diptera: Tephritidae) Population through Augmented Releases of the Parasitoid Diachasmimorpha longicaudata (Ashmead) (Hymenoptera: Braconidae)’, Biolo- gical Control, 6, 177Á185. Wharton, R.A. (1987), ‘Changes in Nomenclature and Classification of Some Opine Braconidae (Hymenoptera)’, Entomological Society of Washington, 89, 61Á73. Zu¨mreog˘lu, A. (1979), ‘Investigations on the Rearing of the Mediterranean Fruit flyon Artificial Mediums in Relation to the Application of Sterile-male Releasing Technique’, Izmir Bo¨lge Zirai Mu¨cadele Aras¸tirma Enstitu¨su¨ Mu¨du¨rlu¨˘gu¨ Aras¸tirma Eserleri Serisi No. 34. 94 p. Downloaded By: [Hendrichs, Jorge] At: 15:51 6 November 2009 Biocontrol Science and Technology, Vol. 19, S1, 2009, 167177

Irradiation of Anastrepha ludens (Diptera: Tephritidae) eggs for the rearing of the fruit fly parasitoids, Fopius arisanus and Diachasmimorpha longicaudata (Hymenoptera: Braconidae) Jorge Cancinoa*, Lia Ru´zı a, Jorge Pe´reza, and Ernest Harrisb

aDesarrollo de Me´todos, Campan˜a Nacional Contra Moscas de la Fruta, Tapachula, Chiapas, Mexico; bUS Pacific Basin Agricultural Research Center, USDA-ARS, Honolulu, HI, USA

Irradiated eggs of Anastrepha ludens were evaluated as hosts of two fruit-fly parasitoids for mass rearing. Three different ages of A. ludens eggs (24-, 48- and 72-h-old) were analyzed for hatchability after being subjected to radiation doses of 2.5, 5, 7.5, 10, 12.5, 15, 17.5, 20, 22.5, 25, 27.5 and 30 Gy. No significant reduction in hatchability occurred with the 72-h-old eggs at any of the radiation dose levels and no adult emergence occurred at radiation doses greater than 25 Gy. Seventy two-h-old eggs irradiated above 25 Gy were found to be the best age and dose for fruit fly egg hosts to be used in mass rearing the egg parasitoid Fopius arisanus. It was demonstrated that larvae hatching from the irradiated A. ludens eggs can also be used as hosts for Diachasmimorpha longicaudata. Parasitoid emergence of both species was not statistically different from the control group (parasitoids emerged from non-irradiated host). The fecundity of parasitoids emerged from irradiated hosts also was similar to that obtained with parasitoids reared with non-irradiated hosts. There were some statistical differences between the curves for longevity. However, these were not clearly correlated with radiation dose. The results of this study will aid in the design of improved methods for mass rearing and release of fruit-fly parasitoids. Keywords: egg irradiation; irradiated host; parasitoid mass-rearing

Introduction

Downloaded By: [Hendrichs, Jorge] At: 15:52 6 November 2009 Irradiated larvae have proven to be a significant resource for the mass rearing of fruit fly parasitoids (Sivinski and Smittle 1990; Cancino, Ru´z,ı Go´mez, and Toledo 2002a). Irradiated hosts that are not parasitized do not emerge, so that it is possible to work only with the parasitoids. Many activities such as host exposure, packing, and mass releases would be extremely difficult to carry out without the use of irradiation. Nowadays, in the mass rearing of parasitoids, the use of irradiated hosts is imperative. However, there may be opportunities to increase the efficiency of the irradiation procedures. Currently Diachasmimorpha longicaudata (Ashmead) (Hy- menoptera: Braconidae) is reared in irradiated larval hosts. These are irradiated before exposure to the parasitoids. Many facilities that raise D. longicaudata use irradiated hosts, for example, in Piracicaba, Brasil, La Molina, Lima, Peru; La Aurora, Guatemala; Florida, USA; and the Moscafrut Plant in Mexico (Sivinski

*Corresponding author. Email: [email protected]

ISSN 0958-3157 print/ISSN 1360-0478 online # 2009 Taylor & Francis DOI: 10.1080/09583150802439827 http://www.informaworld.com 168 J. Cancino et al.

et al. 1996; Rodr´ıguez, Quenta, and Molina 1997; Matrangolo, Nascimento, Carvalho, Melo, and De Jesu´s 1998; Menezes et al. 1998; Cancino et al. 2002a). If fruit fly eggs could be irradiated as opposed to late instar larvae, then smaller volumes would need to be exposed to radiation, decreasing the time required to handle host materials. At the same time, it would be easier to collect eggs from ‘oviposition cages’ than to remove larvae from diet, again resulting in increased savings of time and space. To this end, we examined the ability of the widely used larval-prepupal parasitoid D. longicaudata to develop in larvae derived from irradiated eggs. We compared both the survival of the parasitoid and the propensity of the host to complete its development when eggs are exposed to a variety of radiation doses. At the same time, we determined the ability of another parasitoid, Fopius arisanus (Sonan), an egg-prepupal braconid species, to develop in irradiated eggs. This species has been developed on Anastrepha eggs (Lawrence, Harris, and Bautista 2000; Zenil et al. 2004), although nowadays there are no strong results in its mass-rearing with Anastrepha. This could be a reason to believe that F. arisanus is not a good option as a natural enemy for Anastrepha fly species. However, recently in Mexico, mass-rearing of F. arisanus has been established successfully on Anastrepha ludens (Loew) eggs (Cancino and Ort´ız 2002). This could offer opportunities to propose new ideas for the use of F. arisanus against Anastrepha spp. The literature provides little guidance on parasitoid host egg irradiation. The range of doses reported in this paper was in part derived from previous studies of fly mortality following exposure to radiation (Rigney 1989). Post-harvest radiation treatments have been explored in Anastrepha spp. (Bustos, Enkerlin, Toledo, Reyes, and Casimiro 1992), but none of the studies investigated the effects that radiation might have on parasitoids.

Materials and methods A. ludens eggs used in the evaluations were provided by the Moscafrut Plant located in Metapa de Dom´ınguez, Chiapas, Mexico. The colony was maintained in accordance with the procedures described by Dom´ınguez, Castellanos, Herna´ndez, and Mart´ınez (2000). The F. arisanus parasitoids that were used had developed from host A. ludens eggs. Parasitoids from the 1020th generation were used during the Downloaded By: [Hendrichs, Jorge] At: 15:52 6 November 2009 development of the tests. The colony of the parasitoid D. longicaudata had been maintained for approximately 300 generations under mass-rearing conditions when the tests were initiated. A Gammacell 220 Co-60 irradiator with a dose rate of 2.53.0 Gy/min was used. The eggs were irradiated with an oxygen-free substitute for air (IAEA 1982) in a plastic container (53 cm, highwide) with 1 mL of eggs suspended in 3 mL of water. The times of exposure to the different doses were determined prior by Fricke’s dosimetry according to the Manual High-Dose Dosimetry in Industrial Processing (Zavala, Fierro, Schwarz, Orozco, and Guerra 1985; IAEA 2001; FAO/IAEA/ USADA 2003). The evaluations were performed as follows.

Irradiation of A. ludens eggs at different ages The first step was to evaluate the viability of A. ludens eggs using different ages and different doses of radiation. Eggs incubated for 24, 48 and 72 h were irradiated at Biocontrol Science and Technology 169

2.5, 5, 7.5, 10, 12.5, 15, 17.5 and 20 Gy. The hatching percentage was taken from a sample of 100 eggs placed on a piece of black paper held in a closed Petri dish containing a piece of wet filter paper. With the aid of a dissecting microscope, the number of eggs hatched was recorded every 24 h and was used to calculate the maximum percentage of eggs hatched for each age and dose of radiation. This test was replicated 13 times.

Effects of irradiation on 72-h-old A. ludens eggs The next step was to analyze the data from the previous step in order to select the best age for the egg hatching percentage component of the study. The best age was 72 h. Twelve different doses of radiation were applied to 72-h-old eggs of A. ludens. Doses of 2.5, 5, 7.5, 10, 12.5, 15, 17.5, 20, 22.5, 25, 27.5 and 30 Gy were applied. We increased the doses applied to 72-h-old eggs to assure no emergence of flies. The 72- h-old eggs were not affected by these high doses. The eggs were incubated at 268Cfor 24 h. The eggs hatched in the diet medium (larval diet mix of torula yeast, corn flour, corn cob fractions, sugar, citric acid, nipagin, sodium benzoate and water) and the larvae remained there for 9 days. After 9 days, the larvae were separated from the artificial diet medium by washing. The volume of larvae obtained for each dose was recorded. Samples containing 100 larvae were taken for each treatment. Each sample was placed in a cylindrical plastic container (94.5 cm) packed with fine vermiculite (SUNGRO Horticulture†), held at 268C, and allowed to develop until reaching the pupal stage. One day before emergence (on day 14 of the 15 days needed to complete pupal development), the samples of pupae were weighed for each treatment. The pupae were returned to the container for adult emergence. After emergence, the number of adult flies was counted for each treatment. With a sample of 10 :10 flies within a 10-cm3 glass cage, their longevity and fertility were recorded. Adult longevity was determined by monitoring mortality daily. When the females were 12 days old, their eggs were collected daily as they were laid into a green agar ball (3 cm diameter) covered with a piece of parafilm. Fertility was estimated using the percentage of hatched eggs. Percent hatch was obtained by counting the eclosed eggs on a piece of wet, black paper on the bottom of a Petri Downloaded By: [Hendrichs, Jorge] At: 15:52 6 November 2009 dish. Nine replications for longevity and fertility were done.

Exposure of irradiated eggs to F. arisanus For the 12 irradiated treatments, 72-h-old samples (about 1 mL of eggs) of A. ludens eggs were irradiated at 2.5, 5. 7.5, 10, 12.5, 15, 17.5, 20, 22.5, 25, 27.5 and 30 Gy. One treatment consisting of eggs that were not treated with radiation was used as a control. The 13 treatments were exposed in a ‘papaya unit’ (Harris and Okamoto 1991) which consisted of a piece of papaya (4.54.52.5 cm) with six holes carved into its surface. About 0.5 mL (approximately 12,000 eggs) of eggs per treatment were distributed in the holes. These units were introduced into a ‘Hawaii- type’ cage (272727 cm) (Wong and Ramadan 1992) containing 30 :15 of F. arisanus parasitoids and held for 10 h. Afterwards, the parasitized eggs were seeded onto the larval artificial diet for development. The mature larvae were then 170 J. Cancino et al.

separated from the diet and kept in trays of vermiculite to allow pupation. Fourteen days after pupation, samples of 100 pupae were placed into containers (94.5 cm) and the number and percentage of parasitoids emerging at each treatment dose was recorded. In order to confirm the longevity and fecundity of parasitoids emerged from each dose, a sample of 30 :15 emerged parasitoids was taken per treatment and placed inside a ‘Hawaii-type’ cage. Daily mortality was recorded starting from day 1. The fecundity was recorded by monitoring the daily exposition of eggs from 5- to 10-day-old parasitoids. The eggs were treated as mentioned above. The numbers of offspring emerged were compared with the number of live females per day. This test was replicated 10 times.

Exposure of larvae hatched from irradiated eggs exposed to D. longicaudata Twelve treatments were prepared consisting of 200 fruit fly larvae each, developed from 72-h-old eggs exposed to one of 12 different doses of radiation. The doses were the same as those applied previously. These larvae were put into an oviposition unit (plastic Petri dish covered with a fine mesh) containing artificial diet medium. The unit was placed inside a ‘Hawaii-type’ cage (272727 cm) containing 30 :15 D. longicaudata parasitoids for a period of 2 h. After exposure, the larvae were kept in cylindrical plastic containers (94.5 cm) with fine vermiculite at 268C in order to stimulate pupation. The flies and parasitoids that emerged from each treatment were counted to obtain the percent emergence. Ten replicates were carried out. The adult parasitoids that emerged were sampled, and a group of 30 :15 per treatment were placed into ‘Hawaii-type’ cages. These samples were evaluated to confirm the longevity and fecundity of parasitoids emerged from each dose. Mortality was recorded daily until the 20th day. When D. longicaudata females were 5 days old, an oviposition unit with 200 A. ludens (8-day-old) larvae was exposed to these parasitoids for 2 h. Larval exposure to parasitoids was performed during the following 10 days. The exposed larvae were kept in a container with fine vermiculite for 14 days. The number of adults that emerged each day was related to the respective number of living females in order to obtain the number of offspring per female per day. Downloaded By: [Hendrichs, Jorge] At: 15:52 6 November 2009

Data analysis The means for percentage hatched eggs, larval yield, and adult emergence (parasitoids and flies) were analyzed using ANOVA and Tukey’s multiple range test (a0.05). A Long-rank test (Francis, Green, and Payne 1993) was applied to the longevity data. These data were analyzed with the JMP software (SAS Institute 2002).

Results Irradiation of A. ludens eggs at different ages Figure 1 shows the percentage of eggs hatching at different radiation doses. There were clear differences in the percent hatching in 24- and 48-h-old eggs with increasing doses of radiation (24 h: F30.18, df 8,108, PB0.001; 48 h: F 255.23, df8,108, PB0.001). The 72-h-old eggs did not suffer the effects of Biocontrol Science and Technology 171

100 24 48 72 90 a a a a a a ab a a a a a a 80 ab 70 bc cd 60 c

% 50 cd cd d 40 e e d e d 30 d d

20

10

0 0 2.5 5 7.5 10 12.5 15 17.5 20 Doses (Gy)

Figure 1. Percent hatching in 24-, 48- and 72-h-old Anastrepha ludens eggs subjected to different doses of radiation.

irradiation, with percent egg hatch ranging from 72 to 82% (n13, F0.993, a 0.05). The following evaluations were performed using 72-h-old eggs.

Effects of irradiation on 72-h-old A. ludens eggs The averages of hatching percent, larval yield, pupal weight and fly emergence are shown in Table 1. The hatching percent and volume of larvae yielded were not significantly different for the doses of radiation applied (hatching percent: F0.558, df12,117, P0.871; volume of larvae: F0.571, df12, 102, P0.860).

Table 1. Means (9SE) of percent hatching, larval yield, pupal weight and fly emergence from 72-h-old Anastrepha ludens eggs irradiated at different doses.

Percentage (%) flies emerged Downloaded By: [Hendrichs, Jorge] At: 15:52 6 November 2009 Doses Hatching Larval yield (Gy) percentage (%) (mL) Weight of pupa (mg) Females Males

0 88.8 9 2.0 a 105.3 9 12.15 a 19 9 0.05 ab 41.1 91.44 a 38.6 9 1.7 ab 2.5 88.9 9 2.1 a 90.2 9 14.49 a 20.2 9 0.04 a 40.1 91.22 a 42.6 9 1.2 a 5.0 86.2 9 1.8 a 105.5 9 12.65 a 19.8 9 0.04 ab 38.3 91.44 ab 39.6 9 1.3 ab 7.5 85.8 9 1.8 a 94.6 9 12.16 a 19.5 9 0.05 ab 36.3 91.59 abc 35.5 9 1.8 b 10.0 86.4 9 2.1 a 88.9 9 11.62 a 18.8 9 0.03 ab 34.99 0.85 bc 37.5 9 1.6 ab 12.5 86.4 9 2.1 a 92 9 11.41 a 18.1 9 0.04 abc 31.6 9 0.93 cd 29.1 9 1.8 c 15.0 85 9 2.0 a 90.7 9 10.29 a 18.3 9 0.06 abc 26.6 9 1.31 d 27.6 9 1.7 c 17.5 87.4 9 1.8 a 95.7 9 10.37 a 16.9 9 0.06 bcd 17.1 9 1.79 e 10.0 9 0.9 d 20.0 89 9 1.5 a 82.6 9 15.48 a 15.6 9 0.07 cd 5.0 9 0.66 f 4.4 9 0.6 de 22.5 85.3 9 1.4 a 77.4 9 9.13 a 14.8 9 0.07 de 1.4 9 0.29 f 1.0 9 0.2 e 25.0 86.1 9 1.8 a 89.3 9 8.46 a 11.8 9 0.07 f 0.2 9 0.07 f 0.0 9 0.0 e 27.5 87.6 9 1.5 a 82.7 9 11.87 a 12.6 99 0.07 ef 0.0 9 0. 00 f 0.0 9 0.0 e 30.0 87.6 9 1.6 a 76.3 9 10.48 a 10.4 9 0.09 f 0.0 9 0.00 f 0.0 9 0.0 e

*Means with different letters in each column indicate statistical differences. ANOVA, Tukey’s test (a 0.05). 172 J. Cancino et al.

100 90 80 70 60 50

hatched 40 Percent of eggs 30 12 13 14 15 16 17 20 18 19 20 10 21 adult age (d) 0 0 2.5 5 7.5 10 12.5 15 17.5 Doses (Gy)

Figure 2. Daily decrease in fertility of Anastrepha ludens flies emerged from 72-h-old eggs irradiated at different doses.

However, there were significant differences among the average pupal weights (F 29.24, df12, 385, PB0.0001) at different doses. The decrease in pupal weight progressed as the radiation dose increased. Adult emergence suffered with increasing radiation as well. At 27.5 Gy, flies did not emerge, even though the larvae had satisfactorily achieved pupation. The flies that did emerge after receiving different radiation doses showed decreasing fertility with increasing dose. In Figure 2, the fertility curves, representing the percentage of hatched eggs, show that the flies’ fertility began to decrease at 7.5 Gy. For example, at 10 Gy the hatching average was 54.7%, which was significantly lower than the 81.1% hatching rate obtained in the control group. Above 15 Gy the flies that emerged were sterile.

Exposure of irradiated eggs to F. arisanus In general, there was a high variability in the emergence data obtained. The only

Downloaded By: [Hendrichs, Jorge] At: 15:52 6 November 2009 common thread was that parasitoid emergence occurred in every sample. Although there were significant differences in the parasitoid emergence averages at different doses, there was no relationship between the decrease of emergence and higher doses. Similar variability was obtained in the sex-ratio data of parasitoids (female emergence: F3.340, df12, 117, P0.0003) (Table 2). The longevity presented as averages of live parasitoids at 20 days and fecundity in emerged parasitoids are shown in Table 3. With the exception of the 17.5, 22.5 and 27.5 Gy doses, the percentage of living females at day 20 was upwards of 40% or more. In males, the longevity obtained at 2.5, 15 and 25 Gy was below 15%. For the remainder of the group, longevity was higher. The long-rank analysis reported statistical difference between the longevity results (females: x244.27, df12, PB 0.0001; males: x230.80, df12, P0.0021). However, it did not show a tendency to reduce the longevity at higher radiation levels. The fecundity data were lower than one offspring/female/day at all radiation levels except 17.5 Gy, which had a rate of 1.2 offspring/female/day. Biocontrol Science and Technology 173

Table 2. Means (9SE) for emergence and sex-ratio of Fopius arisanus parasitoids using irradiated 72-h-old eggs as host.

No. of parasitoids emerged

Doses (Gy) Females Males Sex-ratio (female:male)

0 1.5 9 0.7 c 1.4 9 0.68 ef 1.1:1 2.5 5.8 9 1.8 abc 2.8 9 0.90 bcd 2.1:1 569 2.00 ab 2.9 9 0.92 bcd 2.1:1 7.5 5.2 9 2.42 abc 2.6 9 1.16 bcde 2.0:1 10 2.8 9 0.88 abc 2.4 9 0.98 bcde 1.2:1 12.5 6.5 9 2.11 a 4.7 9 1.95 a 1.4:1 15 4.4 9 1.24 abc 3.2 9 1.2 bc 1.4:1 17.5 3.19 0.3 abc 2.1 9 0.3 cde 2.0:1 20 4.1 9 2.68 abc 2.4 9 1.44 bcde 1.7:1 22.5 2 9 1.12 bc 1.9 9 0.88 de 1.1:1 25 1.6 9 0.86 bc 0.4 9 0.22 f 4.0:1 27.5 4.8 9 2.63 abc 0.3 9 1.6 f 1.6:1 30 5.0 9 2.91 abc 3.4 9 1.72 b 1.6:1

*Means followed by different letters in the columns are significantly different. ANOVA, Tukey?s test (a 0.05%).

Exposure of larvae hatched from irradiated eggs to D. longicaudata The average larval yield per treatment was statistically similar (F0.819, df12, 39, P0.630). There were no differences in percentages of male parasitoid emergence (males: F0.971, df12, 117, P0.480). However, for female emergence, there were minor statistical differences (females: F0.739, df12, 117, P0.710) (Table 4). The emergence of flies from unparasitized larvae decreased as the dose of radiation increased. Above 10 Gy, fly emergence dropped considerably. Moreover,

Table 3. Longevity and fecundity of Fopius arisanus parasitoids emerged from 72-h-old eggs treated with different radiation doses.

Live parasitoids at the 20th day (%) Downloaded By: [Hendrichs, Jorge] At: 15:52 6 November 2009 Doses (Gy) Females Males Offspring /female/day

0 60 53 0.45 2.5 70 13.3 0.22 5 43 20 0.20 7.5 40 40 0.12 10 63 47 0.34 12.5 43 27 0.44 15 43 13.3 0.09 17.5 27 27 1.21 20 50 20 0.95 22.5 27 27 0.99 25 67 13 0.74 27.5 27 6.7 0.31 30 60 73 0.50 174 J. Cancino et al.

Table 4. Mean (9SE) volume of larval yield and percentage of Diachasmimorpha long- icaudata parasitoids and flies emerged from larvae developed from 72-h-old eggs irradiated at different doses.

Parasitoids emergence (%) Fly emergence (%)

Doses Volume of larval (Gy) yield (mL) Females Males Females Males

0 125910.41 a 30.793.32 ab 16.593.92 a 10.692.55 a 9.893.32 abc 2.5 101.2917.60 a 30.994.15 ab 1994.15 a 8.391.77 abc 10.792.35 abc 5 117.5917.97 a 28.793.03 ab 11.991.45 a 7.991.50 abc 12.392.98 ab 7.5 13598.66 a 30.293.37 ab 15.492.64 a 7.591.59 abc 10.492.93 abc 10 107.5916.52 a 31.492.85 ab 16.592.85 a 9.291.75 a 12.992.89 a 12.5 105921.02 a 34.493.84 ab 17.991.98 a 5.691.31 bcd 7.591.61 abcd 15 115911.90 a 30.594.90 ab 17.993.53 a 5.591.03 cde 792.07 bcde 17.5 115911.90 a 2592.61 b 15.193.74 a 3.190.71 def 5.191.12 def 20 102.5910.31 a 35.294.51 ab 12.490.73 a 1.990.84 ef 3.391.33 ef 22.5 118.7917.12 a 37.694.19 a 19.494.20 a 1.390.66 f 1.690.86 ef 25 137.5911.09 a 36.593.94 a 12.791.48 a 0.090.00 f 0.090.00 f 27.5 123.7912.81 a 32.694.61 ab 11.390.89 a 0.090.00 f 0.090.00 f 30 97.594.79 a 32.994.90 ab 18.792.86 a 0.090.00 f 0.090.00 f

*Means followed by different letters in the columns are significantly different. ANOVA, Tukey’s test (a 0.005).

there were significant differences between the averages for emergence suppression in males and females. At 25 Gy, total suppression of fly emergence was obtained, while the emergence of parasitoids maintained similar values in comparison with the other doses. The longevity and fecundity of the emerged parasitoids did not appear to be affected by radiation dose (Table 5). These data were widely variable. A range of three to six male offspring/female/day was obtained without any relationship with levels of radiation. Statistic differences were found in the longevity results (females: x2 51.96, df 12, PB0.0001; males: x2 36.68, df 12, P0.0003), but they were not related to the increase in radiation dose. Downloaded By: [Hendrichs, Jorge] At: 15:52 6 November 2009 Discussion The irradiation of host eggs is an important new development in the mass rearing of fruit fly parasitoids. Previously, radiation had been successfully applied to larvae in the mass rearing of fruit fly parasitoids of larvae (Sivinski and Smittle 1990; Cancino et al. 2002a), but the irradiation of eggs improves the mass-rearing of larval parasitoids and makes the rearing of pure egg-parasitoid cohorts possible. The present study was inspired by Rigney (1989) who found that if a fruit fly egg is irradiated, the adult will not emerge. In general, insect eggs are one of the most sensitive stages to irradiation. Radiation effects are inversely proportional to the degree of differentiation. The rate of cell division in insect eggs is high, although during a brief period just before molting, cell division slows down. This could explain why the 72-h-old eggs were less sensitive to radiation. Previous research also suggests that the static stages are less sensitive to radiation (IAEA 1977). After 72 h of development, the A. ludens eggs could be considered ready to start hatching. This Biocontrol Science and Technology 175

Table 5. Mean longevity and fecundity of Diachasmimorpha longicaudata emerged from larvae developed from 72-h-old irradiated eggs at different doses.

Live parasitoids at 20th day (%)

Doses (Gy) Females Males Offspring/female/day

083873 2.5 55 50 4.8 570833 7.5 58 70 4.9 10 68 63 3.7 12.5 50 57 6.1 15 57 80 6.1 17.5 53 87 4.5 20 52 50 5.2 22.5 73 67 3.8 25 78 80 3.6 27.5 65 50 4.1 30 47 43 6

is an advantage for F. arisanus which is a parasitoid of older eggs and young first instar larvae of Tephritidae (Bautista and Harris 1996). In the case of F. arisanus, this is a particularly crucial improvement since it makes it possible to obtain pure cohorts of adult parasitoids for mass-release without any contamination by fertile flies. When C. capitata eggs are used as hosts, parasitoid emergence occurs at almost the same time as adult fly emergence making separation at this point particularly difficult. When A. ludens eggs are used as hosts, flies emerge earlier but the labor to separate the insects and sanitation problems caused by starved host adults still makes egg radiation an attractive alternative (Zenil et al. 2004). The mass rearing of D. longicaudata may also benefit from using irradiated eggs as host. Egg irradiation would require the exposure of less volume of host material and eggs are easier to gather and prepare for oviposition than larvae that must be Downloaded By: [Hendrichs, Jorge] At: 15:52 6 November 2009 removed from the diet. About 22,000 A. ludens eggs can be contained in a volume of 1 mL, while millions can fit into a litre. The irradiation of this volume of eggs requires a smaller and simpler irradiating device than is used for other biological control operations. Again, cages, oviposition units, and procedures for developing immature stages can be designed taking into consideration the emergence of only parasitoids. These advantages have already been implemented in the design of mass rearing procedures for parasitoids reared on irradiated host larvae (Sivinski and Smittle 1990; Cancino, Villalobos, and De la Torre 2002b). The irradiation of biological material at mass rearing facilities already using an irradiator is easily performed. This point is of considerable importance in many cases, since many parasitoid mass rearing facilities have been built away from fruit fly mass rearing facilities. For example, when the larvae are to be subjected to radiation, they are typically crowded into a small container where metabolic heat can collect and cause a rise in the temperature inside the container. Depending on the length of time, this increased temperature can lead 176 J. Cancino et al.

to a reduction in longevity and high mortality of parasitoid hosts 72 h after oviposition. In mass-rearing conditions, a mortality of hosts (dead larvae) greater than 10% brings about a significant reduction in the parasitoid emergence (Cancino et al. 2002b). When host eggs are irradiated, the risks of increasing host mortality due to such packaging procedures are easier to manage. Another significant benefit of using host egg irradiation in the mass rearing of fruit fly parasitoids will be evident in the packing and releasing components of the program. Suppressing fly emergence (from unparasitized hosts) gives new freedom in designing broader methodologies for packing and mass releasing. Recently, several new packing methods have been developed thanks to the use of irradiated larval hosts in the mass rearing of D. longicaudata. However, in the case of the packing and release of F. arisanus, little work has been done. Some methods have been previously designed for the packing of F. arisanus using Bactrocera dorsalis (Hendel) eggs, but since their development takes different lengths of time, separating emerging flies from emerging parasitoids causes fewer problems (Bautista, Harris, and Vargas 2001).

Acknowledgements We thank Edgar de Leo´n for his technical help. We are also grateful to M.C. Yeudiel Go´mez and his team for their help in the irradiation of eggs. This work was supported by the IAEA in cooperation with the Medfly Program SAGARPA under contract No. 10848.

References Bautista, R.C., and Harris, E. (1996), ‘Effect of Fruit Substrates of Parasitization of Tephritid Fruit Flies (Diptera) by the Parasitoid Biosteres arisanus (Hymenoptera: Braconidae)’, Environmental Entomology, 25, 470475. Bautista, R.C., Harris, E., and Vargas, R.I. (2001), ‘The Fruit Fly Parasitoid Fopius arisanus: Reproductive Attributes of Pre-Released Females and the Use of Added Sugar as Potential Food Supplement in the Field’, Entomologia Experimentalis et Applicata, 101, 247255. Bustos, M.E., Enkerlin, W., Toledo, J., Reyes, J., and Casimiro, A. (1992), ‘Irradiation of Mangoes as Quarantine Treatmeant’, Proceedings of the Final Research Coordination Meeting on Use of Irradiation as a Quarantine Treatment, Vienna: IAEA, pp. 7790. Downloaded By: [Hendrichs, Jorge] At: 15:52 6 November 2009 Cancino, J., and Ortiz, S. (2002), ‘Desarrollo de te´cnicas ba´sicas en la adecuacio´n de una cepa de Fopius arisanus (Sonan) (Hymenoptera: Braconidae) a cr´ıa masiva’,inXXV Congreso Nacional de Control Biolo´gico, eds. S.R. Baez and J.J. Juvera, Hermosillo, Son, Me´xico, pp. 121125. Cancino, J., Ruiz, L., Go´mez, Y., and Toledo, J. (2002a), ‘Irradiacio´n de larvas de Anastrepha ludens (Loew) (Diptera: Tephritidae) para inhibir la emergencia de moscas en la cr´ıa del parasitoide Diachasmimorpha longicaudata (Ashmead) (Hymenoptera: Braconidae)’, Folia Entomolo´gica Mexicana, 41, 195208. Cancino, J., Villalobos, P., and De la Torre, S. (2002b), ‘Changes in the Rearing Process to Improve the Quality of Mass Production of the Fruit Fly Parasitoid Diachasmimorpha longicaudata (Ashmead) (Hymenoptera: Braconidae)’, in Quality Control for Mass-Reared Arthropods, Proceedings of the eighth and ninth workshop of the Working Group on Quality Control of Mass-Reared Arthropods, eds. N. Leppla, K.A. Bloem and R. Luck, University of Florida, pp. 7482. Dom´ınguez, J., Castellanos, D., Herna´ndez, E., and Mart´ınez, E. (2000), ‘Me´todos de cr´ıa masiva de moscas de la fruta’, Memorias del XIII Curso Internacional sobre Moscas de la Fruta. Programa Moscamed DGSV-SAGARPA. Metapa de Dom´ınguez, Chiapas, Me´xico, pp. 399414. Biocontrol Science and Technology 177

FAO/IAEA/USDA. (2003), ‘Product Quality Control and Shipping Procedures for Sterile Mass-Reared Tephritid Fruit Flies’, Version 5.0, International Atomic Energy Agency, Vienna, Austria. Francis, B., Green, M., and Payne, C. (1993), Statistical System for Generalized Linear Interactive Modelling, Oxford, UK: Clarendon, Press. Harris, E.J., and Okamoto, R.Y. (1991), ‘A Method for Rearing Biosteres arisanus (Hymenoptera: Braconidae) in the Laboratory’, Journal of Economic Entomology, 84, 417422. IAEA (1977) Laboratory Training Manual on the Use of Isotopes and Radiation in Entomology, Technical report series No. 61, 2nd ed., Vienna, Austria, pp. 120. IAEA (1982), Training Manual on Food Technology Techniques, 2nd edition, Vienna, Austria, 205 pp. IAEA (2001), Gafchromic† Dosimetry System for SIT, Standard Operation Procedure, Joint FAO/IAEA Division of Nuclear Techniques in Food and Agriculture, pp. 42. Lawrence, P.O., Harris, E.J., and Bautista, R.C. (2000), ‘Development and Reproductive Biology of the Egg-Pupal Parasite, Fopius arisanus in Anastrepha suspensa, a New Tephritid Host’,inArea-wide Control of Fruit Flies and Other Pests, ed. T. Keng-Hong, Penerbit Universiti Sains Malasia, pp. 739748. Matrangolo, W.J.R., Nascimento, A.S., Carvalho, R.S., Melo, E.D., and De Jesu´s, M. (1998), ‘Parasitoides de moscas das frutas (Diptera: Tephritidae) asociados a fruteiras tropicais’, Anais de la Sociedade Entomologica do Brasil, 27 (4), 593603. Menezes, E., Sivinski, J., Holler, T., Aluja, M., Jero´nimo, F., and Ram´ırez, E. (1998), ‘Development of Coptera haywardi (Hymenoptera: Diapriidae) in Irradiated and Uni- rradiated Pupae of the Caribbean Fruit Fly and the Mediterranean Fruit Fly (Diptera: Tephritidae)’, Florida Entomologist, 81, 567570. Rigney, C.J. (1989), ‘Radiation-Disinfestation of Fresh Fruit’,inFruit Flies: Their Biology, Natural Enemies and Control, eds. A.S. Robinson and G. Hooper, Amsterdam: Elsiever, pp. 425434. Rodr´ıguez, A., Quenta, F., and Molina, P. (1997), Control integrado de las moscas de la fruta, Ministerio de Agricultura SENASA, Peru´, pp. 54. Sivinski, J., and Smittle, B. (1990), ‘Effect of Gamma Radiation on the Development of the Caribbean Fruit Fly Anastrepha suspensa and the Subsequent Development of its Parasite Diachasmimorpha longicaudata (Ashmead)’, Entomologia Experimentalis et Applicata, 55, 295297. Sivinski, J.M., Calkins, C.O, Baranowski, R., Harris, D., Brambila, J., D´ıaz, J., Burns, R.E., Holler, T., and Dodson, G. (1996), ‘Suppression of a Caribbean Fruit Fly (Anastrepha suspensa (Loew) Diptera: Tephritidae) Population through Augmented Releases of the Parasitoid Diachasmimorpha longicaudata (Ashmead) (Hymenoptera: Braconidae)’, Biolo- gical Control, 6, 177185.

Downloaded By: [Hendrichs, Jorge] At: 15:52 6 November 2009 Wong, T.T.Y., and Ramadan, M.M. (1992), ‘Mass Rearing Biology of Larval Parasitoids (Hymenoptera: Braconidae) of Tephritid Fruit Flies in Hawaii’,inAdvances in Insect Rearing for Research and Pest Management, eds. T. Anderson and N. Leppla, Boulder, CO: Westview Press, pp. 405476. Zavala, J.L., Fierro, M.M., Schwarz, A.J., Orozco, D.H., and Guerra, M. (1985), ‘Dosimetry Practice for the Irradiation of the Mediterranean Fruit Fly Ceratitis capitata (Wied)’,in High Dose Dosimetry in Industrial Processing, IAEA, pp. 2330. Zenil, M., Liedo, P., Williams, T., Valle, J., Cancino, J., and Montoya, P. (2004), ‘Reproductive Biology of Fopius arisanus (Hymenoptera: Braconidae) on Ceratitis capitata and Anastrepha spp. (Diptera: Tephritidae)’, Biological Control, 29, 169178. Biocontrol Science and Technology, Vol. 19, S1, 2009, 179Á191

RESEARCH ARTICLE Effects of gamma radiation on suitability of stored cereal pest eggs and the reproductive capability of the egg parasitoid Trichogramma evanescens (Trichogrammatidae: Hymenoptera) A.S. Tunc¸bilek*, U. Canpolat, and A. Ayvaz

Department of Biology, Erciyes University, Faculty of Arts & Sciences, 38039 Kayseri, Turkey

A study was conducted to determine parasitization suitability and preference of irradiated and untreated eggs of the Mediterranean flour moth (MFM), Ephestia kuehniella Zeller, and the Angoumois grain moth (AGM), Sitotroga cerealella (Olivier) for mass-production and augmentative releases of T. evanescens West- wood in cereal storage and processing facilities. Eggs of both species irradiated with 200 Gy could be effectively used for propagation of T. evanescens in the sterilized host eggs. The irradiated host eggs of E. kuehniella were markedly preferred over eggs of S. cerealella. A dose of 200 and 150 Gy prevented adult emergence of E. kuehniella and S. cerealella, respectively. Treatment of immature stages of the parasitoid inside the host eggs or treatment of adults of T. evanescens with low-level doses (ranging between 0 and 140 Gy) resulted in significant reduction in the number of parasitized eggs, adult emergence and progeny production (F1) as radiation dose increased. Our study showed that MFM and AGM eggs killed by gamma radiation could be used for the rearing and release of T. evanescens into commodity storages without any risk of increasing the pest population. Keywords: parasitization; Trichogramma evanescens; Ephestia kuehniella; Sitotroga cerealella; biological control; gamma radiation; irradiated host eggs; stored cereal pests management

Introduction

Downloaded By: [Hendrichs, Jorge] At: 15:52 6 November 2009 Moth species of the genus Ephestia, especially the Mediterranean flour moth (MFM), Ephestia kuehniella Zeller, as well as the Angoumois grain moth (AGM), Sitotroga cerealella (Olivier) are serious pests in cereal-based food processing facilities, stored maize and other cereals (Anonymous 1995a). They have three to four overlapping generations per year in Turkey (Yildirim, O¨ zbek, and Aslan 2001). Typically, control of these pests is undertaken by regular treatment of infested facility areas with pesticides such as malathion, dichlorvos and methyl bromide (Anonymous 1995b). Use of chemicals is very costly, as it requires the shut-down of the food processing factory and the interruption of the production process. Therefore, research is needed to find improved methods for control. Egg parasitoids are promising biological control agents because they eliminate the pest before it causes damage and they can be economically produced (Knipling 1992; Nurindah and Gordh 1999).

*Corresponding author. Email: [email protected]

First Published Online 21 April 2009 ISSN 0958-3157 print/ISSN 1360-0478 online # 2009 Taylor & Francis DOI: 10.1080/09583150902790269 http://www.informaworld.com 180 A.S. Tunc¸bilek et al.

Recent investigations suggested the use of inundative releases of parasitoid wasps of the genus Trichogramma in cereal stores. Trichogramma is often reared on eggs of the stored product pests, and therefore, releasing large numbers of Trichogramma into a warehouse might greatly suppress the pest population (Brower 1983). The use of Trichogramma requires repeated (inundative) releases at regular time intervals in order to ensure a long-term effect (Gwinner, Harnisch, and Mu¨ck 1996). Trichogramma turkestanica Meyer has been considered as a potential candidate for control of pyralid moths in stored grain and grain processing (Hansen and Jensen 2002). The first step in the evaluation of a proposed biological control agent is definition of the complete group of species on which the agent can survive and develop under controlled conditions (McEvoy 1996; Onstad and McManus 1996). If the agent is not monophagous when tested in a no choice situation, it is useful to assess host preference through choice tests with more than one host species present (Field and Darby 1991; Blossey, Schroeder, Hight, and Malecki 1994; Silva and Stouthamer 1999). Choice tests indicate if the target host is preferred over other physiologically acceptable hosts or if all acceptable hosts are equally suitable. Considerable technological advances, including the use of radiation, have been made in mass rearing of parasitoids and natural enemies. It has been reported that parasitization by egg parasitoids was increased in the eggs of irradiated lepidopteran pests (Marston and Ertle 1969; Mannion, Carpenter, and Gross 1995). Marston and Ertle (1969) tested the acceptability of irradiated moth eggs to Trichogramma minitum Rile, and reported that irradiated eggs were as suitable as non-irradiated eggs for parasitoid development. Indian meal moth Plodia interpunctella Hubner eggs killed by irradiation were used for rearing and release of Trichogramma pretiosum Riley into commodity storage facilities (Brower 1982). Irradiation can also be used to reproductively sterilize hosts to prevent the emergence of non-parasitized hosts or to prolong the development of host stages suitable for parasitization, thus facilitating the use of these hosts under mass rearing conditions. There appear to be significant opportunities for increased use of classical and augmentative biological control through nuclear techniques for production, shipping, and release of biological control agents. Moreover, Tillinger, Hoch, and Schopf (2004) showed that gamma radiation can be used as a tool to study interactions between the Downloaded By: [Hendrichs, Jorge] At: 15:52 6 November 2009 braconid endoparasitoid associated factors and its host larva. On the other hand, when the field performance of laboratory-reared parasitoids is a concern, very low doses of radiation may be useful in improving field performance of laboratory-reared parasitoids which may stimulate some physiological and behavioral processes in parasitoids. The aim of this study was to evaluate the effect of gamma radiation on the eggs of E. kuehniella and S. cerealella on their suitability as hosts of T. evanescens. The study also evaluated the effect of low dose gamma radiation on T. evanescens reproductive capability to increase progeny production of the parasitoid.

Materials and methods Host rearing E. kuehniella and S. cerealella were obtained from the Department of Plant Protection, Faculty of Agriculture of Ankara University and Adana Plant Protection Biocontrol Science and Technology 181

Research Institute, respectively. E. kuehniella was reared in a mixture of 1 kg wheat flour, 5% yeast and 30 g wheat germ (Marec, Kollarova, and Pavelka 1999). S. cerealella were reared on wheat grain. Throughout the rearing, cultures were kept in a rearing room at 27918C and 7095% RH and under a light regime of 14 h L:10 h D. To obtain eggs, large numbers of 1Á2-day-old adults of E. kuehniella and S. cerealella were collected from stock cultures and placed in plastic jars with screen bottoms. Eggs that fell through the screen were collected the following days and sifted to remove insect parts and frass, and placed in Petri dishes. Eggs were removed daily and exposed to parasitoids in glass tubes for 24 h.

Rearing of Trichogramma evanescens T. evanescens used in this experiment were initially obtained from Adana Plant Protection Research Institute. They were collected from Ostrinia nubilalis Hubner (Lep: Pyralidae) eggs in southern Turkey in 1999. In the laboratory, T. evanescens were mass-reared on E. kuehniella eggs for several generations. Throughout the rearing, cultures were kept in the rearing room at 24918C and 7095% RH, under a light regime of 14 h L:10 h D. Parasitoid cultures were started from a single female on E. kuehniella eggs and maintained in glass rearing vials (27.5 cm).

Sterility tests Large numbers of 1-day-old MFM and AMG eggs were placed in glass Petri dishes and irradiated in a calibrated 60Co irradiator (Therathronics 780C) with a source strength of ca 3811 Ci and a dose rate of ca. 1 Gy/min; dose rate was verified with Fricke dosimetry. The eggs were exposed to doses of 0, 50, 100, 150, 200, 250, 300, 350, 400, 450, 500 and 550 Gy, using 120 eggs at each dose (three replicates of 40). Immediately after treatment, the eggs were transferred to 300 mL jars each containing 100 g of medium and incubated in the conditioned chamber. An unirradiated control population was started similarly. The number of eggs hatched and adult emergence was noted. Data were analyzed using analysis of variance (ANOVA) and probit analysis was used to estimate SD50 and SD99 values (SPSS 1999). Downloaded By: [Hendrichs, Jorge] At: 15:52 6 November 2009 In the parasitization test, 1-day-old eggs obtained from irradiated E. kuehniella were glued on pieces of white cardboard (22.5 cm) and placed in individual tubes along with young female adult T. evanescens (five replicates of 50). All females were fed with honey, mated and had no previous contact with host eggs. A single female per tube was obtained by capturing a 24-h-old female from a group of females scattered on a white piece of paper. A single tube was placed over an adult female of medium size and the female was allowed to walk up the vial towards the light. Data collected on the longevity of T. evanescens adults emerged from irradiated host eggs were compared by KruskallÁWallis test.

No choice and choice experiments Á irradiated host eggs In the no choice experiment, 1-day-old MFM and AMG eggs were placed in glass tubes and irradiated with the doses mentioned above, using approximately 250 eggs at each dose level (five replicates of 50). The parasitoid to host ratio was high in these 182 A.S. Tunc¸bilek et al.

tests to ensure that most acceptable eggs would be parasitized (i.e., a female wasp may lay about 35Á40 eggs in each day). The egg cards were prepared using a number of moth eggs as described by Brower (1982). Equal number of eggs were counted and sprinkled on strips of lightweight cardboard (22.5 or 2.54 cm). The cards were glued with gum arabica, and the gum was allowed to dry for at least 1 h. Eggs were placed in tubes along with young female adult T. evanescens. All females had no previous contact with host eggs, were fed with honey and mated. Single females per tube were obtained by capturing a 24-h-old female from a group of females scattered on a white piece of paper. A single tube was placed over an adult female of medium size with open end and the female was allowed to walk up the vial towards the light. Newly emerged Trichogramma that climbed from emergence glass tubes into an experimental tube were used, ensuring that more than 90% were female, since females are more phototactic than males (McDougall and Mills 1997). The glass tube was covered tightly with plastic hardware cloth to prevent the wasp from escaping. After 24 h at 248C, 70% RH and a photoperiod of 14 h L:10 h D, the wasps were removed from the tube cards, and the parasitized eggs were incubated in a controlled environment. Each morning, the tubes were checked for the number of live and dead T. evanescens. To determine whether parasitoid eggs hatched, we counted the eggs that had turned black after 4Á7 days of incubation at 248C. The numbers of parasitized eggs, adult and female emergence were recorded as parameters. Data were analyzed using analysis of polynomial regression analysis, with dose of radiation (independent variable), the number parasitized eggs, adult emergence, and the number of females (dependent variables) as sources of variation (SPSS 1999). In the choice experiment, the procedures were the same as above with the following exceptions. Five categories of radiation doses ranging from 0 to 200 Gy were defined and the experiment was replicated 10 times (50 vs. 50 eggs of each host type in each replication). The following combinations were tested: 0 vs. 50, 50 vs. 0, 50 vs. 50, 150 vs. 150, and 200 vs. 200 Gy, E. kuehniella egg versus S. cerealella egg, respectively. In this test, 1-day-old irradiated and unirradiated eggs (50 eggs from each) were glued on pieces of white cardboard (22.5 cm) 5 mm apart and arranged crosswise.

Downloaded By: [Hendrichs, Jorge] At: 15:52 6 November 2009 This test was performed to determine the relative attractiveness of irradiated E. kuehniella and S. cerealella eggs for T. evanescens. The eggs of irradiated and unirradiated E. kuehniella and S. cerealella were offered simultaneously to single female of T. evanescens in glass tubes. A single female per tube was obtained by capturing a 24-h-old female as mentioned above. After 24 h, the wasps were discarded and the cards were incubated under controlled conditions. Mean parasitization of the corresponding E. kuehniella eggs vs. S. cerealella eggs was compared by independent samples t-test (SPSS 1999).

Irradiation of parasitoids Irradiated T. evanescens immature stage experiments Parasitized host eggs (1.5-h- and 5-day-old, stored at room temperature) were exposed to five different doses ranging from 0 to 20 Gy, using about 250 eggs in each dose (five replicates of 50). Data were analyzed by polynomial regression analysis, Biocontrol Science and Technology 183

with dose of radiation (independent variable), the number parasitized eggs, and the number of adult emergence (dependent variables) as sources of variation (SPSS 1999).

Irradiated T. evanescens adult experiments One-day-old adult T. evanescens were exposed to five different doses ranging from 0 to 140 Gy. After irradiation, untreated eggs of MFM were offered to single irradiated females, using about 250 host eggs in each dose (five replicates of 50). The numbers of parasitized eggs and adult emergence were recorded as parameters. Data were analyzed by polynomial regression analysis, with dose of radiation (indepen- dent variable), number parasitized eggs, and number of adult emergence (dependent variables) as sources of variation (SPSS 1999). Data from longevity of irradiated T. evanescens adults were compared by KruskallÁWallis test.

Irradiated T. evanescens adult and host egg experiments Host eggs were irradiated with 0, 100, 140 and 200 Gy and parasitized by irradiated T. evanescens that were exposed to doses from 0 to 4 Gy, using about 500 eggs in each dose (10 replicates of 50). Data were analyzed using a two-factor analysis of variance (ANOVA), with dose of radiation applied vs. host eggs and dose applied vs. adults parasitoids as factors and the number of parasitized eggs, adult emergence and female emergence as parasitization efficiency dependent variables (PROC ANOVA & PROC GLM) (SPSS 1999). All data were transformed to square root before statistical analysis; when significant differences occurred Tukey-HSD was applied as a means of separation. Back transformed data are presented in regression equations.

Results Sterility test Á irradiated MFM and AGM eggs There was a significant reduction in adult emergence with increasing radiation doses, F146.383; df4; PB0.001 for E. kuehniella and F160.968, df3; PB0.001 for

Downloaded By: [Hendrichs, Jorge] At: 15:52 6 November 2009 S. cerealella. A dose of 200 and 150 Gy prevented adult emergence of E. kuehniella and S. cerealella, respectively. Male-to-female ratios in the adult stage were skewed in favor of male moths with increasing doses (from 0.60:1 to 1.25:1; ß:à). Using probit analysis, estimates were obtained of the doses required for 50 and 99% egg sterility. The SD50 values were 110.9 and 37.1 Gy for E. kuehniella and S. cerealella, respectively, and the SD99 were 210.3 and 188.5 Gy for E. kuehniella and S. cerealella, respectively (x2 27.262; df2; PB0.001; x25.977; df2; P0.050). At least 75% of the Trichogramma females that emerged from irradiated eggs survived until the 8th day of adult life when given access to honey and an unlimited supply of E. kuehniella eggs (Figure 1). The mean female longevity was 4.6193.98, 5.7293.92, 6.593.72, 6.5393.66 and 5.5393.90 days for 0, 50, 100, 150 and 200 Gy, respectively. Longevity of T. evanescens adults from irradiated 1-day-old E. kuehniella eggs was not significantly influenced by increasing doses and was similar to that of the untreated control hosts (Figure 1). The KruskallÁWallis test showed that there was no statistically significant difference among the medians at the 95% 184 A.S. Tunc¸bilek et al.

0 Gy 50 Gy 100 Gy 150 Gy 200 Gy 11 10 9 8 7 6 5

Survival 4 3 2 1 0 1234567891011121314 123456789 Longevity (Day)

Figure 1. Longevity of female and male adults of T. evanescens originating from irradiated 1- day-old E. kuehniella eggs. The host eggs were treated with 0, 50, 100, 150 and 200 Gy.

confidence level (x2 14, df13, P0.374 for females; x28, df7, P0.333 for males).

No choice and choice experiments Á irradiated host eggs The T. evanescens parasitization curves in the no choice experiments are shown in Figure 2. The number of 1-day-old E. kuehniella eggs parasitized by T. evanescens was not affected by the dose of radiation given to host eggs (Y22.580.009X 0.000027X2, R2 0.020; P0.59). Similarly, adult parasitoid emergence as well as female parasitoid emergence were not significantly affected by irradiation (Y 21.620.00009XÁ 0.000015X2, R2 0.027; P0.49 and Y16.590.0048XÁ 0.000024X2; R2 0.02; P0.57, respectively). A different trend was observed for irradiated 1-day-old host eggs of S. cerealella eggs (Figure 2). Here, the number of parasitized eggs was significantly affected by the dose of radiation given to host eggs (Y27.210.0173X0.0000689X2, R2 0.20; P0.003). Likewise, adult and female parasitoid emergence were significantly affected by irradiation (Y21.780.01XÁ0.000063X2; R2 0.29; PB0.0001; Y 19.06Á 0.006X0.000034X2; R20.24; P0.0009, respectively). For E. kuehniella, irradiating eggs had no significant effect on their acceptability Downloaded By: [Hendrichs, Jorge] At: 15:52 6 November 2009 by T. evanescens. However, the acceptability of irradiated eggs of S. cerealella declined significantly as the dose of radiation increased. In choice experiments, when E. kuehniella and S. cerealella eggs had been exposed to different combinations of radiation (0 vs. 50, 50 vs. 0, 50 vs. 50, 150 vs. 150 and, 200 vs. 200 Gy for E. kuehniella eggs versus S. cerealella eggs, respectively), and then T. evanescens females were given a choice between E. kuehniella and S. cerealella eggs, there was no preference (tB0.001, P1.000).

Irradiation of parasitoids Irradiated T. evanescens immature stages experiments When the host eggs were irradiated after parasitization (1.5 h or 5 days post parasitization) by T. evanescens, the survival of parasitoids and adult emergence decreased with increasing doses (Figure 3). Survival of parasitoids differed significantly among doses when parasitized host eggs were treated with gamma Biocontrol Science and Technology 185

Number of parasitized eggs 50 E. kuehniella S. cerealella 40

30

20

10

0 0 50 100 150 200 250 300 350 400 450 500 550 Dose (Gy)

Number of adults emerged 50 E. kuehniella 45 S. cerealella 40 35 30 25 20 15 10 5 0 0 50 100 150 200 250 300 350 400 450 500 550 Dose (Gy)

Number of females emerged 50 E. kuehniella 40 S. cerealella

30 20 10 Downloaded By: [Hendrichs, Jorge] At: 15:52 6 November 2009 0 0 50 100 150 200 250 300 350 400 450 500 550 Dose (Gy)

Figure 2. Effect of radiation dose administered to 1-day-old E. kuehniella and S. cerealella eggs on parasitization, adult emergence and female emergence of T. evanescens.

radiation after 1.5 h (Y0.11669X25.2091X38.743; R20.9734; PB0.0001), but this difference disappeared when parasitized eggs were treated with gamma radiation after 5 days (Y0.0029X20.2349X36.063; R2 0.0351; P0.6749). Adult parasitoid emergence decreased significantly with increasing dose (Y 0.1731X2Á5.0589X33.937; R2 0.9194; PB0.0001 and Y0.056X2 0.168X 27.2; R2 0.6498; PB0.0001 for irradiation at 1.5 h and 5 days post parasitization, respectively). Female parasitoid emergence also decreased significantly with dose (Y0.1623X2 4.6137X29.274; R2 0.9082; PB0.0001 and Y0.0657X2 186 A.S. Tunc¸bilek et al.

Irradiated T. evanescens (1.5 h post parasitization) 50 45 Parasitization 40 Adult emergence 35 Female emergence 30 25 20 15 10 5 0 0 5 10 15 20 Dose (Gy)

Irradiated T. evanescens (5 d post parasitization) 80 Parasitization 70 60 Adult emergence 50 Female emergence 40 30 20 10 0 0 5 10 15 20 Dose (Gy)

Figure 3. Effect of radiation dose administered to the endoparasitic developmental stages of T. evanescens within host eggs on parasitization, adult emergence and female emergence of T. evanescens. Regression equations are given in the text.

0.4463.6137X18.514; R20.6204; PB0.0001 for irradiation at 1.5 h and 5 days post parasitization, respectively). Downloaded By: [Hendrichs, Jorge] At: 15:52 6 November 2009 Irradiated T. evanescens adult experiments The mean number of parasitized eggs, as well as adult emergence of parasitoids, obtained from irradiated T. evanescens adults decreased with increasing dose of radiation. There were no parasitized eggs at doses of 60 Gy and above (Figure 4). There were significant relationships between irradiation doses and number of parasitized eggs (Y0.016X2 1.3227X23.37; R2 0.7594; PB0.0001), adult parasitoid emergence (Y0.0126X21.0112X16.641; R20.7239; PB0.0001), and female parasitoid emergence (Y0.0102X2Á0.8134X13.395; R20.7287; PB 2 0.0001). In the F1 generation egg production, (Y0.0053X 1.0799X42.881; 2 R 08896; PB0.0001), adult parasitoid emergence from F1 generation (Y 0.0124X21.382X37.712; R20.8465; PB0.0001) and female parasitoid emer- gence (Y0.011X21.2168X31.772; R2 0.7977; PB0.0001) also decreased with increasing dose and, there was no progeny production at doses of 40 Gy and above. T. evanescens are more sensitive to gamma radiation than host eggs. All of the Biocontrol Science and Technology 187

Irradiated T. evanescens adults 35 Parasitization 30 Adult emergence 25 Female emergence 20 15 10 5 0 0 102030405060 Dose (Gy)

F1 generation from irradiated T. evanescens 50 Parasitization 40 Adult emergence Female emergence 30

20

10

0 0 102030405060 Dose (Gy)

Figure 4. Effect of radiation dose administered to the adult of T. evanescens on parasitization, adult emergence and female emergence of T. evanescens. Regression equations are given in the text.

T. evanescens adults that were irradiated with 0, 1, 2 and 3 Gy when 1-day-old survived until the 5th day when given access to honey; at 4 Gy this value was 80%. Longevity of T. evanescens adults irradiated with lower doses was not significantly influenced by increasing doses and was similar to that of the adults from untreated control hosts (Figure 5). The KruskallÁWallis test shows that there is no statistically significant difference among the medians at the 95% confidence level

Downloaded By: [Hendrichs, Jorge] At: 15:52 6 November 2009 (x214, df13, P0.374). However, mortality rates of adult wasps irradiated with

100 0 Gy 1 Gy 2 Gy 80 3 Gy 4 Gy 60

40 Survival (%) Survival

20

0 1 2 3 4 5 6 7 8 9 1011121314 Longevity (Day)

Figure 5. Longevity of T. evanescens adults when irradiated with low level gamma radiation of 0, 1, 2, 3, or 4 Gy. 188 A.S. Tunc¸bilek et al.

Parasitizes egg Progeny production 40 0 Gy 35 100 Gy 30 140 Gy 200 Gy 25

20

15 Mean number

10

5

0 0123 0123 Dose (Gy)

Figure 6. Effect of radiation dose administered to T. evanescens and to 1-day-old E. kuehniella host eggs on parasitization (number of parasitized eggs) and progeny production (F1)ofT. evanescens adults.

2 Gy were a little higher than those from control host eggs, but this differences was not statistically significant (P0.005).

Irradiated T. evanescens adult and host egg experiments When 1-day-old T. evanescens adults were irradiated, the two-way ANOVA analysis

showed that the mean number of parasitized host eggs and adult emergence (F1)of T. evanescens were significantly reduced (F56.535; df3; PB0.001; F52.407; df3; PB0.001, respectively). On the other hand, when E. kuehniella eggs were irradiated, the mean number of parasitized eggs was not affected, but progeny production of T. evanescens was significantly decreased by irradiation of host egg (F5.511; df3; PB0.001) (Figure 6).

Downloaded By: [Hendrichs, Jorge] At: 15:52 6 November 2009 Discussion We conducted a series of experiments to determine the acceptability and suitability of irradiated and unirradiated eggs of both E. kuehniella and S. cerealella to parasitization by T. evanescens under choice and no choice experimental design. In practice, unless all of the fertile host eggs were parasitized, which is unlikely, the eggs that hatched would increase the number of pest moth larvae present. This would be unacceptable, and one way to avoid this problem is to kill or sterilize the moth eggs before they are exposed to the T. evanescens adults (Brower 1982; Takada, Kawamura, and Tanaka 2000). The present results with both host species confirmed earlier work (Ayvaz and Tunc¸bilek 2006), which showed that doses of 200 Gy and above prevented adult emergence from irradiated E. kuehniella eggs. Similarly, studies on sterilization of E. kuehniella adults by gamma radiation showed that egg hatch was significantly reduced at 150 Gy (Marec et al. 1999). Irradiating eggs of the two species had no significant effect on their acceptability as hosts by T. evanescens. The test was carried out with 24-h-old females to evaluate Biocontrol Science and Technology 189

the performance of the wasps in respect of host acceptance, fecundity, emergence and longevity. These are the most important traits for the performance of mass-reared Trichogramma parasitoids (Mansfield and Mills 2004) and host suitability was shown to be strongly correlated with host acceptance (Pak and Van Lenteren 1984; Wackers, De Groot, Noldus, and Hassan 1987). Based on the regression analysis of non choice experiments, the irradiated host eggs of E. kuehniella were equally acceptable as control eggs. The choice experiments showed that the irradiated MFM and AGM eggs presented to female T. evanescens were also equally acceptable for oviposition and suitable for parasitoid development. The present results indicate that the tendency of females to attack irradiated host eggs was similar to that of unirradiated eggs. Thus, host fertility did not appear to play a role in host preference of T. evanescens. These findings are in agreement with Brower’s results (1983) on irradiated and unirradiated host eggs. Similar findings were made by Lewis and Young (1972) in Trichogramma spp. and by Makee (2006) in Trichogramma cacoeciae Marchal. Roriz, Oliveira, and Garcı´a (2006) found that when host preference was analyzed by offering eggs of two host species simulta- neously to a single female of T. cordubensis Vargas & Cabello, the mean number of parasitized eggs differed significantly, and the host species with heavier eggs were the most parasitized. Our results on egg hatching after irradiation corresponded with those of Cogburn, Tilton, and Brower (1973) on the eggs of almond moth, Cadra cautella Walker. Thus, eggs irradiated at 200 Gy could be effectively used for propagation of T. evanescens in the sterilized host and could be used in warehouses and mills, without any risk of increasing the pest population. In addition, irradiated moth eggs could be used for laboratory mass rearing programmes in order to avoid any problems posed by hatching host eggs. Data obtained from both irradiated immature and adult stages of T. evanescens experiments indicated that there was no stimulatory effect of irradiation on T. evanescens, rather it caused significant adverse effects on parasitoid competence with increasing doses. Tillinger et al. (2004) found that irradiation had no negative effect on the lifespan of Glyptapanteles liparidis Bouche (Braconidae), but when female wasps were irradiated with 48 Gy, oviposition was significantly Downloaded By: [Hendrichs, Jorge] At: 15:52 6 November 2009 reduced and only 10% of these eggs were viable. This might be due to the vulnerability of sensitive stages of the developing parasitoid within the host at the time of irradiation. Mechanisms underlying any stimulatory response induced by low dose irradiation are still unclear, but several hypotheses have been considered (Yamaoka, Edamatsu, Itoh, and Mori 1994; Cai, Satoh, Tohyama, and Cherian 1999). Earlier studies (Brower 1982; Prozell and Scho¨ller 1998; Scho¨ller and Hassan 2001) support the assumption that the inundative release of Trichogramma egg parasitoids may be practicable for the control of stored product moths. Comparing the results obtained for irradiated eggs from different host species, we found that the acceptability and suitability of irradiated eggs for parasitoid production is a strong argument for the development of biological control strategies for control of MFM and AGM under commodity storage conditions. 190 A.S. Tunc¸bilek et al.

Acknowledgements We acknowledge the support of the project by FAO/IAEA to enable us to carry out this work under Research Contract No: IAEA/TUR-10782 as part of a Coordinated Research Project and The Unit of Scientific Research Projects, Erciyes University (EU¨ BAP Á 02-012-09). We thank Dr G. Hoch and Dr Canhilal for comments on early drafts of the manuscript, Mrs S. O¨ ztemiz for supplying the T. evanescens used in the experiments, and Department of Radiation Oncology for using 60Co irradiator.

References Anonymous (1995a), Technical Instructions for Plant Protection Vol. I, Ministry of Agriculture and Rural Affairs. General Directorate of Protection and Control Ankara, p. 393. Anonymous (1995b), Technical Instructions for Plant Protection Vol. IV, Ministry of Agriculture and Rural Affairs, General Directorate of Protection and Control Ankara, p. 393. Ayvaz, A., and Tunc¸bilek, A.S. (2006), ‘Effects of Gamma Radiation on Life Stages of the Mediterranean Flour Moth, Ephestia kuehniella Zeller (Lepidoptera: Pyralidae)’, Journal of Pest Science, 79, 215Á222. Blossey, B., Schroeder, D., Hight, S.D., and Malecki, R.A. (1994), ‘Host Specificity and Environmental Impact of the Weevil Hylobius transversovittatus, a Biological Control Agent of Purple Loosestrife’, Weed Science, 42, 128Á133. Brower, J.H. (1982), ‘Parasitization of Irradiated Eggs from Irradiated Adults of the Indian Meal Moth (Lepidoptera: Pyralidae) by Trichogramma pretiosum (Hymenoptera: Tricho- grammatidae)’, Journal of Economic Entomology, 75, 939Á944. Brower, J.H. (1983), ‘Utilization of Stored-product Lepidoptera Eggs as Hosts by Tricho- gramma pretiosum Rilay (Hymenoptera: Trichogrammatidae)’, Journal of Kansas Entomo- logical Society, 56, 50Á54. Cai, L., Satoh, M., Tohyama, C., and Cherian, M.G. (1999), ‘Metallothionein in Radiation Exposure: Its Induction and Protective Role’, Toxicology, 132, 85Á98. Cogburn, R.R., Tilton, E.W., and Brower, J.H. (1973), ‘Almond Moth: Gamma Radiation Effect of the Life Stages’, Journal of Economic Entomology, 66, 745Á751. Field, R.P., and Darby, S.M. (1991), ‘Host Specificity of the Parasitoid, Sphecophaga vesparum (Curtis) (Hymenoptera: ), a Potential Biological Control Agent of the Social Wasps, Vespula germanica (Fabricius) and V. vulgaris (Linnaeus) (Hymenoptera: Vespidae) in Australia’, New Zealand Journal of Zoology, 18, 193Á197. Gwinner, J., Harnissh, R., and Mu¨ck, O. (1996), Manual of the Prevention of Post-harvest Grain Losses, by Post-Harvest Project Deutsche Gesellschaft fu¨r Technische Zusammenar- Downloaded By: [Hendrichs, Jorge] At: 15:52 6 November 2009 beit (GTZ) GmbH Postfach 5180, D-65726 Eschborn, Federal Republic of Germany, p. 338 Hansen, L.S., and Jensen, K.-M.V. (2002), ‘Effect of Temperature on Parasitism and Host- feeding of Trichogramma turkestanica (Hymenoptera: Trichogrammatidae) on Ephestia kuehniella (Lepidoptera: Pyralidae)’, Journal of Economic Entomology, 95, 50Á56. Knipling, E.F. (1992), Principle of Insect Parasitism Analyzed from New Perspectives: Practical Implications of Regulating Insect Populations by Biological Means, Agricultural Handbook No. 693, United States Department of Agriculture, Washington, DC. Lewis, W.J., and Young, J.R. (1972), ‘Parasitism by Trichogramma evanescens of Eggs from Tepa-sterilized and Normal Heliothis zea’, Journal of Economic Entomology, 65, 341Á348. Makee, H. (2006), ‘Effect of Host Egg Viability on Reproduction and Development of Trichogramma cacoeciae and T. principium (Hymenoptera: Trichogrammatidae)’, Biocontrol Science and Technology, 16, 195Á204. Mannion, C.M., Carpenter, J.E., and Gross, H.R. (1995), ‘Integration of Inherited Sterility and a Parasitoid, Archytus marmoratus (Diptera: Tachinidae), for Managing Heliocoverpa zea (Lepidoptera: Noctiidae)’, Environmental Entomology, 24, 1679Á1684. Mansfield, S., and Mills, N.J. (2004), ‘A Comparison of Methodologies for the Assessment of Host Preference of the Gregarious Egg Parasitoid Trichogramma platneri’, Biological Control, 29, 332Á340. Biocontrol Science and Technology 191

Marec, F., Kollarova, I., and Pavelka, J. (1999), ‘Radiation-induced Inherited Sterility Combined with a Genetic Sexing System in Ephestia kuehniella (Lepidoptera: Pyralidae)’, Annals of the Entomological Society of America, 92, 250Á259. Marston, N., and Ertle, L.R. (1969), ‘Host Age and Parasitism by Trichogramma minitum (Hymenoptera: Trichogrammatidae)’, Annals of the Entomological Society of America, 62, 1776Á1782. McDougall, S.J., and Mills, N.J. (1997), ‘The Influence of Hosts, Temperature and Food Sources on the Longevity of Trichogramma platneri’, Entomologia Experimentalis et Applicata, 83, 195Á203. McEvoy, P.B. (1996), ‘Host Specificity and ’, BioScience, 46, 401Á405. Nurindah, B.W.C., and Gordh, G. (1999), ‘Effects of Physiological Condition and Experience on Oviposition Behavior of Trichogramma australicum Girault (Hymenoptera: Tricho- grammatidae) on Eggs of Helicoverpa armigera Hu¨bner (Lepidoptera: Noctuidae)’, Australian Journal of Entomology, 38, 104Á114. Onstad, D.W., and McManus, M.L. (1996), ‘Risks of Host Range Expansion by Parasites of Insects’, BioScience, 46, 430Á435. Pak, G.A., and van Lenteren, J.C. (1984), ‘Selection of Candidate Trichogramma sp. Strain for Inundative Releases against Lepidopteran Pests of Cabbage in the Netherlands’, Mededelin- gen van de Faculteit Landbouwwetenschappen Rijksuniversiteit Gent, 49, 827Á837. Prozell, S., and Scho¨ller, M. (1998), ‘Insect Fauna of Bakery, Processing Organic Grain and Applying Trichogramma evanescens Westwood. Integrated Protection of Stored Products’, IOBC Bulletin, 21, 39Á44. Roriz, V., Oliveira, L., and Garc´ıa, P. (2006), ‘Host Suitability and Preference Studies of Trichogramma cordubensis (Hymenoptera: Trichogrammatidae)’, Biological Control, 36, 331Á336. Scho¨ller, M., and Hassan, S.A. (2001), ‘Comparative Biology and Life Tables of Tricho- gramma evanescens and Trichogramma cacoeciae with Ephestia elutella as Host at Four Constant Temperatures’, Entomologia Experimentalis et Applicata, 98, 35Á40. Silva, I.M.M., and Stouthamer, R. (1999), ‘Do Sympatric Trichogramma Species Parasitize the Pest Insect Helicoverpa armigera and the Beneficial Insect Chrysoperla carnea in Different Proportions?’, Entomologia Experimentalis et Applicata, 92, 101Á107. SPSS (1999), SPSS Version 10.0, 233 S. Wacker Drive, Chicago, IL: USA. Takada, Y., Kawamura, S., and Tanaka, T. (2000), ‘Biological Characteristics: Growth and Development of the Egg Parasitoid Trichogramma dendrolimi (Hymenoptera: Trichogram- matidae) on the Cabbage Armyworm Mamestra brassicae (Lepidoptera: Noctudiae)’, Applied Entomology and Zoology, 35, 369Á379. Tillinger, N., Hoch, G., and Schopf, A. (2004), ‘Effects of Parasitoid Associated Factors of the Endoparasitoid Glyptapanteles liparidis (Hymenoptera: Brachonidae)’, European Journal of Entomology, 101, 243Á249.

Downloaded By: [Hendrichs, Jorge] At: 15:52 6 November 2009 Yamaoka, K., Edamatsu, R., Itoh, T., and Mori, A. (1994), ‘Effects of Low-dose X-ray Irradiation on Biomembrane in Brain Cortex of Aged Rats’, Free Radical Biology & Medicine, 16, 529Á534. Yildirim, E., O¨ zbek, H., and Aslan, I. (2001), Stored Product Pests, Faculty of Agriculture of Atatu¨rk University, No: 191, Erzurum, Turkey. Wackers, F.L., De Groot, I.J.M., Noldus, L.P.J.J., and Hassan, S.A. (1987), ‘Measuring Host Preference of Trichogramma Egg Parasites: An Evaluation of Direct and Indirect Methods’, Mededelingen van de Faculteit Landbouwwetenschappen Rijksuniversiteit Gent, 52, 339Á348. Biocontrol Science and Technology, Vol. 19, S1, 2009, 193Á209

Rearing of five hymenopterous larval-prepupal (Braconidae, Figitidae) and three pupal (Diapriidae, Chalcidoidea, Eurytomidae) native parasitoids of the genus Anastrepha (Diptera: Tephritidae) on irradiated A. ludens larvae and pupae Jorge Cancinoa,Lı´a Ruı´za, John Sivinskib, Fredy O. Ga´lveza, and Martı´n Alujac*

aDesarrollo de Me´todos, Campan˜a Nacional Contra Moscas de la Fruta, Tapachula, Chiapas, Me´xico; bCenter for Medical, Agricultural and Veterinary Entomology, USDA-ARS, Gainesville, FL, USA; cInstituto de Ecologia, Xalapa, Veracruz, Me´xico

The aim of this study was to ascertain if eight species of native larval-prepupal and pupal Anastrepha (Diptera: Tephritidae) parasitoids which have been recently domesticated and colonized (Aluja et al. in press) could be reared on irradiated larvae and pupae, and if such was the case, determine the optimal irradiation dose so that only adult parasitoids (not flies) would emerge. The species considered were: Doryctobracon crawfordi, Utetes anastrephae, Opius hirtus (all larval- prepupal braconids), Aganaspis pelleranoi, Odontosema anastrephae (both lar- val-prepupal figitids), Coptera haywardi, Eurytoma sivinskii and Dirhinus sp. (diapriid, eurytomid and chalcidoid pupal parasitoids). Eight-day-old A. ludens larvae or 3-day-old A. ludens pupae were irradiated with 0, 5, 10, 15, 20, 25, 30, 35, 40, 50, 60 and 70 Gy under free oxygen and then subjected to parasitoid attack. Emergence of the unparasitized host was completely halted at 20Á25 Gy but such was not the case with the three braconid parasitoids that emerged even if subjected to doses as high as 70 Gy. In the case of the figitids, the emergence of the host and the parasitoids was completely halted at 20 and 25 Gy, respectively. Some parasitoid emergence was recorded at 5Á15 Gy but at this irradiation dose, fly adults also emerged rendering the fly/parasitoid separation procedures impractical. Finally, in the case of the pupal parasitoids, A. ludens adults emerged from unparasitized pupae irradiated at 15 Gy. Beyond this dose, only parasitoids Downloaded By: [Hendrichs, Jorge] At: 15:53 6 November 2009 emerged. With the exception of the figitid larval-prepupal parasitoids, irradiation did not negatively affect adult longevity or fecundity. Our results show that parasitoid mass rearing with irradiated hosts is technically feasible. Keywords: fruit fly parasitoids; mass rearing; host irradiation; Tephritidae; Braconidae; Figitidae; Diapriidae; Eurytomidae; Chalcidoidae

Introduction In the New World, some species of fruit flies in the genus Anastrepha (Diptera: Tephritidae) (e.g. A. grandis [Macquart], A. fraterculus [Wiedemann], A. obliqua [Macquart], A. ludens [Loew], A. serpentina [Wiedemann], A. suspensa [Loew]) represent important agricultural pests that also significantly hinder fruit exports (Aluja 1994). With increasing public resistance to widespread insecticide use (Clark,

*Corresponding author. Email: [email protected]

First Published Online 10 October 2008 ISSN 0958-3157 print/ISSN 1360-0478 online # 2009 Taylor & Francis DOI: 10.1080/09583150802377423 http://www.informaworld.com 194 J. Cancino et al.

Steck, and Weems 1996), regional efforts are underway attempting to combine the use of the sterile insect technique (SIT) and augmentative releases of parasitoids. For example, in Mexican mango and citrus growing regions (e.g. Nayarit, Sinaloa, Nuevo Leo´n), sterile A. obliqua and A. ludens adults are being released in conjunction with the exotic parasitoid, Diachasmimorpha longicaudata (Ashmead) (Anonymous 2003). Despite the fact that this parasitoid has been proven effective at significantly lowering A. suspensa and A. ludens populations when repeatedly released in large numbers (Sivinski et al. 1996; Montoya et al. 2000) and that it is easily and cheaply mass-reared (Montoya and Cancino 2004), there has been a recent upsurge in interest at determining the potential of native parasitoids which had been so far neglected in fruit fly biological control programs. Native parasitoids, given their long-term evolutionary interaction with their host, could prove quite effective at lowering fly populations under certain circumstances (e.g. Sivinski, Aluja, and Lo´pez 1997; Eitam, Sivinski, Holler, and Aluja 2004). For example, in fruit growing regions, officially declared as low fruit fly prevalence areas, a native parasitoid may be better suited at detecting and parasitizing the few larvae present. Furthermore, some authors have proposed that releasing large numbers of exotic parasitoids may be detrimental to native, non-target insects (e.g. Williamson 1996). In this sense, native parasitoids may represent a more environmentally friendly alternative. There are three fundamental prerequisites to the use of native parasitoids in Anastrepha biological control programs. The first is to obtain basic knowledge of their natural history, ecology and behavior, and significant progress in this field has been made over the past 10 years (Sivinski et al. 1997; Aluja, Lo´pez, and Sivinski 1998; Sivinski, Aluja, and Holler 1999; Sivinski, Vulinec, and Aluja 2001; Guille´n, Aluja, Equihua, and Sivinski 2002; Ovruski and Aluja 2002; Aluja et al. 2003; Eitam et al. 2004; Guimara˜es and Zucchi 2004; Ovruski, Schliserman, and Aluja 2004; Ovruski, Wharton, Schliserman, and Aluja 2005). The second one is related to their domestication and colonization. Recently, Eitam et al. (2004) described some rearing techniques useful in the initial stages of the colonization of D. areolatus in Florida. Related to the work being reported here, we have successfully domesticated and colonized D. areolatus, D. crawfordi, U. anastrephae, O. hirtus (all larval-prepupal braconids), A. pelleranoi (Bre`thes), O. anastrephae (both larval-prepupal figitids), Coptera haywardi (Oglobin), E. sivinskii and Dirhinus sp. at the Instituto de Ecologia, Downloaded By: [Hendrichs, Jorge] At: 15:53 6 November 2009 A.C. in Xalapa, Veracruz, Mexico (Aluja et al. in press). Thirdly, on top of having access to an established colony, parasitoids need to be mass-reared. Two efforts stand out in this respect. A fairly recent effort by Menezes et al. (1998) aimed at rearing the native pupal parasitoid C. haywardi in irradiated A. suspensa and Ceratitis capitata (Wiedemann) larvae. The other, is a yet unpublished but successful effort, directed at mass-rearing D. crawfordi in Mexico (L.R., unpublished data). Our aim here was to ascertain if eight of the species of native larval-prepupal and pupal Anastrepha (Diptera: Tephritidae) parasitoids recently domesticated and colonized at the Instituto de Ecologia, A.C., in Xalapa, Veracruz, Mexico (Aluja et al. 2008) could be reared on irradiated larvae and pupae and if such was the case, to determine the optimal irradiation dose. Our approach was based on the pioneering effort by Sivinski and Smittle (1990), who successfully tested the idea of mass rearing the exotic parasitoid D. longicaudata on irradiated A. suspensa larvae. We also wanted to develop a technique that would facilitate the use of excess mass reared larvae that sometimes are left over in mass rearing facilities with the idea Biocontrol Science and Technology 195

of finding an irradiation dose that would allow healthy adult parasitoids but not flies to emerge, as the latter would greatly facilitate handling procedures and reduce costs of production.

Materials and methods Study site All experiments were carried out under controlled environmental conditions in facilities and laboratories belonging to the Subdireccio´n de Desarrollo de Me´todos and the Programas MoscaMed/MoscaFrut, Campan˜a Nacional Contra Moscas de la Fruta in Metapa de Dom´ınguez, Chiapas, Me´xico. Mean temperature, relative humidity and illumination regime were as follows: 24928C, 60Á80% RH, and 12:12 h. Fly rearing and irradiation procedures took place in separate buildings.

Insects All parasitoids were reared on A. ludens larvae stemming from a laboratory strain that had been kept for over 300 generations (Dom´ınguez, Castellanos, Herna´ndez, and Mart´ınez 2000). Doryctobracon crawfordi, U. anastrephae, O. hirtus, A. pelleranoi, O. anastrephae, C. haywardi, and E. sivinskii colonies were obtained from the Instituto de Ecolog´ıa, A.C. in Xalapa, Veracruz, Mexico and reared for over 25 generations in our laboratories in Metapa de Dom´ınguez, Chiapas before being used for this study. Dirhinus sp. was discovered during a parasitoid survey in the Soconusco region (near the city of Tapachula, Chiapas) and subsequently domesticated and colonized in our laboratories.

Irradiation procedures Eight-day-old larvae and 3-day-old pupae of A. ludens were exposed to 0, 5, 10, 15, 20, 25, 30, 35, 40, 50, 60 and 70 Gy, respectively. Experiments were replicated 30 (braconids), 50 (figitids), 25 (C. haywardi), 35 (E. sivinskii) and 20 (Dirhinus sp.) times (replication level determined on the basis of result variability (e.g. high in the Downloaded By: [Hendrichs, Jorge] At: 15:53 6 November 2009 case of the two figitids, low in the case of Dirhinus sp.)). We used a Gammacell 220 irradiator (g radiation with a Co 60 source), applying a dose ranging between 2.5 and 3.0 Gy/min under free oxygen. Exposure times were determined by Fricke’s dosimetry (IAEA 1977). Before being exposed to radiation, larvae were removed from their rearing medium (artificial diet in a plastic washbowl) and rinsed with tap water until all diet residues had been washed away. In the case of pupae, we removed excess vermiculite (pupation medium) with the aid of a sieve.

Exposure of A. ludens larvae to parasitoids The method used to expose irradiated larvae or pupae to parasitism was tailored to the idiosyncrasies of the parasitoids. In the case of the braconids, 100 A. ludens larvae mixed with diet (same diet used for rearing them) were placed in a Petri dish that was covered with organza cloth kept in place with a rubber band. The parasitization unit was then placed in a Hawaii-type holding cage (272727-cm 196 J. Cancino et al.

wooden structure cage covered with 0.5-mm caliber mesh) (Wong, Ramadan, Herr, and McInnis 1992) into which 60 (30à and 30ß)5Á10-day-old parasitoids had been released. Exposure periods were 4, 6 and 8 h for D. crawfordi, O. hirtus and U. anastrephae, respectively. Given that not all species are equally adapted to the artificial rearing conditions, varying exposure times are required to, on the one hand avoid superparasitism (case of D. crawfordi) and on the other, secure minimally acceptable rates of parasitism (case of U. anastrephae). In the case of the two figitids that preferentially parasitize larvae in fallen fruit where they seek them out by penetrating the fruit, we did not cover the Petri dish to allow the female’s direct access to the larvae. In this case, 100 larvae were exposed to 100 adults (50à:50ß) inside a 303030-cm Plexiglass cage. Exposure periods were 4 and 6 h for A. pelleranoi and O. anastrephae, respectively. Finally, in the case of the three pupal parasitoids, 100 pupae were mixed with vermiculite after irradiation and placed in a Petri dish with a paper ‘roof’ to secure a darkened environment for the foraging females. In all cases (i.e. all three species), we released 100 parasitoids (50à:50ß) and allowed them to parasitize pupae over a 24-h period.

Parasitoid developmental times and emergence After exposure to parasitoid attack, larvae were again rinsed with tap water (to remove all diet residues) and placed in 48-cm plastic containers with moistened vermiculite as a pupation medium. Exposed pupae were also handled as described for larvae but were not rinsed with water. After 15 days had elapsed, the vermiculite was removed to facilitate emergence of fly and parasitoid adults, which varied among parasitoid species. After all insects had emerged, we counted the number of females and males, and transferred the insects into cages as described in what follows.

Determination of parasitoid longevity and fecundity After emergence, parasitoid adults were sorted out by species and irradiation treatment, and transferred to holding cages to determine their longevity and fecundity on a per treatment and replicate basis (three per species). Type of cage, parasitization unit and exposure period also varied according to species (details in Downloaded By: [Hendrichs, Jorge] At: 15:53 6 November 2009 section 2.4). Cohort size in each cage was 10à:5ß in all cases (i.e. all species). Survival was measured over a 30-day period from the moment of emergence. Fecundity was measured over a 10-day period starting at age 5 days by offering females a parasitization unit that contained non-irradiated larvae and that was replaced daily after the exposure period was covered. Exposed larvae and pupae were then handled as described in Parasitoid developmental times and emergence. Parasitoids had ad libitum access to water and honey throughout the test period.

Statistical analyses Mean number of flies and parasitoids that emerged, sex ratio, and number of offspring per female per day (i.e. fecundity; OFD), were subjected to a one-way ANOVA (each variable analyzed independently). Quadratic trends in OFD data were also ascertained but given extremely low r2 values (B0.05), results are not reported. To compare means, we used Bonferroni’s test (Snedecor and Cochran 1980). OFD Biocontrol Science and Technology 197

values were obtained by dividing the number of offspring by the number of live mothers per day. The proportion of living parasitoids per day (i.e. longevity) was analyzed by means of a log-rank test (Francis, Green, and Payne 1993).

Results Emergence patterns of irradiated and non-irradiated hosts Developmental times (egg to adult) varied sharply among parasitoid species: 15 days for U. anastrephae, O. hirtus, E. sivinskii,20daysforD. crawfordi, A. pelleranoi, O. anastrephae and Dirhinus sp., and 30 days for C. haywardi. Furthermore, we found that development of irradiated A. ludens larvae or pupae not subjected to parasitism by any of the eight parasitoid species under study here was completely halted at 25 Gy (Table 1). In the case of the braconid parasitoids and their host (exposed to parasitism), highly significant differences were found when comparing the effect of irradiation on emergence patterns of the host (A. ludens) but not the parasitoid (A. ludens,

F11 19.31, P0.0001, D. crawfordi, F11 0.6543, P0.781; A. ludens, F11 20.58, P0.0001, U. anastrephae, F11 0.7321, P0.7075; A. ludens, F11 15.74, P 0.0001, O. hirtus, F11 1.7138, P0.0708). Complete suppression of adult emergence for irradiated A. ludens larvae exposed to unsuccessful parasitism was achieved at doses of 20 Gy (Table 2). In the case of the parasitoids, over 30% emergence was recorded at doses as high as 70 Gy (Table 2). With respect to sex ratio, there were no statistically significant differences among any of the three

parasitoid species under study (D. crawfordi, F11 0.999, P0.447; U. anastrephae, F11 0.394, P0.957; O. hirtus, F11 0.6154, P0.815). Despite the latter, sex ratio was consistently skewed towards females. The two figitid species were much more susceptible to irradiation. As shown in Table 3, emergence was completely halted at 25 Gy, with a highly significant drop apparent at 20 Gy. At lower doses, even though emergence was observed in both

Table 1. Mean proportion (9SE) of A. ludens adults emerging from unparasitized larvae and pupae that were subjected to irradiation. Downloaded By: [Hendrichs, Jorge] At: 15:53 6 November 2009 Dose (Gy) Larva Pupa

0 82.5892.12 a 88.3792.52 a 5 81.5192.60 a 85.3092.46 a 10 38.5194.73b 2.8095.97b 15 5.9593.12c 0.3090.02c 20 0.1390.09c 0.1190.02c 25 0c 0c 30 0c 0c 35 0c 0c 40 0c 0c 50 0c 0c 60 0c 0c 70 0c 0c

Means within columns followed by the same letter are not significantly different (one-way ANOVA, followed by Bonferroni’s test). 198

Table 2. Mean number (9SE) of flies and parasitoids and sex-ratio of three species of Opiinae parasitoids that emerged from irradiated fruit fly larvae that were subjected to parasitism (values are means9SE).

Parasitoid species

D. crawfordi U. anastrephae O. hirtus

Mean number emerged Sex-ratio Mean number emerged Sex-ratio Mean number emerged Sex-ratio .Cancino J. Dose (Gy) Flies Parasitoids (à: ß) Flies Parasitoids (à: ß) Flies Parasitoids (à: ß)

0 20.37 1.91a 32.22 3.77a 2.70 0.44a 31.07 2.86a 40.30 3.23a 1.28 0.13a 36.31 3.37a 36.87 4.38a 1.00 0.16a

Downloaded By: [Hendrichs, Jorge] At: 15:53 6 November 2009 6 November 15:53 At: Jorge] [Hendrichs, By: Downloaded 9 9 9 9 9 9 9 9 9 5 10.6492.22b 27.6692.51a 2.1390.26a 27.9293.24a 42.9693.39a 1.0490.11a 29.5493.29a 42.0090.91a 1.0390.08a tal et 10 3.7591.20b 36.6192.90a 2.2390.45a 8.4292.07b 41.9694.03a 1.4190.42a 14.2092.03b 27.7391.64a 1.2290.16a

15 0.0790.07c 35.6893.02a 2.0190.17a 0.4690.26b 33.5792.97a 1.5190.47a 0.3690.30c 31.7691.70a 1.0790.12a . 20 0 c 36.6894.04a 1.6390.13a 0c 38.5193.00a 1.1390.15a 0 32.6091.77a 1.2790.16a 25 0 c 35.1392.66a 1.8590.15a 0c 35.6892.93a 1.1790.12a 0 33.3291.61a 1.1290.09a 30 0 c 31.2592.70a 2.1990.66a 0c 35.2792.89a 1.1390.17a 0 32.1691.61a 1.3190.12a 35 0 c 34.8292.97a 2.1790.23a 0c 38.8493.24a 1.1990.13a 0 32.4491.45a 1.0890.10a 40 0 c 35.3993.06a 2.9490.55a 0c 37.2893.87a 1.0190.14a 0 32.7291.50a 1.1790.13a 50 0 c 34.5093.19a 2.7190.60a 0c 39.6493.60a 1.3390.23a 0 30.4891.36a 1.0290.10a 60 0 c 33.8693.21a 1.9990.15a 0c 40.5093.42a 1.3190.12a 0 31.2491.59a 1.0690.13a 70 0 c 34.8793.09a 1.9290.29a 0c 40.8493.06a 1.2590.17a 0 30.1391.73a 1.0690.12a

Means within columns followed by the same letter are not significantly different (one-way ANOVA, followed by Bonferroni’s test). Table 3. Mean number (9SE) of flies and parasitoids and sex-ratio of two species of Figitidae parasitoids that emerged from irradiated fruit fly larvae that were subjected to parasitism (values are means9SE).

Parasitoid species

A. pelleranoi O. anastrephae Technology and Science Biocontrol

Mean number emerged Sex-ratio Mean number emerged

Dose (Gy) Flies Parasitoids (à: ß) Flies Parasitoids Mean no. a females Mean no. b males

0 7.2491.19a 22.6892.08a 2.6790.65a 32.1192.08a 35.2391.84a 35.0891.85a 0.1490.10a 5 8.8491.29a 23.0291.84a 2.3690.27ab 16.6791.81b 32.0091.87a 31.9391.87a 0.0790.04a Downloaded By: [Hendrichs, Jorge] At: 15:53 6 November 2009 6 November 15:53 At: Jorge] [Hendrichs, By: Downloaded 10 3.0990.57b 17.8691.87a 2.8290.49a 7.5491.41c 15.4291.73b 15.4291.73b 0a 15 1.6191.19b 5.1591.45b 1.0690.28bc 0.6590.31d 2.8291.10c 2.0490.69c 0.0290.02a 20 0 c 0.3490.15c 0.1090.05c 0 d 0.0290.02d 0.0290.02c 0a 25 0c 0d 0c 0d 0d 0c 0a 30 0c 0d 0c 0d 0d 0c 0a 35 0c 0d 0c 0d 0d 0c 0a 40 0c 0d 0c 0d 0d 0c 0a 50 0c 0d 0c 0d 0d 0c 0a 60 0c 0d 0c 0d 0d 0c 0a 70 0c 0d 0c 0d 0d 0c 0a

a,b Instead of sex-ratio, we provide actual emergence values for each sex to highlight fact that almost all emerged adults were females (apparently because we are dealing with a thelytokous strain). Means within columns followed by a common letter are not significantly different (one-way ANOVA, followed by Bonferroni’s test). 199 200

Table 4. Mean number (9SE) of flies and parasitoids and sex-ratio of three species of fruit fly pupal parasitoids that emerged from irradiated pupae that were subjected to parasitism (values are means9SE).

Parasitoid species

C. haywardi E. sivinskii Dirhinus sp.

Mean number emerged Sex-ratio Mean number emerged Sex-ratio Mean number emerged Sex-ratio .Cancino J. Doses (Gy) Flies Parasitoids (à: ß) Flies Parasitoids (à: ß) Flies Parasitoids (à: ß)

0 22.2191.22a 37.7892.07a 1.3590.10a 61.9092.46a 24.8891.38a 1.1490.07a 49.3292.74a 31.0593.62a 0.9290.11a Downloaded By: [Hendrichs, Jorge] At: 15:53 6 November 2009 6 November 15:53 At: Jorge] [Hendrichs, By: Downloaded 5 18.3292.00a 38.7091.39a 1.2690.06a 63.1092.46a 20.2791.54a 1.2990.14a 46.2492.88a 27.0593.46a 0.9490.12a tal et 10 8.9591.30b 38.5291.47a 1.3290.09a 14.5092.42b 20.0891.89a 1.6990.37a 17.0592.86b 28.8093.41a 0.8790.10a

15 2.4091.57c 36.0797.30ab 1.7490.13a 0.9490.47c 22.4891.55a 1.1490.08a 0.2090.2c 32.6293.48a 0.8890.10a . 20 0d 31.2591.99abc 1.7890.22a 0d 22.1991.74a 1.1090.08a 0c 29.7392.43a 1.0690.12a 25 0d 35.7291.50ab 1.4290.07a 0d 23.0092.20a 1.1890.15a 0c 32.4492.65a 0.9890.09a 30 0d 30.9091.4abcd 2.9290.56b 0d 24.6492.02a 1.0190.07a 0c 33.5292.91a 1.0590.10a 35 0d 34.0291.73ab 2.0790.15ab 0d 24.6991.88a 1.2890.19a 0c 33.3792.82a 1.5290.34a 40 0d 34.1991.47ab 1.799 0.19a 0d 24.2492.23a 1.1490.08a 0c 36.2092.85a 1.1290.10a 50 0d 29.3191.99bcd 2.269 0.2ab 0d 25.7492.22a 1.1990.13a 0c 31.3593.20a 1.1190.09a 60 0d 23.7392.04cd 1.899 0.1ab 0d 23.1891.97a 1.1490.10a 0c 31.4593.70a 0.9690.14a 70 0d 23.2591.18d 2.209 0.2ab 0d 24.3891.44a 1.1690.11a 0c 23.3592.78a 1.4690.28

Means within columns followed by the same letter are not significantly different (one-way ANOVA, followed by Bonferroni’s test). Biocontrol Science and Technology 201

species, a highly significant effect of irradiation was also detected, particularly in the case of O. anastrephae (A. ludens, F11 8.91, P0.0003, A. pelleranoi, F11 20.89, P0.0001; A. ludens, F11 47.15, P0.0001, O. anastrephae, F11 33.72, P 0.0001). Sex ratios were highly skewed towards females in both species, with statistically significant differences detected when adult emergence was recorded (B 25 Gy) (A. pelleranoi, F11 8.26, P0.001; O. anastrephae, F11 134.64, P 0.0001). Finally, in the case of the pupal parasitoids, emergence was observed at doses as high as 70 Gy, while the host exposed to unsuccessful parasitism (in this case irradiated in the pupal stage) totally ceased emerging at doses of 20 Gy (Table 4). The effect of irradiation dose was highly significant with respect to emergence of the host and in the case of C. haywardi (A. ludens,F11 19.97, P0.0001, C. haywardi, F11 9.44, P 0.0001; A. ludens, F11 128.4, P0.0001, E. sivinskii, F11 0.9568, P0.48; A. ludens, F11 38.89, P0.0001, Dirhinus sp., F11 1.148, P0.324). With respect to sex ratios, significant differences were also only detected in the case of C. haywardi (C. haywardi, F11 4.11, P0.0001; E. sivinskii, F11 0.439, P0.937; Dirhinus sp., F11 0.849, P0.590) (details in Table 4).

Fecundity and longevity of parasitoid offspring Mean fecundity of the braconid species studied was significantly affected by irradiation (D. crawfordi: F11 3.51, P0.0003, U. anastrephae: F11 2.32, P 0.013, O. hirtus: F11 3.99, PB0.0001) (Figure 1). With respect to longevity, there was no statistically significant effect of irradiation in any of the three species (D.

3 D. crawfordi O. hirtus a 2.5 ab ab ab U. anastrephae ab ab 2

Downloaded By: [Hendrichs, Jorge] At: 15:53 6 November 2009 ab ab 1.5 ab ab b b 1 a Offspring/female/day ab ab a a ab ab a ab ab ab 0.5 a b a b a a b a a a a b a 0 0 5 10 15 20 25 30 35 40 50 60 70 Irradiation dose (Gy)

Figure 1. Fecundity of D. crawfordi, O. hirtus and U. anastrephae (Braconidae: Opiinae) stemming from larvae irradiated at varying gamma radiation doses. The larvae offered to the adult parasitoids were not irradiated. 202 J. Cancino et al.

1 D. crawfordi

0.8

0.6

0.4

0.2

0 0 3 6 9 12 15 18 21 24 27 30 Age (day)

1 U. anastrephae

0.8

0.6

0.4

Proportion alive 0.2

0 0 3 6 9 12 15 18 21 24 27 30 Age (day)

1 O. hirtus 0.8 0.6 0.4 Downloaded By: [Hendrichs, Jorge] At: 15:53 6 November 2009 0.2 0 0 3 6 9 12 15 18 21 24 27 30 Age (day)

0 Gy 5 Gy 10 Gy 15 Gy 20 Gy 25 Gy 30 Gy 35 Gy 40 Gy 50 Gy 60 Gy 70 Gy

Figure 2. Longevity of D. crawfordi, O. hirtus and U. anastrephae (Braconidae: Opiinae) stemming from larvae irradiated at varying gamma radiation doses.

2 2 2 crawfordi, x11 9.23, P0.60, U. anastrephae, x11 45.97, P0.001, O. hirtus, x11  16.32, P0.129). Of the latter, D. crawfordi lived the longest (Figure 2). In the case of A. pelleranoi and O. anastrephae, no statistically significant influence of irradiation on fecundity was detected in the few cases where adequate Biocontrol Science and Technology 203

0.6 A. pelleranoi

0.5 O. anastrephae

0.4

0.3

0.2 Offspring/female/day 0.1

0 0 5 10 15 20 25 30 35 40 50 60 70 Irradiation dose (Gy)

Figure 3. Fecundity of A. pelleranoi and O. anastrephae (Figitidae: Eucoilinae) stemming from larvae irradiated at varying gamma radiation doses. The larvae offered to the adult parasitoids were not irradiated.

emergence was observed (up to 15 Gy) (A. pelleranoi, F11 0.726, P0.542; O. anastrephae, F11 0.142, P0.934; details in Figure 3). With respect to longevity, and particularly in the case of A. pelleranoi, irradiation had a marginally significant 2 2 effect (A. pelleranoi, x3 7.54, P0.056, O. anastrephae, x30.272, P0.965) (Figure 4). As for pupal parasitoids, fecundity was only influenced by irradiation in the case of C. haywardi and E. sivinskii (C. haywardi, F11 5.595, P0.0001; E. sivinskii, F11 3.824, P0.0001; Dirhinus sp., F11 0.26, P0.99). Remarkably, offspring was produced even at doses as high as 70 Gy (Figure 5). With respect to longevity, no statistically significant differences were detected when comparing the different

Downloaded By: [Hendrichs, Jorge] At: 15:53 6 November 2009 2 irradiation doses in all three species (C. haywardi, x11 4.58, P0.949; E. sivinskii, 2 2 x11 11.74, P0.383; Dirhinus sp., x11 12.84, P0.303). As can be seen in Figure 6, large numbers of adults were still alive after 30 days.

Discussion Several points of basic physiological and applied significance emerged from our study: (1) irradiating larvae or pupae to mass rear native Anastrepha larval-prepupal and pupal parasitoids appears technically feasible in all but two of the species under study here. With the exception of A. pelleranoi and O. anastrephae (both figitids), host emergence was completely halted at doses that did not negatively affect parasitoid emergence, fecundity or survival. This capacity to develop in irradiated hosts is paralleled in certain Old World species such as D. longicaudata (Sivinski and Smittle 1990; Cancino, Ruiz, Go´mez, and Toledo 2002). (2) Sex ratios were consistently (albeit not significantly) female biased, and did not vary when compared 204 J. Cancino et al.

1 A. pelleranoi

0.8

0.6

0.4

0.2

0 0 3 6 9 12151821242730 Age (day)

1 O. anastrephae Proportion alive 0.8

0.6

0.4

0.2

0 0 369 12151821242730 Age (day)

Downloaded By: [Hendrichs, Jorge] At: 15:53 6 November 2009 0 Gy 5 Gy 10 Gy 15 Gy

Figure 4. Longevity of A. pelleranoi and O. anastrephae (Figitidae: Eucoilinae) stemming from larvae irradiated at varying gamma radiation doses.

to the control. The latter adds significantly to the practical benefit of irradiation on native parasitoid mass rearing. (3) All three species of pupal parasitoids developed on irradiated hosts, although C. haywardi seemed the most sensitive to host irradiation, perhaps due to its unusual endoparasitic feeding habits and possible damage to host organs and physiology. (4) While C. haywardi is unable to develop in pupae resulting from irradiated larvae (Menezes et al. 1998), it was found to develop in irradiated pupae, suggesting some necessary early pupal development in the host. (5) Finally, it appears that A. ludens is highly susceptible to irradiation, as is A. obliqua (Toledo, Rull, Oropeza, Herna´ndez, and Liedo 2004), highlighting the urgent need to reexamine currently used irradiation doses that seem unnecessarily high. Biocontrol Science and Technology 205

C. haywardi 1 0.8 0.6 0.4 0.2 0 0 3 6 9 12 15 18 21 24 27 30 Age (day)

E. sivinskii 1 0.8 0.6 0.4 0.2 Proportion alive 0 0 3 6 9 12 15 18 21 24 27 30 Age (day) Dirhinus sp. 1 0.8 0.6 0.4 0.2 0 Downloaded By: [Hendrichs, Jorge] At: 15:53 6 November 2009 0 369 12151821242730 Age (day)

0 Gy 5 Gy 10 Gy 15 Gy 20 Gy 25 Gy 30 Gy 35 Gy 40 Gy 50 Gy 60 Gy 70 Gy

Figure 5. Longevity of Dirhinus sp., C. haywardi and E. sivinskii (Chalcidoidea, Diapriidae and Eurytomidae, respectively) stemming from pupae irradiated at varying gamma radiation doses. The pupae offered to the adult parasitoids were not irradiated.

The opiine braconids contribute a number of important fruit fly biological control agents (Wharton and Marsh 1978; Wharton and Gilstrap 1983; Ovruski, Aluja, Sivinski, and Wharton 2000). Our present experiments found that, like the Old World species D. longicaudata and D. kraussii (Fullaway), the New World species D. crawfordi, U. anastrephae and O. hirtus develop as well or better in irradiated host- 206 J. Cancino et al.

Dirhinus sp. 3 C. haywardi Eurytomidae 2.5

2

1.5

1

Offspring/female/day a a a a a b ab a b b b a a a b a 0.5 ab b b b b b b b 0 0 5 10 15 20 25 30 35 40 50 60 70 Irradiation dose (Gy)

Figure 6. Fecundity of Dirhinus sp., C. haywardi and E. sivinskii (Chalcidoidea, Diapriidae and Eurytomidae, respectively) stemming from pupae irradiated at varying gamma radiation doses.

larvae (see Sivinski and Smittle 1990). However, this capacity is not universal in the subfamily. Attempts to rear Psyttalia spp. on irradiated hosts have been unsuccessful (E. Harris, unpublished data). It is possible that irradiation prevents some important developmental process in the host that subsequently prevents parasitoid develop- ment. For example, Thomas and Hallman (2000) documented that irradiating late third instar A. ludens larvae at 20 Gy (gamma radiation), retarded protein metabolism and arrested development at the transition from cryptocepahlic to

Downloaded By: [Hendrichs, Jorge] At: 15:53 6 November 2009 phanerocephalic pupa. Evidence of required host development for maturation of the endoparasitic pupal parasitoid C. haywardi can be obtained by comparing the capacity of the insect to develop in pupae derived from irradiated larvae and pupae. In the first instance, C. haywardi is unable to develop (Sivinski et al. 1999), while development is completed if radiation is applied after pupation (our data here). In addition to retarding host development, irradiation might damage vital structures in the host required by the immature parasitoid. For example, radiation damages the nervous and endocrine systems of Anastepha suspensa (Loew) larvae (Nation, Smittle, Milne, and Dykstra 1995). None of the two figitid parasitoids emerged at doses above 20 Gy. These species have a longer developmental period, 20Á25 days, than braconids, and this relatively slow development could be a disadvantage when irradiated hosts eventually begin to decompose. In addition, apparently larvae start as endoparasitoids but move outside the host with increasing size. Given that parasitoid larvae may need to use the empty spaces between the host and the puparium (Ovruski 1994), unsatisfactory formation of the pupae might Biocontrol Science and Technology 207

result from irradiation. However, damage to the host need not be detrimental to the developing parasitoid. Increasing levels of irradiation could possibly suppress the immune system of the host and inhibit its ability of for example, encapsulate the parasitoid developing inside. In conclusion, the results obtained here represent a significant step forward in the use of native parasitoids in fruit fly biological control. Although their augmentative release has to date not been formally tested, the use of irradiated hosts may provide various advantages in other activities. For example, tests to determine movement ability with artificial traps and studies of foraging behavior using irradiated hosts may be carried out under field conditions without the risk of releasing pests.

Acknowledgements We thank two anonymous reviewers and the editor for helping us produce a more polished final product. Francisco D´ıaz-Fleischer (Subdireccio´n de Desarrollo de Me´todos, Campan˜a Nacional Contra Moscas de la Fruta [CNCMF]), Mariano Ordano, Juan Rull, Ricardo Ram´ırez and Larissa Guille´n (all Instituto de Ecolog´ıa, A.C. [INECOL]) also made many useful comments on an earlier draft. Javier Valle Mora (El Colegio de la Frontera Sur) and Francisco D´ıaz-Fleischer (CNCMF) made important suggestions on data analyses. We appreciate the technical support provided by Edelfo Pe´rez, Mario Pineda, Javier Robledo, Floriberto Pe´rez, Velizario Ribera and Marbel Monjaraz (all Subdireccio´n de Desarrollo de Me´todos, CNCMF). We are also grateful to Yeudiel Go´mez and his staff (Programa MoscaMed) for having irradiated all the larvae and pupae used in this study. Thanks are due to Nicoletta Righini and Alberto Anzures (both INECOL) for formatting and final preparation of this manuscript. This work was financed by the International Atomic Energy Agency (IAEA) through contract No. 10848, the Mexican Campan˜a Nacional Contra Moscas de la Fruta (Secretar´ıa de Agricultura, Ganader´ıa, Desarrollo Rural y Pesca Á Instituto Interamericano de Cooperacio´n para la Agricultura [SAGARPA-IICA]) and the Instituto de Ecolog´ıa, A.C. (INECOL). MA also acknowledges support from CONACyT through a Sabbatical Year Fellowship (Ref. 79449) and thanks Benno Graf and Jo¨rg Samietz (Forschungsanstalt Agroscope Changins-Wa¨denswil ACW), for providing ideal working conditions to finish writing this paper.

References Aluja, M. (1994), ‘Bionomics and Management of Anastrepha’, Annual Review of Entomology, 39, 155Á178. Downloaded By: [Hendrichs, Jorge] At: 15:53 6 November 2009 Aluja, M., Lo´pez, M., and Sivinski, J. (1998), ‘Ecological Evidence for Diapause in Four Native and One Exotic Species of Larval-Pupal Fruit Fly (Diptera: Tephritidae) Parasitoids in Tropical Environments’, Annals of the Entomological Society of America, 91, 821Á833. Aluja, M., Rull, J., Sivinski, J., Norrbom, A.L., Wharton, R.A., Mac´ıas-Ordo´n˜ez, R., D´ıaz- Fleischer, F., and Lo´pez, M. (2003), ‘Fruit Flies of the Genus Anastrepha (Diptera: Tephritidae) and Associated Native Parasitoids (Hymenoptera) in the Tropical Rainforest Biosphere Reserve of Montes Azules, Chiapas, Mexico’, Environmental Entomology, 32, 1377Á1385. Aluja, M., Sivinski, J., Ovruski, S., Guille´n, L., Cancino, J., Lo´pez, M., Torres-Anaya, A., Gallegos-Chan, G., and Ru´ız, L. ‘Colonization and Domestication of Seven Species of Native New World Hymenopterous Larval-Prepupal and Pupal Fruit Fly (Diptera: Tephritidae) Parasitoids’, Biocontrol Science and Technology (in press). Anonymous (2003), Informe Anual Planta Moscafrut. Metapa de Dom´ınguez, Chiapas, Me´xico. SAGARPA-DGSV, 56 p. Cancino, J., Ruiz, L., Go´mez, Y., and Toledo, J. (2002), ‘Irradiacio´n de larvas de Anastrepha ludens (Loew) (Diptera: Tephritidae) para inhibir la emergencia de moscas en la cr´ıa del parasitoide Diachasmimorpha longicaudata (Ashmead) (Hymenoptera: Braconidae)’, Folia Entomologica Mexicana, 41, 195Á208. 208 J. Cancino et al.

Clark, R., Steck, G., and Weems, H. (1996), ‘Detection, Quarantine and Eradication of Exotic Fruit Flies in Florida’,inPest Management in the Subtropics, eds. D. Rosen, F.D. Bennett and J.L. Capinera, Andover, UK: Intercept Ltd, pp. 29Á54,. Dom´ınguez, J., Castellanos, D., Herna´ndez, E., and Mart´ınez, E. (2000), ‘Me´todos de cr´ıa masiva de moscas de la fruta’, Memorias del XIII Curso Internacional sobre Moscas de la Fruta. CICMF, Metapa de Dom´ınguez, Chiapas, Me´xico, pp. 399Á414. Eitam, A., Sivinski, J., Holler, T., and Aluja, M. (2004), ‘Biogeography of Braconid Parasitoids of the Caribbean Fruit Fly, Anastrepha suspensa (Diptera: Tephritidae) in Florida’, Annals of the Entomological Society of America, 97, 928Á939. Francis, B., Green, M., and Payne, C. (1993), Statistical System for Generalized Linear Interactive Modelling, Oxford, UK: Clarendon Press. Guille´n, L., Aluja, M., Equihua, M., and Sivinski, J. (2002), ‘Performance of Two Fruit Fly (Diptera: Tephritidae) Pupal Parasitoids (Coptera haywardi [Hymenoptera: Diapriidae] and Pachycrepoideus vindemiae [Hymenoptera: Pteromalidae]) under Different Environmental Soil Conditions’, Biological Control, 23, 219Á227. Guimara˜es, J.A., and Zucchi, R.A. (2004), ‘Parasitism Behavior of Three Species of Eucoilinae (Hymenoptera: Cynipoidea: Figitidae) Fruit Fly Parasitoids (Diptera) in Brazil’, Neotro- pical Entomology, 33, 217Á224. IAEA (1977), ‘Laboratory Training Manual on the Use of Isotopes and Radiation in Entomology’, Technical Report Series No. 61, 2nd ed. Vienna, Austria, pp. 120. Menezes, E., Sivinski, J., Holler, T., Aluja, M., Jeronimo, F., and Ramirez, E. (1998), ‘Development of Coptera haywardi (Hymenoptera: Diapriidae) in Irradiated and Non- Irradiated Pupae of the Caribbean Fruit Fly and the Mediterranean Fruit Fly (Diptera: Tephritidae)’, Florida Entomologist, 81, 567Á570. Montoya, P., and Cancino, J. (2004), ‘Control biolo´gico por aumento en moscas de la fruta (Diptera: Tephritidae)’, Folia Entomologica Mexicana, 43, 257Á270. Montoya, P., Liedo, P., Benrey, B., Cancino, J., Barrera, J.F., Sivinski, J., and Aluja, M. (2000), ‘Biological Control of Anastrepha spp. (Diptera: Tephritidae) in Mango Orchards through Augmentative Releases of Diachasmimorpha longicaudata (Ashmead) (Hymenoptera: Braconidae)’, Biological Control, 18, 216Á224. Nation, J.L., Smittle, B.J., Milne, K., and Dykstra, T.M. (1995), ‘Influence of Irradiation on Development of Caribbean Fruit Fly (Diptera: Tephritidae) Larvae’, Annals of the Entomological Society America, 88, 348Á352. Ovruski, S.M. (1994), ‘Immature Stages of Aganaspis pelleranoi (Bre`thes) (Hymenoptera: Cynipoidea: Eucoilidae), a Parasitoid of Ceratitis capitata and Anastrepha spp. (Diptera: Tephritidae)’, Journal of Hymenoptera Research, 3, 233Á239. Ovruski, S.M., and Aluja, M. (2002), ‘Mating Behavior of Aganaspis pelleranoi (Bre`thes) (Hymenoptera: Figitidae, Eucoilinae), a Fruit Fly (Diptera: Tephritidae) Larval Parasitoid’, Journal of Insect Behavior, 15, 139Á151. Downloaded By: [Hendrichs, Jorge] At: 15:53 6 November 2009 Ovruski, S.M., Aluja, M., Sivinski, J., and Wharton, R. (2000), ‘Hymenopteran Parasitoids on Fruit-Infesting Tephritidae (Diptera) in Latin America and the Southern United States: Diversity, Distribution, Taxonomic Status and their Use in Fruit Fly Biological Control’, Integrated Pest Management Reviews,5,81Á107. Ovruski, S.M., Schliserman, P., and Aluja, M. (2004), ‘Indigenous Parasitoids (Hymenoptera) Attacking Anastrepha fraterculus and Ceratitis capitata (Diptera: Tephritidae) in Native and Exotic Host Plants in Northwestern Argentina’, Biological Control, 29, 43Á57. Ovruski, S.M., Wharton, R.A., Schliserman, P., and Aluja, M. (2005), ‘Abundance of Anastrepha fraterculus (Diptera: Tephritidae) and its Associated Native Parasitoids (Hymenoptera) in ‘‘Feral’’ Guavas Growing in the Endangered Northernmost Yungas Forests of Argentina with an Update on the Taxonomic Status of Opiine Parasitoids Previously Reported in this Country’, Environmental Entomology, 34, 807Á818. Sivinski, J., and Smittle, B. (1990), ‘Effect of Gamma Radiation on the Development of the Caribbean Fruit Fly Anastrepha suspensa, and the Subsequent Development of its Parasite Diachasmimorpha longicaudata (Ashmead)’, Entomologia Experimentalis et Applicata, 55, 295Á297. Sivinski, J., Calkins, C., Baranowski, R., Harris, D., Brambila, J., Diaz, J., Burns, R., Holler, T., and Dodson, G. (1996), ‘Suppression of a Caribbean Fruit Fly (Anastrepha suspensa Biocontrol Science and Technology 209

(Loew) Population through Releases of the Parasitoid Diachasmimorpha longicaudata (Ashmead) (Hymenoptera: Braconidae)’, Biological Control, 6, 177Á185. Sivinski, J., Aluja, M., and Lo´pez, M. (1997), ‘Spatial and Temporal Distributions of Parasitoids of Mexican Anastrepha Species (Diptera: Tephritidae) Within Canopies of Fruit Trees’, Annals of the Entomological Society of America, 90, 604Á618. Sivinski, J., Aluja, M., and Holler, T. (1999), ‘The Distributions of the Caribbean Fruit Fly, Anastrepha suspensa (Tephritidae) and its Parasitoids (Hymenoptera: Braconidae) Within the Canopies of Host Trees’, Florida Entomologist, 82, 72Á81. Sivinski, J., Vulinec, K., and Aluja, M. (2001), ‘Ovipositor Length in a Guild of Parasitoids (Hymenoptera: Braconidae) Attacking Anastrepha spp. Fruit Flies (Diptera: Tephritidae) in Southern Mexico’, Annals of the Entomological Society of America, 94, 886Á895. Snedecor, G.W., and Cochran, W.G. (1980), Statistical Methods, 7th ed., Ames, IA: The Iowa State University Press. Thomas, D.B., and Hallman, G.J. (2000), ‘Radiation-induced Pathology in the Metamorphosis of the Mexican Fruit Fly (Diptera: Tephritidae)’, Journal of Entomological Science, 35, 267Á 278. Toledo, J., Rull, J., Oropeza, A., Herna´ndez, E., and Liedo, P. (2004), ‘Irradiation of Anastrepha obliqua (Diptera: Tephritidae) Revisited: Optimizing Sterility Induction’, Journal of Economic Entomology, 97, 383Á389. Wharton, R.A., and Gilstrap, F.E. (1983), ‘Key to and Status of Opiinae Braconid (Hymenoptera) Parasitoids Used in Biological Control of Ceratitis and Dacus s.l. (Diptera: Tephritidae)’, Annals of the Entomological Society of America, 76, 721Á742. Wharton, R.A., and Marsh, P.M. (1978), ‘New World Opiinae (Hymenoptera: Braconidae) Parasitic on Tephritidae (Diptera)’, Journal of the Washington Academy of Sciences, 68, 147Á 167. Williamson, M. (1996), Biological Invasions, London: Chapman & Hall. Wong, T.T.Y., Ramadan, M.M., Herr, J.C., and McInnis, D.O. (1992), ‘Suppression of a Mediterranean Fruit Fly (Diptera: Tephritidae) Population with Concurrent Parasitoid and Sterile Fly Releases in Kula, Maui, Hawaii’, Journal of Economic Entomology, 85, 1671Á 1681. Downloaded By: [Hendrichs, Jorge] At: 15:53 6 November 2009 Biocontrol Science and Technology, Vol. 19, S1, 2009, 211Á224

RESEARCH ARTICLE Control of the olive fruit fly, Bactrocera oleae, (Diptera: Tephritidae) through mass trapping and mass releases of the parasitoid Psyttalia concolor (Hymenoptera: Braconidae) reared on irradiated Mediterranean fruit fly Bahriye Hepdurgun*, Tevfik Turanli, and Aydin Zu¨mreog˘lu

Plant Protection Research Institute Genc¸lik cad. No. 6, 35040 Bornova, I˙zmir, Turkey

Field studies were performed from 2002 to 2004 on Go¨kc¸eada Island, Turkey, to determine the effectiveness of releases of the larvalÁpupal parasitoid Psyttalia concolor Szepligeti against the olive fruit fly, Bactrocera oleae (Gmelin), alone and in combination with mass trapping, using EcoTraps†. For this, the parasitoid was reared on a factitious host, irradiated larvae of Ceratitis capitata (Wiedemann). Preparatory to making open-field augmentative releases, initial parasitoid releases were conducted throughout the 2001 season using confinement cages over branches bearing naturally infested olives into which parasitoids were introduced, using 1 : pair per 3 fruit oviposition punctures. Percent reduction in fly emergence due to parasitization in these cages was 26.9, 27.6, 18.0, and 24.7% from the first to the fourth olive fruit fly generations during the season, respectively. In 2002, open-field experiments were conducted in an experimental area (EA-1) containing 2500 olive trees. In this area, augmentative parasitoid releases and mass-trapping (MT) were combined, using 2000 EcoTraps. Following the first fruit oviposition punctures, parasitoids were released throughout the season, using ca. 26Á40 parasitoids per tree per occasion. Damage of olives was reduced from an average of 87.6% in the control areas to only 18.1% in areas receiving mass trapping plus parasitoids. In 2003, the experiments were conducted in two areas. In EA-1 (EA-1 in 2002), only parasitoid releases were made throughout the season, using ca. 27Á34 parasitoids per tree per occasion. In EA-2, which contained 2000 olive trees, parasitoid releases were combined with 2000 EcoTraps and the parasitoids were released using ca. 20Á30 parasitoids per tree per occasion throughout the Downloaded By: [Hendrichs, Jorge] At: 15:54 6 November 2009 season after occurrence of the first fruit oviposition punctures. Overall fruit damage rates of 10.6 and 10.1% were recorded in EA-1 (parasitoids only) and EA-2 (parasitoids  MT), respectively. Damage in the control area was 35.5%. In 2004, only parasitoid releases were conducted in EA-1. However, that year releases were begun early (in May) to attack the spring olive fly generation. Early season releases were made at ca. seven parasitoids per tree and late releases involved ca. 35 parasitoids per tree. Overall damage throughout the season was 12.2% in EA-1 vs. 37.9% in the control area. Our studies suggest that parasitoid releases are not enhanced by use of EcoTraps at the times and rates they were deployed. Despite the positive effects of both mass trapping and parasitoid releases, the reduction of damage by these means alone was not adequate to meet the requisite economic threshold of one to six larvae per fruit for table- and oil-varieties, respectively. Keywords: Bactrocera oleae; Ceratitis capitata; Psyttalia concolor; EcoTraps; pest management; gamma radiation

*Corresponding author. Email: [email protected]

ISSN 0958-3157 print/ISSN 1360-0478 online # 2009 Taylor & Francis DOI: 10.1080/09583150903056926 http://www.informaworld.com 212 B. Hepdurgun et al.

Introduction The olive fruit fly, Bactrocera oleae (Gmelin) (Diptera: Tepritidae), is the major pest of olives in Turkey, as it is in other Mediterranean countries (Haniotakis 2005 and references therein). In the Mediterranean region, it has been one of the most devastating olive pests for more than 2000 years. Infestation of olive fruit by the larvae causes premature fruit drop and reduces fruit quality for both table olives and for olive oil production (Michelakis and Neuenschwander 1983). Many studies and experiments have been carried out to suppress the pest (Haniotakis 2005). Among the methods used, chemical control measures are most widely applied, both as general cover sprays and as aerial and ground bait sprays. However, because of the detrimental effects of these chemicals on the environment and beneficial insects, an increasing effort is being made to develop biorational control strategies, including more biotechnical approaches. For this purpose, the sex pheromone of the olive fruit fly has been synthesized and used with traps developed to help monitor and control the pest (Haniotakis, Mazomenos, and Tumlinson 1977; Mazomenos, Haniotakis, Ioannou, Spanakis, and Kozirakis, 1983; Montiel 1986; Broumas and Haniotakis 1987; Cristofaro, Cristofaro, Tenaglia, Fenio, and Tronci 2007). In recent years, in order to protect the beneficial fauna in the ecosystem, mass trapping has become an important management tool (Haniotakis, Kozyrakis, and Fitsakis 1991; Montiel and Jones 2002; Ragoussis 2002; Mazomenos, Haniotakis, Ioannou, Spanakis, and Kozirakis 2002; Lentini, Delrio, and Foxi 2003; Tedeschini, Isufi, Uka, Bacaj, and Pfeffer 2003; Rizzi, Petacchi, and Guidotti 2005; Caleca, Rizzo, Battaglia, and Piccionello 2007; Iannotta, Pellegrino, Perri, Perri, and de Rose 2007). Biological controls using parasitoids are also being developed. In the Mediterra- nean and sub-Saharan Africa, the olive fruit fly is attacked by a number of parasitoid species including a braconid wasp, Psyttalia concolor Szepligeti, which was introduced into Italy from Tunisia in 1914 (Clausen 1978). This larvalÁpupal parasitoid was later introduced into France, Greece and other Mediterranean countries. Numerous studies have been conducted to develop improved rearing techniques for P. concolor to facilitate its use against B. oleae (Biliotti and Delanoue 1959; Jannone and Binaghi 1959; Monastero 1959; Monastero and Genduso 1962; Fenili and Pegazzano 1971; Bmetic 1973; Kapatos, Fletcher, Pappas, and Laudeho Downloaded By: [Hendrichs, Jorge] At: 15:54 6 November 2009 1977; Liaropoulos, Louskas, Canard, and Laudeho 1977; Kapatos and Fletcher 1984; Jimenez, Castillo, and Lorite 1990). This species is believed to be relatively ineffective as a classical biological control agent in Europe. One reason for its poor performance may be a lack of synchronization between the life cycles of the parasitoid and fly (Clausen 1978). However, because of increasing concerns for the environment and maintaining a more natural balance of plants and in managed ecosystems, P. concolor is still routinely used in the Mediterranean region for inoculative (Delrio, Lentini, and Satta 2003, 2005) and augmentative releases against the olive fly (Kimani-Njogu, Trostle, Wharton, Woolley, and Raspi 2001). After the first discovery of olive fruit fly in California in 1998, rearing of P. concolor was also initiated in Guatemala using the Mediterranean fruit fly (medfly) as a host and shipments of the adult parasitoids were made to California (Yokoyama, Rendon, and Sivinski 2008 and references therein). Continuing efforts are being made in California to use P. cf. concolor from Kenya in inoculative releases against B. oleae, importing parasitoid adults reared in Guatemala on C. capitata larvae (Yokoyama Biocontrol Science and Technology 213

et al. 2008). Efforts also are being made in Europe to use augmentative releases of P. concolor alone and in combination with other control tactics such as mass trapping (Liaropoulos, Mavraganis, Broumas, and Ragoussis 2005). To facilitate augmentative releases, rearing of P. concolor on C. capitata was achieved by rearing P. concolor for multiple generations on this factitious host to achieve a medfly-adapted strain (Loni and Canale 2005). Detailed studies on host suitability of C. capitata for P. concolor were conducted by Mohamed, Overholt, Lux, Wharton, and Eltoum (2007). Hepdurgun, Turanli, and Zu¨mreog˘lu (2009) demonstrated that P. concolor could be successfully reared on irradiated C. capitata larvae, which would allow field release of parasitized host puparia without fear of release of viable C. capitata adults from unparasitized pupae. In Turkey, efforts are now underway to test combinations of tactics and to apply control measures on an area wide basis in an attempt to find more effective and environmentally safe olive fly control strategies. In this study, we examined augmentative releases of P. concolor alone and in combination with mass trapping using EcoTraps.

Materials and methods Experimental area Field experiments were conducted on Go¨kc¸eada Island (Imbroz), which is ca. 30 km from continental Turkey and 300 km2 in size. It has ca. 120 000 productive olive trees. Although the olive oil variety ‘Megaritiki’ is dominant, the table olive variety ‘Gemlik’ also is present. Treatment orchards were located in the Zeytinli zone of the island, while the untreated control orchard was located ca. 2 km distant in the Baraj zone. Tree height in all of the orchards was 7Á10 m and trees were planted at an average density of 150 trees/ha. No olive fly control measures other than the experimental treatments were applied in the orchards from 2001 to 2004.

Parasitoid rearing The parasitoid P. concolor was acquired from the National Agricultural Research

Downloaded By: [Hendrichs, Jorge] At: 15:54 6 November 2009 Foundation (NAGREF), Greece, and was adapted to continuous rearing on medflies using procedures described in USDA manuals (Anonymous 1997, 1998). Further details on parasitoid rearing using irradiated C. capitata larvae are given in Hepdurgun et al. (2009).

Field-cage studies on the efficacy of Psyttalia concolor As a first step in preparing for eventual augmentative open-field releases of P. concolor to control olive fly, the efficacy of the parasitoid was initially investigated in cage tests, using confinement cages over fruit-bearing branches. The cages served to exclude wild olive flies and to confine introduced parasitoids. For this purpose, 80 mesh organdy cloth tree-branch cages 51 cm in height and 28 cm in diameter were used. All 80 cages were installed over fruit-bearing branches prior to observing the first oviposition punctures, on June 28, 2001, but 20 were not closed initially, so that they allowed access for oviposition by wild olive fly females. 214 B. Hepdurgun et al.

For the first generation, these 20 cages were closed after samples revealed that they contained sufficient punctured fruit as of July 11, 2001. Eleven days after observing the first punctures, mated individuals of P. concolor were released into 10 of these cages. The other 10 cages were maintained as controls. The number of parasitoids released was determined according to the number of punctures on the fruits inside the cages. The release ratio was one : pair of parasitoids per 3 punctures (potentially representing 3 host larvae). Small plastic tubes containing wet wicks and a honeyÁwater solution were placed into the branch-cages to provide food and water for the enclosed parasitoids. For the second and subsequent generations, 20 additional cages out of the total of 80 cages were opened to allow access for oviposition by the time the number of olive fly adults of second and subsequent generations were present, as evidenced by trap captures. The cages were closed again after 7Á9 days, when enough olive fruit fly punctures were seen on the fruits. Parasitoids were released into 10 cages 10Á12 days later. The other 10 cages were left as controls. Parasitoids were released into the cages right away in second generation tests because we noticed that larval development had begun. This process was also followed for the third and fourth olive fly generations, except that for the fourth generation, the test was replicated only 7 times (7 release and 7 control cages) due to inadvertent damage to some of the cages. Thus, for each olive fruit fly generation, 10 experimental and 10 control replicates/cages were employed (except that only 7 cages were used for the fourth generation).

Assessment of parasitoid efficacy in cages For each generation, the cages were opened 11Á20 days after parasitoid introduction and punctured fruits were collected and placed into labeled culture boxes with a pupation medium (sand) for any larvae completing development. Upon adult eclosion, olive fruit flies were counted along with unemerged puparia so that the percent successful development of B. oleae could be determined and the influence of parasitoid activity assessed.

Downloaded By: [Hendrichs, Jorge] At: 15:54 6 November 2009 Field tests Mass-trapping applications against Bactrocera oleae For mass trapping, EcoTraps† (Vioryl, S.A. Athens, Greece) were used. These traps were made of a 1520 cm green paper envelope with an internal plastic lining for water and air proofing. Each trap contained 70 g ammonium bicarbonate salt, a powerful food attractant for both sexes, and on its surface, 15 mg a.i. of deltamethrin especially formulated for protection of the active ingredient from natural UV light. A pheromone dispenser contained 80 mg of synthetic racemic 1,7-Dioxaspiro (5-5) undecane. In 2002, the experiment was conducted in Experimental Area (EA) 1, which had 2500 olive trees, and 2000 EcoTraps were distributed on June 26th at a density of ca. one trap per tree. In 2003, 2000 EcoTraps were hung on 2000 olive trees in EA-2 (across the road from EA-1) on 29th July. EcoTraps were placed approx. 2 m high and in the middle the canopy of the olive trees, in the shade, without coming in contact with leaves or branches. EcoTraps were hung in the trees before the emergence of the first generation of olive flies and before the olive fruits became susceptible to infestation. Traps were Biocontrol Science and Technology 215

applied once during the season. This application was intended to decrease the B. oleae population before releasing P. concolor.

Parasitoid releases Parasitoid releases were combined with mass-trapping in EA-1 in 2002 and in EA-2 in 2003, respectively, as shown in Table 1. In 2003 and 2004, in EA-1, only parasitoid releases were made; i.e., without mass-trapping.

Table 1. Number of Psyttalia concolor released in experimental areas throughout study.

Releasing No. of parasitoids released

Average no. Date parasitoids Year Area (dd/mo) Total released/tree

2002 Experimental Area 05.09 37,84 28,571 66,455 26.6 1 (combination of mass-trapping and parasitoid releases) 18.09 52,940 50,933 103,873 41.5 02.10 45,308 20,385 65,693 26.3 24.10 37,961 28,896 66,857 26.7 Total 174,093 128,785 302,878 121.1 2003 Experimental Area 23.09 36,023 32,718 68,741 27.5 1 (parasitoid releases only) 07.10 39,371 22,168 61,539 24.6 21.10 45,287 38,827 84,114 33.6 04.11 45,144 39,868 85,012 34.0 Total 140,825 113,581 254,406 101.7 Experimental Area 23.09 22,050 19,380 41,430 20.7 2 (combination of mass-trapping and Downloaded By: [Hendrichs, Jorge] At: 15:54 6 November 2009 parasitoid releases) 07.10 23,335 19,882 43,217 21.6 21.10 33,336 28,336 61,672 30.8 04.11 30,423 22,998 53,421 26.7 Total 109,144 90,596 199,740 99.9 2004 Experimental Area 26.05 10,681 7,478 18,159 7.3 1 (parasitoid releases only) 23.06 11,054 8,820 19,874 7.9 29.07 15,301 9,910 25,211 10.1 18.08 32,677 26,151 58,828 23.5 07.09 62,345 56,541 118,886 47.6 22.09 48,550 42,326 90,876 36.4 07.10 44,827 41,499 86,326 34.5 20.10 49,814 38,556 88,370 35.3 Total 275,249 231,281 506,530 202.6 216 B. Hepdurgun et al.

Parasitoid releases were initiated after the number of olive fly trap catches increased commensurate with discovery of fruit punctures in 2002 and 2003. However, releases were begun in May 2004 against the overwintering individuals and spring generation. Laboratory-reared P. concolor individuals were released into the groves after they had mated and matured. Before releases, the adult eclosion rate and sex ratio were determined by using eclosion grids. Thus, the numbers of parasitoids released in 2002Á2004 were estimates (Table 1). For assessment, before every release and at harvest, a total of 1000 randomly selected fruit were collected from the trees and the number of infested fruits was determined by checking for oviposition punctures. This was done in each area, core (orchard centre), buffer (15 rows beyond core), neighboring (10 rows beyond buffer) and control (2 km distant). Additionally, 500 fruits were collected from the ground. Infested fruits were held under laboratory conditions. After 15Á20 days, infested fruits were checked and the number of eclosed olive flies and P. concolor were recorded. In addition, McPhail and EcoTraps baited with pheromone were checked to observe the population trend of B. oleae in experimental areas and in the control. Mass-trapping and parasitoid releases were evaluated together where these two techniques were combined.

Population trend of Bactrocera oleae The population trend of B. oleae was observed by use of McPhail traps containing DAP 2% and yellow sticky traps with pheromone. Six traps of each were distributed as pairs in the experimental and control areas. The traps were checked weekly and the McPhail trap solutions were renewed after 3 weeks, along with the pheromone capsules (Vioryl, S.A. Athens, Greece).

Results Branch cage trials (2001) When parasitoid release cage and control cage results are compared, the reduction in olive fly survival as a result of parasitoid activity in the cages was 26.8, 27.6, 18.0 and Downloaded By: [Hendrichs, Jorge] At: 15:54 6 November 2009 24.7%, respectively, for the four generations (Table 2) The highest parasitization rate occurred in the third and forth generations, when adult emergence of B. oleae also was highest.

Open field trials In 2002, the average fruit damage throughout EA-1, which received parasitoids plus mass trapping, was only 14.1% in the core area, 22.1% in the buffer area, and 44.3% in the neighboring area (18.1% overall) vs. 87.6% in the control (Table 3). In 2003, the damage in EA-1, which received only parasitoids, averaged 10.4% (core), 10.9% (buffer) and 18.9% (neighboring). Damage in EA-2, which received parasitoids plus mass trapping, averaged 10.1% (core), 10.2% (buffer), and 18.9% in the same neighboring area vs. 35.5% damage in the control area (Table 4). The overall fruit damage rates for all regions of EA-1 and EA-2 were 10.6 and 10.1%, respectively. Biocontrol Science and Technology 217

Table 2. Overall results of the branch cage studies.

Mean no. B. oleae adult eclosion Reduction in B. oleae Generation punctured fruit from recovered survival associated with no. Treatment per cage puparia (%) parasitoid activity (%)

I Control 11.0 54.3 26.8 Parasitoids 12.1 27.4 released II Control 14.3 37.7 27.6 Parasitoids 14.8 10.1 released III Control 16.3 66.3 18.0 Parasitoids 17.3 48.3 released IV Control 17.4 86.5 24.7 Parasitoids 18.7 61.8 released

In 2004, since the parasitoid releases were made early (in May instead of September or October as in 2002 and 2003) because trapping results indicated the fly population was peaking early, beginning in August, as compared to September in earlier years. Fruit counting was initiated on August 18, when the first fruit punctures were seen. According to the assessment made at harvest, the average damage in the core, buffer and neighboring areas was 11.7, 12.7 and 18.3% (12.2% overall) vs. 37.9% in the control (Table 5).

Discussion The preliminary field cage studies conducted in 2001 clearly established that P. concolor reared on irradiated C. capitata larvae (Hepdurgun et al. 2009) were able to find and successfully attack B. oleae under our conditions. Consequently, we initiated larger-scale trials involving releases of this parasitoid alone and in combination with EcoTraps during 2002Á2003. Downloaded By: [Hendrichs, Jorge] At: 15:54 6 November 2009 In 2002, when mass trapping and P. concolor releases were combined in EA-1, it was found that by using a combination of parasitoids and traps, damage was reduced to a marked degree, culminating in only 14.1% damage in the core area at harvest vs. 87.6% damage in the untreated control area. In 2003, comparisons between use of only parasitoids vs. parasitoids plus mass trapping showed similar outcomes, with only about 10% damage in the core areas of each experimental area vs. 35.5% damage in the control area. This indicated that the use of EcoTraps might not be cost effective to use along with parasitoid releases. Tests performed in 2004 using only parasitoids again indicated that the parasitoid release strategy (without traps) could substantially reduce damage, in this case to around 10% vs. about 38% in the control area, reaffirming the earlier study. It is generally accepted that almost all alternate pest management methods, including mass trapping, depend upon a large experi- mental site (of at least 1000 trees) for success to be achieved. Mass-trapping experiments carried out in Greece (Mazomenos et al. 2002) showed that high population densities decreased the success of this technique. 218

Table 3. Effectiveness of combining Psyttalia concolor releases and mass-trapping (MT).

Parcel Damage (%)

Experimental Area 1 (Parasitoids  MT) .Hepdurgun B. Core Buffer Neighboring Control

Date Tree Ground Mean Tree Ground Mean Tree Ground Mean Tree Ground Mean Downloaded By: [Hendrichs, Jorge] At: 15:54 6 November 2009 6 November 15:54 At: Jorge] [Hendrichs, By: Downloaded 2002 05.09 0.5 ÁÁ 1.3 ÁÁ2.0 ÁÁ3.4 ÁÁ

18.09 1.1 1.2 1.1 2.1 2.0 2.1 5.0 4.9 4.9 8.0 10.9 9.5 al. et 02.10 1.2 1.5 1.4 2.9 3.3 3.1 6.1 5.1 5.6 12.3 18.7 15.5 24.10 11.3 14.4 12.9 14.8 25.3 20.0 20.1 26.5 23.3 36.2 44.5 40.4 Harvest 12.4 15.8 14.1 16.1 28.2 22.1 39.9 48.7 44.3 81.0 94.2 87.6 20.11 Table 4. Efficacy of releasing Psyttalia concolor alone and in combination with EcoTraps (MT).

DAMAGE (%) Technology and Science Biocontrol

Experimental Area 1 Á Parasitoids only Experimental Area 2 (Parasitoids  MT)

Parcel Core Buffer Neighboring Area Core Buffer Control

Date Tree Ground Mean Tree Ground Mean Tree Ground Mean Tree Ground Mean Tree Ground Mean Tree Ground Mean Downloaded By: [Hendrichs, Jorge] At: 15:54 6 November 2009 6 November 15:54 At: Jorge] [Hendrichs, By: Downloaded 2003 23.09 1.2 ÁÁ1.4 ÁÁ1.3 ÁÁ0.4 ÁÁ0.4 ÁÁ1.8 ÁÁ 07.10 1.5 1.9 1.7 2.0 1.8 1.9 2.3 1.4 1.9 0.7 1.0 0.9 0.7 1.0 0.9 3.8 3.0 3.4 21.10 4.3 2.2 3.2 5.1 3.2 4.1 6.2 4.6 5.4 1.2 5.0 3.1 1.4 3.8 2.6 7.2 7.4 7.3 04.11 9.1 10.3 9.7 9.8 8.9 9.3 13.6 10.9 12.3 8.7 10.8 9.7 9.2 10.2 9.7 22.7 28.4 25.6 Harvest 9.5 11.2 10.4 10.6 11.2 10.9 18.2 19.6 18.9 9.2 10.9 10.1 9.6 10.8 10.2 31.5 39.6 35.5 19.11 219 220

Table 5. Efficacy of releasing Psyttalia concolor alone.

Damage (%)

Experimental Area 1 (parasitoid release area) .Hepdurgun B. Parcel Core Buffer Neighboring Control

Date Tree Ground Mean Tree Ground Mean Tree Ground Mean Tree Ground Mean Downloaded By: [Hendrichs, Jorge] At: 15:54 6 November 2009 6 November 15:54 At: Jorge] [Hendrichs, By: Downloaded 2004 18.08 0.1 ÁÁ0.1 ÁÁ0.1 ÁÁ0.1 0 Á

07.09 0.8 0.8 1.4 1.1 1.6 1.4 1.8 1.9 1.9 2.2 2.0 2.1 al. et 23.09 1.4 1.2 1.3 1.9 1.7 1.8 2.2 2.6 2.4 5.6 8.2 6.9 07.10 2.0 2.8 2.4 2.2 3.1 2.7 3.2 5.4 4.3 8.8 11.4 10.1 20.10 5.2 6.4 5.3 5.8 6.9 6.4 12.3 10.8 11.6 28.0 32.1 30.1 Harvest 10.1 13.2 11.7 10.6 14.8 12.7 17.8 18.6 18.3 36.4 39.3 37.9 03.11 Biocontrol Science and Technology 221

Accordingly, it was stated that in such circumstances, additional control methods would be needed to achieve an acceptable result (Mazomenos et al. 2002). Similarly, Liaropoulos, Mavraganis, Broumas, and Ragoussis (2002) performed an EcoTrap trial in an olive orchard containing 150 trees, using 1 trap per tree and released a total of 17,000 P. concolor pupae (113,333/tree) on two different dates in October 2002. Despite this, damage rates were 78.4 and 68.1% in control and experimental areas, respectively because of the high population density. Because the olive fly population density generally is low from early May to late August in our test area, the parasitoids released within this period suppressed the olive fly population fairly effectively, but still did not achieve the local economic threshold of about 1% for table varieties or 6Á8% for oil varieties. Conclusions drawn from our experiments may be outlined as follows: in Go¨kc¸eada Island, since the growers do not apply any control measures due to the ecological/organic farming techniques undertaken, olive fly populations typically exceed the economic threshold, which is considered to be only 1% damage for table variety olives and 6Á8% for oil production olives. Although the overall results obtained were encouraging in experimental areas where either parasitoids and mass trapping were used together, or only parasitoid releases were employed, damage rates below the generally accepted economic threshold level were not achieved. This illustrates the importance of an area-wide application strategy in the control of pests like the olive fruit fly, which has high dispersal capabilities and a high reproduction potential. Additional biotechnical tools may need to be used to achieve adequate control. By way of example, in Hawaii, the related species Psyttalia fletcheri (Silvestri) was used along with sterile flies to suppress the melon fly, Bactrocera curcurbitae (Coquillet) (McInnis et al. 2004; Vargas et al. 2004). More recently, inclusion of sanitation (Klungness et al. 2005), reduced-risk protein bait sprays, and male lure treatments also have been integrated into effective IPM systems for fruit flies in Hawaii (Vargas et al. 2001; Prokopy et al. 2003; Mau, Jang, and Vargas 2007; Vargas, Mau, Jang, Faust, and Wong 2008). GF-120 NF Naturalyte Fruit Fly BaitTM (Peck and McQuate 2000; Vargas et al. 2001; Prokopy et al. 2003), and male lures have been used on organically certified farms in Hawaii (Vargas et al. 2008). To achieve the desired degree of economic control of the olive fly, it may be possible to utilize additional tools such as those used synergistically in Hawaii to complement Downloaded By: [Hendrichs, Jorge] At: 15:54 6 November 2009 use of parasitoids and traps.

Acknowledgements The authors would like to thank the International Atomic Energy Agency, Vienna, Austria for their support of the project that is number 10783/TUR and the Vioryl firm (Athens, Greece) to support the EcoTraps during the study.

References Anonymous (1997), Procedural Manual for Mass Rearing Six Species of Tephritid Fruit Fly Parasitoids, USDA ARS, Honolulu, HI. Anonymous (1998), Mass Rearing Production Manual ‘Fruit Fly Parasitoids, Aurora Rearing Facility, USDA APHIS PPQ, Guatemala. Biliotti, E., and Delanoue, P. (1959), ‘Contribution a l’etude Biologique d’Opius concolor Szepl. (Hym. Braconidae) en Elevage de Laboratoire’, Entomophaga,4,7Á14. 222 B. Hepdurgun et al.

Bmetic, D. (1973), ‘Artifical Rearing of the Wasp Opius concolor Szepl. and Possibility of its Employment to Control the Olive Fly (Dacus oleae Gmel.) in Dalmatia’s Olive Groves’, Final Rep. of the project No. E 30-ENT 8, Institute for Adriatic Agriculture and Karst Reclamation, Split, Yugoslavia, 193 pp. Broumas, T., and Haniotakis, G. (1987), ‘Further Studies on the Control of the Olive Fruit Fly by Mass-Trapping’, Proceedings of the II International Symposium on Fruit Flies, Crete, Sept. 1986, pp. 561Á565. Caleca, V., Rizzo, R., Battaglia, I., and Manuela, P.P. (2007), ‘Tests on the Effectiveness of Mass Trapping by Eco-Trap (Vyoril) in the Control of Bactrocera oleae (Gmelin) in Organic Farming’, Proceedings of the Working Group on Integrated Protection of Olive Crops, Florence, Italy: IOBC/WPRS Bulletin, 30, 139Á145. Clausen, C.P. (1978), ‘Introduced Parasitoids and Predators of Arthropod Pests and Weeds: A World Review’, U.S. Department of Agriculture Handbook 480. Washington, DC. Cristofaro, A de, Cristofaro, M., Tenaglia, F., Fenio, A., and Tronci, C. (2007), ‘Field Assessment of Different Combinations of Ammonia-Based Attractants and a Synthetic Female Sex Pheromone for the Monitoring and Control of the Olive Fruit Fly, Bactrocera oleae Gmel. (Diptera: Tephritidae) in Apulia, Southern Italy’, Proceedings of the Working Group on Integrated Protection of Olive Crops, Chania, Greece: IOBC/WPRS Bulletin, 28, 31Á40. Delrio, G., Lentini A., and Satta A. (2003), ‘Biological Control of Olive Fruit Fly with Inundative Releases of Opius concolor’,1st European Meeting of the IOBC/WPRS study group ‘Integrated Control in Olives’, 2003 May 29Á31; MAICh-Chania, Crete/Greece. Delrio, G., Lentini, A., and Satta, A. (2005), ‘Biological Control of Olive Fruit Fly Through Inoculative Releases of Opius concolor Szepl.’, Proceedings of the Working Group on Integrated Protection of Olive Crops, Florence, Italy: IOBC/WPRS Bulletin, 30, 53Á58. Fenili, G.A., and Pegazzano, F. (1971), ‘Contributo alla conoscenza dei parassiti del Dacus oleae Gmel.’, Ricerche eseguite in Toscana negli anni 1967 e 1968’, Redia, 52, 1Á29. Haniotakis, G.E. (2005), ‘Olive Pest Control: Present Status and Prospects’, Proceedings of the Working Group on Integrated Protection of Olive Crops, Chania, Greece: IOBC/WPRS Bulletin, 28, 1Á9. Haniotakis, G.E., Mazomenos, B.E., and Tumlinson, J.M. (1977), ‘A Sex Attractant of the Olive Fruit Fly Dacus oleae and its Biological Activity under Laboratory and Field Conditions’, Experimental and Applied Entomology, 21, 81Á87. Haniotakis, G., Kozyrakis, M., and Fitsakis, T. (1991), ‘An Effective Mass Trapping Method for the Control of Dacus oleae (Diptera: Tephritidae)’, Journal of Economic Entomology, 84, 564Á569. Hepdurgun, B., Turanli, T., and Zu¨mreog˘lu, A. (2009), ‘Parasitism Rate and Sex Ratio of Psyttalia (Opius) concolor Szepl. (Hymenoptera: Braconidae) Reared on Irradiated

Downloaded By: [Hendrichs, Jorge] At: 15:54 6 November 2009 Ceratitis capitata (Wiedemann) larvae (Diptera: Tephritidae)’, Biocontrol Science and Technology (this volume). Iannotta, N., Pellegrino, M., Perri, E., Perri, L., and De Rose, F. (2007), ‘Mass Trapping Experiments with Two Different ‘Attract and Kill’ Devices for Bactrocera oleae (Gmelin)’, in Proceedings of the Working Group on Integrated Protection of Olive Crops, Florence, Italy: IOBC/WPRS Bulletin 30, 135Á135. Jannone, G., and Binaghi, G. (1959), ‘Primi Esperimenti di Introduzione in Liguna di un Endofago della Mosca delle Oliye: Opius concolor Szepl. (O.siculus Mon.) (Hymenoptera: Braconidae) della Sicilia’, Bollettino del Laboratorio di Entomologia Agraria Portici, 17, 89Á123. Jimenez, A., Castillo, E., and Lorite, P. (1990), ‘Supervivancia del Himenoptero Braconido Opius concolor Szep. Parasito de Dacus oleae Gmelin en Olivares de Jaen’, Boldtin de Sanidad Vegetal Plagas, 16, 97Á103. Kapatos, E., Fletcher, B.S., Pappas, S., and Laudeho, Y. (1977), ‘The Release of Opius concolor var. siculus (Hym: Braconidae) Against the Spring Generation of Dacus oleae (Diptera.: Trypetidae) on Corfu’, Entomophaga, 22, 265Á270. Kapatos, E.T., and Fletcher, B.S. (1984), ‘The Phenology of Olive Fly, Dacus oleae Gmel. Diptera, Tephritidae) in Corfu’, Zeitschrift fu¨r Angewandte Entomologie, 97, 360Á370. Biocontrol Science and Technology 223

Kimani-Njogu, S.W., Trostle, M.K., Wharton, R.A., Woolley, J.B., and Raspi, A. (2001), ‘Biosystematics of the Psyttalia concolor Species Complex (Hymenoptera: Braconidae: Opinae): The Identity of Populations Attacking Ceratitis capitata (Diptera: Tephritidae) in Coffee in Kenya’, Biologica1 Control, 20, 167Á174. Klungness, L.M., Jang, E.B., Mau, R.F.L., Vargas, R.I., Sugano, J.S. and Fujitani, E. (2005), ‘New Approaches to Sanitation in a Cropping System Susceptible to Tephritid Fruit Flies (Diptera: Tephritidae) in Hawaii’, Journal of Applied Science and Environmental Manage- ment, 9, 5Á14. Lentini, A. Delrio, G. and Foxi, C. (2003), ‘Experiments for the Control of Olive Fly in Organic Agriculture’, First European Meeting of the IOBC/WPRS Study Group ‘Integrated Control in Olives’, May 29Á31; MAICh-Chania, Crete/Greece. Liaropoulos, C., Louskas, C., Canard, M., and Laudeho, Y. (1977), ‘Lachers d’Opius concolor (Hym:Braconidae) dans des Populations de Printemps de Dacus o/eae (Diptera: Trypeti- dae)’, Entomophaga, 22, 259Á264. Liaropoulos, C., Mavraganis, V.G., Broumas, T., and Ragoussis, N. (2002), ‘Field Tests on the Efficacy of Mass Trapping Combined with the Release of Opius concolor (Hymenoptera: Braconidae) Parasites for the Control of the Olive Fruit Fly Bactrocera o/eae (Diptera: Tephritidae)’, VIIth European Congress of Entomology, 2002 October 7Á13; Thessaloniki, Greece, 342 p. Liaropoulos, C., Mavraganis, V.G., Broumas, T., and Ragoussis, N. (2005), ‘Field Tests on the Combination of Mass Trapping with the Release of the Parasite Opius concolor (Hymenoptera: Braconidae), for the Control of the Olive Fruit Fly Bactrocera oleae (Diptera: Tephritidae)’,inProceedings of the Working Group on Integrated Protection of Olive Crops, Florence, Italy: IOBC/WPRS Bulletin, 30, 77Á81. Loni, A., and Canale, A. (2005), ‘Reproductive success of Psyttalia concolor (Szepligeti) Hymenoptera: Braconidae) on Different Hosts’, Frustula Entomologica, 28/29, 166Á171. Mau, R.F.L., Jang, E.B., and Vargas, R.I. (2007), ‘The Hawaii Fruit Fly Area-wide Fruit Fly Pest Management Programme: Influence of Partnership and a Good Education Pro- gramme’,inArea-Wide Control of Insect Pests: From Research to Field Implementation, eds. J.J. B. Vreysen, A. S. Robinson and J. Hendrichs, Dordrect, The Netherlands, Springer, pp. 671Á683. Mazomenos, B.E., Haniotakis, G.E., Ioannou, A., Spanakis, I., and Kozirakis, A., (1983), ‘Field Evaluation of the Olive Fruit Fly Pheromone Traps with Various Dispensers and Concentrations’,inFruit Flies of Economic Importance, ed. R. Cava1loro, Proceedings of the CEC/IOBC, International Symposium, 1982 November 16Á19; Athens Greece, 642 p. Mazomenos, B.E., Pantazi-Mazomenou, A., and Stefonu, D. (2002), ‘Attract and Kill of the Olive Fruit Fly Bactrocera o/eae in Greece as a part of an Integrated Control System’, Use of Pheromones and Other Semiochemicals in Integrated Production. IOBC/WPRS Bulletin,

Downloaded By: [Hendrichs, Jorge] At: 15:54 6 November 2009 25, 137Á146. McInnis, D.O., Tam, S.Y., Lim, R.R., Komatsu, J., Kurashima, R.S., Albrecht, C.P., and Shelly, T. (2004), ‘Development of a Pupal Color-Based Genetic Sexing Strain of the Melon Fly, Bactrocera cucurbitae (Diptera: Tephritidae)’, Annals of the Entomological Society of America, 97, 1026Á1033. Michelakis, S.E., and Neuenschwander, P. (1983), ‘Estimates of the Crop Losses Caused by Dacus o/eae (Gmel.) (Diptera, Tephritidae) in Crete, Greece’,inFruit Flies of Economic Importance, in Proceedings of the CEC/IOBC, International Symposium, , ed. R. Cava1loro, November 16Á19 1982; Athens, Greece, 642 p. Mohamed, S.A., Overholt, W.A., Lux, S.A., Wharton, R.A., and Eltoum, E.M. (2007), ‘Acceptability and Suitability of Six Fruit Fly Species (Diptera: Tephritidae) for Kenyan Strains of Psyttalia concolor (Hymenoptera: Braconidae)’, Biocontrol Science and Technol- ogy, 17, 247Á259. Monastero, S. (1959), ‘Altra Straordinaria Cattura di Opius Parassiti di Dacus o/eae Gmel.’,in Sicilia nel 1959, Bollettino d.Istituto Entomologia e Osservatorio Fitopatologia di Palermo, III, 261Á269. Monastero, S., and Genduso, P. (1962), ‘La Lotta Biologica Contro la Mosca delle Olive’, Bollettino d’, Istituto Entomologia e Osservatorio Fitopatologia di Palermo,V,31Á53. 224 B. Hepdurgun et al.

Montiel, A. (1986), ‘The Use of Sex Pheromone for Monitoring and Control of Olive Fruit Fly. II’. International. Symposium Fruit Flies, Crete Sept, 1986, pp. 483Á494. Montiel, A.B., and Jones, O. (2002), ‘Alternative Methods for Controlling the Olive Fly, Bactrocera o/eae, Involving Semiochemica1s’,inUse of Pheromones and Other Semiochem- ica1s in Integrated Production, IOBC/WPRS Bulletin, 25, 147Á156. Peck, S.L., and McQuate, G.T. (2000), ‘Field Tests of Environmentally Friendly Malathion Replacements to Suppress Wild Mediterranean Fruit Fly (Diptera: Tephritidae) Popula- tions’, Journal of Economic Entomology, 93, 280Á289. Prokopy, R.J., Miller, N.W., Pinero, J.C., Barry, J.D., Tran, L.C., Oride, L., and Vargas, R.I. (2003), ‘Effectiveness of GF-120 Fruit Fly Bait Spray Applied to Border Area Plants for Control of Melon Flies (Diptera: Tephritidae)’, Journal of Economic Entomology, 96, 1485Á1493. Ragoussis, N. (2002), ‘Eco-Trap: Efficient Toll for the Control of the Olive Fruit Fly Bactrocera o/eae in the Mediterranean Area’,inUse of Pheromones and Other Semiochem- ica1s in Integrated Production, IOBC/WPRS Bulletin, 25, 195Á202. Rizzi, I., Petacchi, R., and Guidotti, D. (2005), ‘Mass Trapping Technique in Bactrocera oleae Control in Tuscany Region: Results Obtained at Different Territorial Scale’,inProceedings of the Working Group on Integrated Protection of Olive Crops, Chania, Greece: IOBC/WPRS Bulletin, 28, pp. 83Á90. Tedeschini, J. Isufi, E. Uka, R. Bac¸aj, M. and Pfeiffer, D. (2003), ‘Attract and KilI’ Method using EcoTraps for Controlling Olive Fruit Fly in Albania’, 1st European Meeting of the IOBC/WPRS study group ‘Integrated Control in Olives’, 2003 May 29Á31, MAICh-Chania, Crete/Greece. Vargas, R.I., Peck, S.L., McQuate, G.T., Jackson, C.G., Stark, J.D., and Armstrong, J.W. (2001), ‘Potential for Area-Wide Integrated Management of Mediterranean Fruit Fly (Diptera: Tephritidae) with a Braconid Parasitoid and a Novel Bait Spray’, Journal of Economic Entomology, 94, 817Á825. Vargas., R.I., Long, J., Miller, N.W., Delate, K., Jackson, C.G., Uchida, G.K., Bautista, R.C., and Harris, E.J. (2004), ‘Releases of Psyttalia fletcheri (Hymenoptera: Braconidae) and Sterile Flies to Suppress Melon Fly (Diptera: Tephritidae) in Hawaii’, Journal of Economic Entomology, 97, 1531Á1539. Vargas, R.I., Mau, R.F.L., Jang, E.B., Faust, R.M., and Wong, L. (2008), ‘The Hawaii Fruit Fly Areawide Pest Management Programme’,inAreawide Pest Management: Theory and Implementation, eds O. Koul, G. Cuperus, and N. Elliott, CAB International, London, pp. 300Á325. Yokoyama, V.Y., Rendon, P.A., and Sivinski, J. (2008), ‘Psytalia cf. concolor (Hymenoptera: Braconidae) for Biological Control of Olive Fruit Fly (Diptera: Tephritidae) in California’, Environmental Entomology, 37, 764Á773. Downloaded By: [Hendrichs, Jorge] At: 15:54 6 November 2009 Biocontrol Science and Technology, Vol. 19, S1, 2009, 225Á234

RESEARCH ARTICLE Effect of early oviposition experience on host acceptance in Trichogramma (Hymenoptera: Trichogrammatidae) and application of F1 sterility and T. principium to suppress the potato tuber moth (Lepidoptera: Gelechiidae) George Saour*

Atomic Energy Commission (AECS), Department of Biotechnology, P.O. Box 6091, Damascus, Syria

Laboratory experiments with Trichogramma principium Sugonyaev and Sorokina females offered potato tuber moth Phthorimaea operculella (Zeller) eggs demon- strated that wasps’ rates of oviposition were highest the first day and decreased gradually thereafter. In addition, when T. principium females were sequentially offered eggs from 250 Gy irradiated parents or obtained from non-irradiated moths, the probability of host acceptance was not influenced by treatment of host eggs. In a concurrent laboratory study, a large cage test with combinant releases of T. principium and 250 Gy irradiated moths produced the greatest reduction in potato tuber moth F3-emerged progeny. Reductions obtained with irradiated moths alone, single release of irradiated moths with T. principium, and one or three releases of parasitoids were significantly higher than those in the control. From a pest management perspective, T. principium releases would synergistically complement the effects of F1 sterility against potato tuber moth infestation.

Keywords: Phthorimaea operculella; Trichogramma;F1 sterility; pest management; nuclear techniques; gamma radiation

Introduction The cultivated potato Solanum tuberosum L. is one of the world’s major food crops. Potatoes are widely grown over many latitudes and elevations over a wide range of

Downloaded By: [Hendrichs, Jorge] At: 15:54 6 November 2009 agro-ecological zones (FAO 2008). The potato tuber moth (PTM), Phthorimaea operculella (Zeller), has been reported as a pest of major economic importance in almost all the potato producing areas in the tropics and subtropics (Sporleder, Kroschel, Gutierrez Quispe, and Lagnaoui 2004). Systemic organophosphates, plant products or insect growth regulators (IGRs) are widely used to control the PTM in both field and unrefrigerated, rural storage (Das 1995; Edomwande, Schoeman, Brits, and Van Der Merwe 2000; Symington 2003). Releases of a biological control agent such as Trichogramma spp. (an oophagous parasitoid) have been used successfully to control various lepidopteran pests and will likely continue to be a primary component of lepidopterous management programmes (Smith 1996; Saour 2004a). Nevertheless, Trichogramma spp. releases may be more effective over a wide range of conditions if they are integrated with compatible control methods (i.e., sterile insect technique). Other control options,

*Email: [email protected]

ISSN 0958-3157 print/ISSN 1360-0478 online # 2009 Taylor & Francis DOI: 10.1080/09583150802522838 http://www.informaworld.com 226 G. Saour

however, are limited by the extreme sensitivity of adult Trichogramma spp. to various environmental factors (Fournier and Boivin 2000) or pesticide drift (Hassan, Hafes, Degrande, and Herai 1998; Hewa-Kapuge, McDougall, and Hoffmann 2003). Accordingly, a previous study in our laboratory demonstrated that a single release of T. principium along with irradiated moths over potatoes in † small Plexiglas boxes was more effective in reducing PTM F1-emerged progeny than using T. principium or F1 sterility (or inherited sterility) employed separately (Saour 2004b). It is well documented that the success of combining Trichogramma spp. and the sterile insect technique to suppress lepidopteran pests can be influenced by several biotic and abiotic factors (Bloem, Bloem, and Knight 1998; Cossentine and Jensen 2000; Carpenter, Bloem, and Hofmeys 2004). For PTM, data have been published on the acceptability and suitability of host eggs from irradiated parental crosses for three generalist Trichogramma species (Saour 2004b). However, data on wasp oviposition rates, in relation to age, early ovipositional experience (the ability to learn to discriminate between host eggs of different qualities), and treatment synergistic effects on subsequent generations are a necessary prerequisite to achieve efficient control of PTM through combined releases of egg parasitoids and sterile moths. In the current study, females of Trichogramma principium Sugonyaev & Sorokina

(arrhenotokous species) were provided with normal or F1 sterile PTM eggs arising from 250 Gy irradiated parents to study the effects of repeated exposure of T. principium to host eggs on oviposition rates and the early ovipositional experience. Furthermore, the effects of single or repeated releases of T. principium combined with

250 Gy irradiated and non-irradiated moths on PTM F3-emerged progeny in large cages under laboratory conditions were determined.

Materials and methods Potato tuber moth Moths used in the experiments were obtained from a laboratory stock culture (45th generation), which was renewed each year with a field collection of PTM individuals. Larvae were reared at a constant temperature of 25928C with 7095% relative Downloaded By: [Hendrichs, Jorge] At: 15:54 6 November 2009 humidity (RH) and a photoperiod of 12:12 h (L:D) on wax-coated potato slices as delineated by Saour and Makee (1997).

Trichogramma source The T. principium colony used in the experiment originated from the Biological Control Laboratory at the Aleppo University, Syria. T. principium was cultured on eggs of the Mediterranean flour moth, Ephestia kuehniella Zeller. Rearing of E. kuehniella was performed in a controlled climate chamber at 25918C, 7095% RH and photoperiod of 16:8 h (L:D). Eggs were obtained by placing adult E. kuehniella † in a transparent Plexiglas cage (502525 cm) with wire mesh covering the bottom side. Eggs fell through the wire mesh into a container and were stored at 5 28C until use. Further details of the rearing method are outlined in Daumal, Voegele, and Brun (1975). Biocontrol Science and Technology 227

Experimental procedures Individuals of T. principium were captured by releasing them onto a large sheet of white paper and then inverting 91 cm glass tubes over dispersing individuals. When the captured specimen moved upward off the paper, the tube was quickly turned upright and plugged with cotton wads that contained small droplets of pure honey on the inner walls as food medium. Male and female T. principium were then separated by sex by examining antennal characteristics under a binocular microscope at 15 magnification (Kyowa Optical, Japan). The parasitoids were set aside for testing. For subsequent tests, two males also were transferred to each test tube to ensure female mating. Newly emerged adult moths were irradiated at a dose of 250 Gy in order to obtain completely sterile females and partially sterile males (Makee and Saour 2004). Irradiated males and females were singly paired in 350 ml transparent plastic boxes provided with an oviposition support (filter paper) and source of food (10% sucrose solution) and held at a constant temperature of 23918C with 7095% RH, and a photoperiod of 12:12 h (L:D). Females and males were kept together until death. Eggs were removed daily, counted, and left until used. A 60Co source (Gamma irradiator, model Issledova, Techsnabexport Co. Ltd, Moscow, Russia) was used to provide the gamma radiation throughout this study at a dose rate of Â42 Gy/min95%. The absorbed dose was measured using an alcoholic chlorobenzene dosimeter.

Female oviposition rate T. principium parasitization activity, i.e., mean number of daily parasitized eggs per female, was evaluated by isolating newly emerged females in glass tubes (91 cm) and offered 40Á50 PTM eggs on a filter paper (Â24 h old) each day until parasitoid death. Females were fed honey droplets. The exposed eggs were removed daily and placed in a glass vial (42 cm). Within 72 h after exposure, eggs were evaluated for parasitism (parasitized eggs turned black). Daily parasitism and the percentage of cumulative parasitism were determined. Each female that failed to oviposit or parasitized fewer than 6 host eggs throughout the entire experimental period was

Downloaded By: [Hendrichs, Jorge] At: 15:54 6 November 2009 discarded. Generally, a T. principium female should parasitize more than 15 PTM eggs after 24 h of exposure (Saour 2004a). Experiments were held at 25918Cwitha photoperiod of 16:8 h (L:D) and consisted of four replicates, each consisting of 15 T. principium females.

Female early ovipositional experience The experiment consisted of two treatments. In the first treatment, 1-day-old inexperienced single T. principium females were offered sequentially for 24 h unlimited numbers (50) of 24-h-old eggs obtained either from 250 Gy irradiated parents (less preferred host) or from non-irradiated moths (preferred host). In the second treatment, T. principium females were either exposed to 10 preferred eggs or 10 less preferred, irradiated hosts for 6 h, then afterwards they were allowed sequentially to oviposit on 50 preferred or less preferred hosts for 24 h. Viable and non-viable PTM eggs appeared visually identical within 24 h of oviposition. 228 G. Saour

The mean number of parasitized eggs was recorded. In each treatment, females werekeptat25918C with a photoperiod of 16:8 h (L:D). All tests were repeated so that there were a total of three replications, with 30 females per replicate for each test.

Effect of multiple releases of T. principium and sterile moths on PTM F3-emerged progeny Irradiated (250 Gy) and non-irradiated moths with or without T. principium females were released in large polypropylene mesh cages (111.2 m) over 12 kg of intact potatoes placed on a thin layer of sand. Seven experimental treatments involving combinations of normal and irradiated moths were assigned to the cages (Table 1). The cages were maintained at a constant temperature of 25928C and a photoperiod of 16:8 h (L:D). Two days after the release of moths, Trichogramma females were introduced into the experimental cages. The releases of T. principium and irradiated moths for the subsequent generations were performed

2 days after the emergence of F1 or F2 moths. Fourteen cages that represent two replicates were maintained. The experiment was conducted twice to give a total of

four replicates in time. The number of F3-emerged adults (expressed as moths per non-irradiated P1 female) and the fresh weight of potatoes were recorded for each replicate. The corrected losses of potato fresh weight during the experimental period were calculated according to Kogan (1986) using the formula: CIe Á Fe (I0/F0). For this, Ccorrected loss of potato fresh weight, Ietreatment initial weight, I0control initial weight, Fetreatment final weight, and F0control final weight.

Table 1. Releases of 250 Gy irradiated and non-irradiated potato tuber moth adults with or without Trichogramma principiumfor 2 generations througout the experiment.

Treatment Initial generation (P1) 1st generation 2nd generation

1 Uninfested potatoes. ÁÁ

Downloaded By: [Hendrichs, Jorge] At: 15:54 6 November 2009 2 15 pairs of non-irradiated moths. ÁÁ 3 15 pairs of 250 Gy irradiated ÁÁ moths  3 pairs of non-irradiated moths. 4 15 pairs of non-irradiated moths ÁÁ  60 T. principium females. 5 15 pairs of 250 Gy irradiated ÁÁ moths  3 pairs of non-irradiated moths  60 T. principium females. 6 15 pairs of non-irradiated moths 60 T. principium females. 60 T. principium  60 T. principium females. females. 7 15 pairs of 250 Gy irradiated 15 pairs of 250 Gy Á moths  3 pairs of non-irradiated irradiated moths  60 moths  60 T. principium females. T. principium females. Biocontrol Science and Technology 229

Statistical analyses Analysis of variance (ANOVA) at the 5% level (P]0.05) was carried out to evaluate the differences in the means of parasitized eggs for female early

ovipositional experience, number of F3 emerged moths, and corrected losses in tuber fresh weight. Significant ANOVAs were followed by Fisher’s protected least significant differences method (PLSD) at a0.05. Data met the assumptions of normality before the analysis. All analyses were performed using StatView Statistical Software (version 4.02, Abacus Concepts 1994).

Results Female oviposition rate Wasps’oviposition rates were highest the first day and decreased gradually thereafter, but females continued to parasitize low numbers of PTM eggs each day until their death. However, it appears that T. principium females concentrated their activity during the first 5 days after their emergence, with a cumulative parasitization rate of 75% (Figure 1).

25 100

20 80

15 60 Daily parasitism Accumulative Parasitism Downloaded By: [Hendrichs, Jorge] At: 15:54 6 November 2009 10 40 Eggs parasitized/female day Accumulative parasitism (%)

5 20

0 0 12345678910 Days

Figure 1. Mean 9 SD daily parasitized host eggs per female wasp and percentage accumulative parasitism of Trichogramma principium reared on potao tuber moth eggs at 258C and 16:8 h (L:D) photoperiod. 230 G. Saour

Table 2. Mean number (9SD) of potato tuber moth parasitized eggs when Trichogramma principium single female was sequentially offered two different types of eggs resulting either from 250 Gy irradiated parents or obtained from non-irradiated moths at 258C and 16: 8 h (L:D) photoperiod.

Exposure duration (h) No. of exposed eggs Eggs type sequence Parasitized eggs

1st period Á 2nd 1st period Á 2nd 1st period Á 2nd period period period 1st period Á 2nd period

24 Á 24 50Á50 Irradiated Á Normal* 15.995.0Aa 13.796.3Aa Normal Á Irradiated 19.295.4Ab 11.594.2Ba Irradiated Á Irradiated 15.395.2Aa 11.796.0Aa Normal Á Normal 19.596.7Ab 12.995.4Ba 6 Á 24 10Á50 Irradiated Á Normal* 7.694.3Aa 14.992.1Bac Normal Á Irradiated 8.692.2Aa 11.894.1Bb Irradiated Á Irradiated 6.994.1Aa 10.494.0Bb Normal Á Normal 8.492.3Aa 15.193.0Bc

*Eggs from non-irradiated parents. Means in row for each exposure duration followed by the same uppercase letter are not significantly different (P50.05, Fisher LSD); means in column for each exposure duration followed by the same lowercase letter are not significantly different (P50.05, Fisher LSD). Mean of three replicates, 30 T. principium females per replicate.

Female early ovipositional experience When T. principium females were exposed for 24 h to eggs obtained from 250 Gy irradiated or normal parents in the first exposure period, they parasitized similar numbers of eggs arising either from irradiated or non-irradiated parents in the second exposure period. The observed differences in the mean numbers of parasitized eggs in the second exposure were not significant (P0.11). In contrast, parasitoids that foraged for only 6 h in eggs from either irradiated or non-irradiated moths during the first exposure period significantly (P0.05) parasitized more eggs arising from non-irradiated moths during the second exposure period, i.e., they appeared to gain experience and preferred normal eggs during the second period. The mean number of parasitized eggs was not influenced by the irradiated parental crosses after 6 h of exposure and the differences were not significant (P0.52). As

Downloaded By: [Hendrichs, Jorge] At: 15:54 6 November 2009 expected, T. principium females significantly preferred eggs from non-irradiated parents compared to eggs from 250Gy irradiated moths during 24 h of exposure period (P0.01) (Table 2).

Effect of multiple releases of T. principium and sterile moths on PTM F3-emerged progeny

In general, the mean number (97.8) of PTM F3-emerged progeny in the control treatment differed significantly versus all other treatments (F121.4; df 5, 18) (Table 3). However, the multiple releases of parasitoids reduced by 78% the number of F3- emerged moths compared to the single release of T. principium. A low emergence of

moths per non-irradiated P1 female (0.2) was obtained after two releases of T. principium and 250 Gy irradiated and non-irradiated moths at a 5:1 ratio. The mean

number of F3 adults that were obtained after a single release of T. principium and irradiated moths (4.1) was not significantly different (P0.42) than that from two Biocontrol Science and Technology 231

Table 3. Mean total number (9SD) of potato tuber moth F3-emerged progeny and mean losses in tubers fresh weight resulted from one or multiple releases of 250 Gy irradiated and non-irradiated moths with or without Trichogramma principium over 12 kg of intact tubers placed inside large polypropylene mesh cages. The introduction of T. principium was done 2 days after the initial moths release or the emergence of F1 and F2 adults, at 258C and 16: 8 h (L: D) photoperiod.

No. of F3-emerged moths Corrected losses in tubers Treatments per-non-irradiated fresh weight (g/non-irra-

P1 female diated P1 female)

1- Uninfested potatoes. ÁÁ 2- 15 pairs of non-irradiated moths. 97.898.6a 431.0939.8a 3- 15 pairs of 250 Gy irradiated moths3 pairs 28.7912.8b 225.8956.9b of non-irradiated moths. 4- 15 pairs of non-irradiated moths60 29.993.2b 109.3930.2c T. principium females. 5-15 pairs of 250 Gy irradiated moths3 pairs 4.193.0c 58.7922.8cd of non-irradiated moths60 T. principium females. 6- 15 pairs of non-irradiated moths3 releases of 6.691.9c 49.3921.8cd 60 T. principium females. 7- 15 pairs of 250 Gy irradiated moths3 pairs 0.290.1c 17.599.6d of non-irradiated moths females (2 releases 60 T. principium).

Means within each column followed by the same letter are not significantly different (PB0.05, Fisher LSD). Mean of four replicates.

releases of T. principium and irradiated moths. Significant losses in potato fresh

weight per non-irradiated P1 female occurred among the treatments (F180; df 5, 18); however, the corrected loss in tuber weight was determined by the intensity of infestation and relative effects of the different treatments (Table 3).

Discussion Trichogramma species are considered to be pro-ovigenic, or partially synovigenic

Downloaded By: [Hendrichs, Jorge] At: 15:54 6 November 2009 (Pak and Oatman 1982; Volkoff and Daumal 1994). In this study, we found that T. principium females laid a significant number of their eggs shortly after emergence, but it also seemed that the parasitization was distributed over 10 days. The propensity for T. principium females to lay eggs soon after emergence and the observation that females continued oviposition up to their death are in agreement with the results reported for T. principium offered Sitotroga cerealella Olivier (Lepidoptera: Gelechiidae) eggs (Reznik, Voinovich, and Umarova 2001) or with other trichogrammatids (Leatemia, Laing, and Corrigan 1995; Consoli and Parra 1996). Several studies have assessed Trichogramma spp. early oviposition experience and its consequences on host acceptance (Reznik, Umarova, and Voinovich 1997; Keasar, Ney-Nifle, Mangel, and Swezey 2001). High- and low-quality hosts were indis- pensable to carry out such an experiment. Usually, fresh hosts are considered to be good (in the sense that Trichogramma spp. favors them) and older hosts to be bad. In our experiment, two different types of PTM eggs were obtained from the crosses 232 G. Saour

between 250 Gy irradiated or non-irradiated parents. The results of the present study showed that T. principium females tend to maintain parasitization even when sequentially exposed to different types of 1-day-old host eggs. This finding corroborates results that T. principium parasitization behaviour was stable when young (preferred) and old (less preferred) grain moth eggs S. cerealella were offered in sequence (Reznik and Umarova 1991; Reznik et al. 1997). Theoretically, under release conditions involving irradiated moths and para- sitoids, Trichogramma spp. females may randomly contact fertile and sterile eggs. However, under field or storage conditions, higher parasitization is expected in sterile eggs resulting from irradiated parents than in wild eggs. This may be due to the fact that fertile eggs continue to develop and are therefore acceptable for Trichogramma spp. oviposition for a shorter period. For this reason, using sterile host eggs could enhance natural T. principium population build-up. F1 sterility (also known as inherited-partial sterility) and Trichogramma spp. have been developed and separately field-tested against lepidopterous pests. However, combinations of parasitoid and sterile insect releases have been evaluated for suppression of Ceratitis capitata (Wiedemann) and Bactrocera cucurbitae (Coquil- lett) (Diptera: Tephritidae) populations in Hawaii (Wong, Ramadan, Herr, and McInnis 1992; Vargas et al. 2004). Data reported by Mannion, Carpenter, and Gross (1994, 1995) suggest that the use of F1 sterility and the tachinid parasitoid Archytas marmoratus (Townsend) (Diptera: Tachinidae), are compatible strategies for mana- ging early-season population of Helicoverpa zea (Boddie) (Lepidoptera: Noctuidae). Accordingly, Bloem et al. (1998), Cossentine and Jensen (2000) and Saour (2004b) demonstrated that the combined use of egg parasitoids and F1 sterility was very efficient as a control method against the codling moth Cydia pomonella (L.) and the PTM, respectively. Based on our cage studies, the data substantiate this potential and suggest that a single release of irradiated and non-irradiated moths at a 5:1 over-flooding ratio (treatment no. 3), or T. principium females and non-irradiated PTM pairs at 4:1 ratio would have detrimental effects on PTM subsequent generations (reductions of 70.7 and 69.4%, respectively). The combination of T. principium and 250 Gy irradiated moths produced the greatest numerical reduction in PTM F3 emergence from tubers, particularly when two properly timed releases (2Á3 days after moth release or Downloaded By: [Hendrichs, Jorge] At: 15:54 6 November 2009 emergence) were performed. Moreover, the larval mining of the tubers reduced the weight and quality of the potatoes, but when T. principium and irradiated moths were released, the potatoes did not suffer feeding damage. Tuber weight loss was only 13.6 and 4.1%, respectively, when T. principium and irradiated moths were used together on a single or repetitive basis (treatments 5 and 7, Table 1). The data presented provide positive evidence regarding the synergistic effect resulting from combinations between egg parasitoids and sterile moth releases. Nonetheless, the greatest likelihood for crosses occurring in the field under area-wide releases of sterile moths would be irradiated malesirradiated females and normal malesnormal females. Of course, because F1 eggs from the irradiated maleswild females possess the potential of inherited sterility (based upon radiation-induced deleterious effects passed on to F1 generation), it would be advantageous if T. principium females seek out and preferentially parasitize fertile eggs produced by feral moths rather than the F1 eggs present in the treated areas, which are crucial to sustain the F1 population for sterility promotion and subsequent population Biocontrol Science and Technology 233

collapse. However, our findings should be further tested in order to determine whether PTM sterile moths and T. principium releases will be more cost-effective in unrefrigerated storage potatoes.

Acknowledgements I thank Dr. I. Othman (General Director) and Dr. N. Sharabi for their help and support. Technical and financial assistances were provided, in part, by contract no. 10781 of the International Atomic Energy Agency, Joint FAO/IAEA Division of Nuclear Techniques in Food and Agriculture, Vienna, Austria.

References Bloem, S., Bloem, K.A., and Knight, A.L. (1998), ‘Oviposition by Sterile Codling Moth, Cydia pomonella (Lepidoptera: Tortricidae) and Control of Wild Populations with Combined Releases of Sterile Moth and Egg Parasitoids’, Journal Entomological Society British Columbia, 95, 99Á109. Carpenter, J.E., Bloem, S., and Hofmeys, J.H. (2004), ‘Acceptability and Suitability of Eggs of False Codling Moth (Lepidoptera: Tortricidae) from Irradiated Parents to Parasitism by Trichogrammatoidea cryptophlebiae (Hymenoptera: Trichogrammatidae)’, Biological Con- trol, 30, 351Á359. Consoli, F.L., and Parra, J.R.P. (1996), ‘Biology of Trichogramma galloi and T. pretiosum (Hymenoptera: Trichogrammatidae) Reared in vitro and in vivo’, Annals of the Entomolo- gical Society of America, 89, 828Á834. Cossentine, J.E., and Jensen, L.B.M. (2000), ‘Releases of Trichogramma platneri (Hymenop- tera: Trichogrammatidae) in Apple Orchards under a Sterile Codling Moth Release Program’, Biological Control, 3, 179Á186. Das, G.P. (1995), ‘Plants Used in Controlling the Potato Tuber Moth, Phthorimaea operculella (Zeller)’, Crop Protection, 14, 631Á636. Daumal, J., Voegele, J., and Brun, P. (1975), ‘Les Trichogrammes. II. Untie´ de Production Massive et Quotidenne d’un Hoˆte de Substitution, Ephestia kuehniella Zeller (Lepidoptera, Pyralidae’, Annales de Zoologie Ecologie Animale,7,45Á59. Edomwande, E.O., Schoeman, A.S., Brits, J.A., and Van Der Merwe, M. (2000), ‘Laboratory Evaluation of Lufenuron on Immature Stages of Potato Tuber Moth (Lepidoptera: Gelechiidae)’, Journal of Economic Entomology, 93, 1741Á1743. FAO (2008), International Year of the Potato, www.potato2008.org/en/index.html Fournier, F., and Boivin., G. (2000), ‘Comparative Dispersal of Trichogramma evanescens and Downloaded By: [Hendrichs, Jorge] At: 15:54 6 November 2009 Trichogramma pretiosum (Hymenoptera: Trichogrammatidae) in Relation to Environmental Conditions’, Environmental Entomology, 29, 55Á63. Hassan, S.A., Hafes, B., Degrande, P.E., and Herai, K. (1998), ‘The Side-Effects of Pesticides on the Egg Parasitoid Trichogramma cacoeciae Marchal (Hymenoptera, Trichogrammati- dae), Acute DoseÁResponse and Persistence Tests’, Journal of Applied Entomology, 122, 569Á573. Hewa-Kapuge, S., McDougall, S., and Hoffmann, AA. (2003), ‘Effects of Methoxfenozide, Indoxacarb, and Other Insecticides on the Beneficial Egg Parasitoid Trichogramma nr. brassicae (Hymenoptera: Trichogrammatidae) Under Laboratory and Field Conditions’, Journal of Economic Entomology, 96, 1083Á1090. Keasar, T., Ney-Nifle, M., Mangel, M., and Swezey, S. (2001), ‘Early Oviposition Experience Affects Patch Residence Time in a Foraging Parasitoid’, Entomologia Experimentalis et Applicata, 98, 123Á132. Kogan, M. (1986), Bioassays for Measuring Quality of Insect Food, in Insect-Plant Interactions , eds. J.R. Miller and T.A. Miller, New York: Springer-Verlag, pp. 155Á189. Leatemia, J.A., Laing, J.E., and Corrigan, J.E. (1995), ‘Effects of Adult Nutrition on Longevity, Fecundity, and Offspring Sex Ratio of Trichogramma minutum Riley (Hyme- noptera: Trichogrammatidae)’, Canadian Entomologist, 127, 245Á254. 234 G. Saour

Makee, H., and Saour, G. (2004), ‘Efficiency of Inherited Sterility Technique Against Phthorimaea operculella Zeller (Lepidoptera: Gelechiidae) as Affected by Irradiation of Females’, Journal of Vegetable Crop Production, 10, 11Á22. Mannion, C.M., Carpenter, J.E., and Gross, H.R. (1994), ‘Potential of the Combined Use of Inherited Sterility and a Parasitoid, Archytas marmoratus (Diptera: Tachinidae), for Managing Helicoverpa zea (Lepidoptera: Noctuidae)’, Environmental Entomology, 23, 41Á 46. Mannion, C.M., Carpenter, J.E., and Gross, H.R. (1995), ‘Integration of Inherited Sterility and a Parasitoid, Archytas marmoratus (Diptera: Tachinidae), for Managing Helicoverpa zea (Lepidoptera: Noctuidae): Acceptability and Suitability of Hosts’, Environmental Entomology, 24, 1679Á1684. Pak, G.A., and Oatman, E.R. (1982), ‘Comparative Life Table, Behavior and Competition Studies of Trichogramma brevicapillum and T. pretiosum’, Entomologia Experimentalis et Applicata, 32, 68Á79. Reznik, S.Ya, and Umarova, T.Ya (1991), ‘Host Population Density Influence on Host Acceptance in Trichogramma’, Entomologia Experimentalis et Applicata, 58, 49Á54. Reznik, S.Ya, Umarova, T.Ya, and Voinovich, N.D. (1997), ‘The Influence of Previous Host Age on Current Host Acceptance in Trichogramma’, Entomologia Experimentalis et Applicata, 82, 153Á157. Reznik, S.Ya, Voinovich, N.D., and Umarova, T.Ya (2001), ‘Long-Term Egg Retention and Parasitization in Trichogramma principium (Hymenoptera, Trichogrammatidae)’, Journal Applied Entomology, 125, 169Á175. Saour, G. (2004a), ‘Efficacy Assessment of Some Trichogramma species (Hymenoptera: Trichogrammatidae) in Controlling the Potato Tuber Moth Phthorimaea operculella Zell. (Lepidoptera: Gelechiidae)’, Journal Pest Science, 77, 229Á234. Saour, G. (2004b), ‘Parasitization of Potato Tuber Moth Eggs (Lepidoptera: Gelechiidae) from Irradiated Adults by Trichogramma (Hymenoptera: Trichogrammatidae) and Control of Moth Population with Combined Releases of Sterile Insect and Egg Parasitoid’, Journal Applied Entomology, 128, 681Á686. Saour, G., and Makee, H. (1997), ‘Radiation Induced Sterility in Male Potato Tuber Moth Phthorimaea operculella (Lep., Gelechiidae)’, Journal Applied Entomology, 121, 411Á415. Smith, S.M. (1996), ‘Biological Control with Trichogramma: Advances, Successes, and Potential of Their Use’, Annual Review of Entomology, 41, 375Á406. Sporleder, M., Kroschel, J., Gutierrez Quispe, M.R., and Lagnaoui, A. (2004), ‘A Temperature-Based Simulation Model for the Potato Tuberworm, Phthorimaea operculella Zeller (Lepidoptera; Gelechiidae)’, Environmental Entomology, 33, 477Á486. StatView Statistical Software (1994), version 4.02. Berkeley, CA: Abacus Concepts. Symington, C.A. (2003), ‘Lethal and Sublethal Effects of Pesticides on the Potato Tuber Moth, Phthorimaea operculella (Zeller) (Lepidoptera: Gelechiidae) and its Parasitoid

Downloaded By: [Hendrichs, Jorge] At: 15:54 6 November 2009 Orgilus lepidus Muesebeck (Hymenoptera: Braconidae)’, Crop Protection, 22, 513Á519. Vargas, R.I., Long, J., Miller, N.M., Delate, K., Jackson, C.G., Uchida, G.K., Bautista, R.C., and Harris, E.J. (2004), ‘Releases of Psyttalia fletcheri (Hymenoptera: Braconidae) and Sterile Flies to Suppress Melon Fly (Diptera: Tephritidae) in Hawaii’, Journal of Economic Entomology, 97, 1531Á1539. Volkoff, A.N., and Daumal, J. (1994), ‘Ovarian Cycle in Immature and Adult Stages of Trichogramma cacoeciae and T. brassicae (Hymenoptera: Trichogrammatidae)’, Entomo- phaga, 39, 303Á312. Wong, T.T.Y., Ramadan, M.M., Herr, J.C., and McInnis, D.O. (1992), ‘Suppression of a Mediterranean Fruit Fly (Diptera: Tephritidae) Population with Concurrent Parasitoid and Sterile Fly Releases in Kula, Maui, Hawaii’, Journal of Economic Entomology, 85, 1671Á 1681. Biocontrol Science and Technology, Vol. 19, S1, 2009, 235Á242

RESEARCH ARTICLE Evaluating the use of nuclear techniques for colonization and production of Trichogramma chilonis in combination with releasing irradiated moths for control of cotton bollworm, Helicoverpa armigera Endong Wanga, Daguang Luc*, Xiaohui Liub, and Yongjun Lib

aChina Agricultural University, Beijing 100193, China; bInstitute of Plant Protection, Chinese Academy of Agricultural Sciences (CAAS), Beijing 100193, China; cDepartment of International Cooperation, Chinese Academy of Agricultural Sciences (CAAS), Beijing 100081, China

Gamma radiation was tested as a means of increasing production of the egg parasitoid Trichogramma chilonis Ishii by improving the suitability of host eggs and by stimulating reproduction of the parasitoid females. For manipulation of the host eggs’ suitability, radiation was used to either (a) produce developmen- tally-inactivated (DI) eggs incapable of hatching, or (b) to produce F1 sterile host eggs. For treatment of the parasitoid females with the intent of stimulating reproduction, parasitoid pupae were exposed to very low dose radiation (250 mGray). For tests on host suitability using radiation-induced DI host eggs, newly- laid (B8 h old) host eggs (Helicoverpa armigera Hubner) were exposed to 300 Gy 60 of Co gamma radiation. For tests of F1 sterile host eggs, H. armigera moths were mated with individuals exposed to 250 Gy as pupae. Tests were performed with eggs resulting from all possibilities of normal (N) and sterile (S) àß matings. Both types of DI host eggs (irradiated or sterile), along with untreated host eggs (controls), were exposed to T. chilonis females, using the following host egg-to-parasitoid ratios: 1:10, 1:30, 1:60 and 1:90. Developmentally-inactivated host eggs exposed to 300 Gy did not differ in suitability from normal host eggs at a 1:10 parasitoidÁhost ratio, but were significantly more suitable at the higher hostÁparasitoid ratios. F1 sterile eggs were not significantly different in suitability from normal eggs at a 1:10 hostÁparasitoid ratio but were marginally better at the higher hostÁparasitoid ratios. In tests performed using T. chilonis females exposed Downloaded By: [Hendrichs, Jorge] At: 15:55 6 November 2009 to low-dose radiation (250 mGy), no effects were observed when cohorts of 5 T. chilonis females were provided with only 50 host eggs, but when more hosts were provided (ratios of 1:30, 1:60 and 1:90), significantly higher rates of parasitization were noted for the parasitoids exposed to low-dose radiation. This effect prevailed using both normal host eggs and DI host eggs exposed to 300 Gy. The stimulatory effect also was noted when F1 sterile host eggs were provided to the irradiated T. chilonis females. These results suggest that release of T. chilonis irradiated with 250 mGy may complement release of irradiated H. armigera moths, which produce sterile F1 eggs that can serve as supplemental hosts in the field and thereby enhance the pest management system. Keywords: Trichogramma; Helicoverpa; biological control; pest management; parasitoids; irradiation; low dose irradiation; radiation hormesis

*Corresponding author. Email: [email protected]

First Published Online 5 May 2009 ISSN 0958-3157 print/ISSN 1360-0478 online # 2009 Taylor & Francis DOI: 10.1080/09583150902790293 http://www.informaworld.com 236 E. Wang et al.

Introduction Releases of insect natural enemies along with irradiated, sterile pest insect eggs were evaluated to achieve improved control of the cotton bollworm, Helicoverpa armigera Hubner. The irradiated, sterile pest eggs were intended to serve as innocuous supplemental hosts for use as a field insectary or to sustain parasitoid populations at times of low host populations. Studies by Saour (2004) on the potato tuber moth, Phthorimaea operculella (Zeller) showed that release of irradiated moths along with three Trichogramma spp. were complementary and provided an integrated control approach by using inherited sterility in conjunction with these Trichogramma spp. for P. operculella suppression. One way in which radiation might be helpful for egg parasitoids such as Trichogramma spp. would be to inhibit host egg development beyond a point of optimal suitability. Tunc¸bilek and Ayvaz (2003) and Pak (1986) cite a number of studies in which host age was shown to be important for acceptance by egg parasitoids. In studies on the influence of host age on parasitism by Trichogramma evanescens Westwood, Tunc¸bilek and Ayvaz (2003) found that newly-laid Ephestia kuhniella Zeller eggs were preferred as compared with older eggs. Tunc¸bilek, Canpolat, and Ayvaz (2009) found that eggs of E. kuehniella and Sitotroga cerealella (Olivier) irradiated with 200 Gy gamma radiation and exposed to T. evanescens for 24 h were suitable for parasitoid development. Harwalkar, Rananavare, and Rahaikar (1987) found that Trichogramma brasiliense could be successfully reared on radiation-sterilized potato tubermoth (Phthorimaea operculella Zeller) eggs and that even after rearing 10 generations of the parasitoid on such eggs, no adverse effects were evident. Brower (1982) also found that successful parasitization by T. pretiosum Riley on Plodia interpunctella (Hu¨bner) eggs could be increased by exposing the eggs at 350 Gy (but not to 500 or 1000 Gy). He also found that eggs from P. interpunctella adults irradiated at 150 Gy, as might be used to achieve F1 sterility, were successfully parasitized at the same rate as control eggs. Thus, it appeared that if the eggs are to be used strictly for parasitoid production, the better option would be to irradiate the eggs directly rather than to irradiate the adult moths, but eggs from moths exposed to up to 150 Gy to achieve inherited sterility would still be suitable for parasitoids in the field. Downloaded By: [Hendrichs, Jorge] At: 15:55 6 November 2009 Cossentine and Jensen (2000) also reported that sterile eggs from released, irradiated Cydia pomonella (L.) were successfully parasitized by Trichogramma platneri Nagarkatti, and these eggs could be used to sustain a population of T. platneri at times of low C. pomonella density. These papers indicated that either irradiated insect eggs or eggs from irradiated moths may be at least equally acceptable for parasitization by Trichogramma spp. as compared with normal eggs. Liu and Chen (1983) reported that embryos of eggs from 300 Gy irradiated male and normal female adults did not finish development, and were developmentally suspended with abundant yolk at this stage egg, resulting in Trichogramma spp. finding irradiated eggs suitable for parasitization beyond the point of acceptability for non-irradiated eggs. Another topic of interest is the possible use of low-dose radiation to stimulate reproduction by parasitoids. The phenomenon known as radiation hormesis (Luckey 1991) refers to the stimulatory effect of very low-dose radiation on biological processes. It often refers to accelerated growth of plants due to low-dose radiation, but Biocontrol Science and Technology 237

it also has been found to influence biological processes in insects. For example, studies by Yusifov, Kuzin, Agaev, and Alieva (1990) showed that low doses of ionizing radiation (100Á4000 times exceeding natural background radiation, or about 2 Gy) stimulated embryogenesis in the silkworm, Bombyx mori L. Stimulatory effects on growth of B. mori larvae also were found by Abdel-Salam and Mahmoud (1995) in response to low levels of gamma radiation (0.01Á1 Gy, with the greatest effects at 1 Gy). The goals of the present study were to assess the influence of irradiation of H. armigera eggs and eggs resulting from irradiated pupae on their acceptability and suitability for parasitization by Trichogramma chilonis (Ishii). We also sought to evaluate the influence of very low dose radiation of the parasitoids themselves on their parasitization potential in normal H. armigera eggs and in eggs resulting from moths irradiated as pupae.

Methods and materials Colonies The strain of T. chilonis used in these tests was obtained from the Institute of Plant Protection, Beijing Academy of Agricultural Sciences, where it was reared for many generations on eggs of Antheraea pernyi Guer. The strain of H. armigera maintained in our laboratory was collected from a cotton field in Gaoyang county, Hebei province in July, 2002. Tests were performed at 28928C, 6595% RH with a 14 h L:10 h D photoperiod regimen.

Test methods

Testing irradiated Helicoverpa armigera eggs and F1 sterile eggs for suitability for Trichogramma chilonis Newly-laid (B8 h old) eggs oviposited by normal mated H. armigera moths were collected on napkin paper (‘egg-paper’) and irradiated at a dose of 300 Gy, using a 60Co gamma radiation source located at the Institute for Application of Atomic Energy, CAAS, producing a dose rate of 3.14 Gy/min. These ‘egg-papers’ were

Downloaded By: [Hendrichs, Jorge] At: 15:55 6 November 2009 placed in glass tubes and 5 T. chilonis female adults were transferred into these tubes and held together for 24 h at a T. chilonis female-to-host egg ratio of 1:10, 1:30, 1:60 or 1:90 (viz. 5:50, 5:150, 5:300 or 5:450 for no. of T. chilonis to irradiated host eggs, respectively), using irradiated host eggs. Control (unirradiated, newly-laid) eggs also were exposed in the same fashion and at the same parasitoid-to-host egg ratios. The experiment was repeated 5 times. For tests involving exposure of F1 eggs from irradiated pupae, the day before expected emergence, H. armigera pupae were irradiated using 250 Gy. After emergence, treated females (Sà); treated males (Sß); normal, untreated females (Nà); and normal, untreated males (Nß) were confined in mating and oviposition cages in the following combinations: NàNß,NàSß,SàNß, and SàSß. The ‘egg-papers’ were like those used for normal H. armigera moths described above but only F1 eggs from these crosses were used. The exposure ratios for T. chilonis females to H. armigera eggs were the same as above for 1:10, 1:30, 1:60 and 1:90. The experiment was repeated 5 times. 238 E. Wang et al.

Low-dose irradiation of Trichogramma chilonis adults For tests on the influence of low dose radiation on T. chilonis reproductive potential, H. armigera eggs parasitized by T. chilonis were irradiated when the parasitoids were in the pupal stage. We used a dose of only 250 mGy, with a dose rate of only 38.7 mGy per min to minimize potential damage to the parasitoids. After emergence, irradiated T. chilonis females and normal T. chilonis were given an opportunity to parasitize normal H. armigera eggs and host eggs exposed to 300 Gy, using the same type of oviposition chamber as above. The ratios of T. chilonis females to H. armigera eggs employed also were 1:10, 1:30, 1:60 and 1:90. The experiment was repeated 5 times.

Statistical analyses The data on the parasitization rates of T. chilonis on H. armigera eggs were analyzed using an independent sample t-test. The remaining data were analyzed using one-way ANOVA followed by a Duncan’s test; these data were checked by homogeneity of variance test analysis. All analyses were carried out using SPSS (SPSS Inc. 2004). Rejection level was set when P0.05.

Results

The influence of irradiation of Helicoverpa armigera eggs and F1 sterile eggs on parasitizing potential of Trichogramma chilonis parasitoids The parasitization rate by T. chilonis on 300 Gy irradiated H. armigera eggs was significantly higher than that on normal eggs at the following ratios of T. chilonis female adults to H. armigera eggs: 1:30, 1:60 and 1:90, but not at a ratio of 1:10 (Table 1). This suggests that it may be possible to increase the opportunity of T. chilonis parasitoids by using host eggs previously exposed to 300 Gy irradiation when ratios of at least 30 host eggs per parasitoid female are used. When normal and irradiated H. armigera males and females were crossed and the resulting eggs were exposed to T. chilonis females, no significant differences in parasitization rates were noted among the crosses when only a 1:10 ratio of Downloaded By: [Hendrichs, Jorge] At: 15:55 6 November 2009 parasitoid to host eggs were tested (F0.370, df3, P0.776) (Table 2). However, as the egg-to-parasitoid ratio increased, differences in parasitization rates were observed among the crosses. Notably, the parasitization rate at ratios of 1:30 (F 10.177, df3, P0.001) and 1:60 (F9.346, df3, P0.001) was highest for eggs

Table 1. The parasitization rate (%) of Trichogramma chilonis on normal and 300 Gy irradiated Helicoverpa armigera eggs.

No. T. chilonis female adults: H. armigera eggs

Type of H. armigera eggs 1:10 1:30 1:60 1:90

Control 51.697.3a 18.091.6a 9.590.8a 6.190.3a 300 Gy 52.894.2a 43.294.6b 31.591.8b 22.890.7b

Note: Means (9SE) followed by different letters in the same column are significantly different (t-test with significance level 0.05). Biocontrol Science and Technology 239

Table 2. The parasitization rate (%) of Trichogramma chilonis on normal Helicoverpa armigera eggs and eggs resulting from 250 Gy irradiated H. armigera adults mated with normal adults.

No. T. chilonis female adults/H.armigera eggs

Cross 1:10 1:30 1:60 1:90

NàNß 55.694.7a 17.691.3a 8.790.4a 6.290.4a NàSß 53.693.4a 24.391.2b 21.593.5b 10.990.4c SàNß 49.694.4a 19.390.2a 13.390.4a 12.690.7d SàSß 52.094.15a 18.790.37a 12.690.3a 9.290.4b

Note: Means (9SE) followed by different letters in the same column are significantly different (Multiple range tests: Duncan test with significance level 0.05).

from NàSß relative to the other treatments, but the ratio of 1:90 (F31.107, df 3, P0.000) was highest for eggs from SàNß relative to the other treatments.

Influence of low dose radiation of Trichogramma chilonis on their reproductive potential in normal and F1 sterile Helicoverpa armigera eggs The parasitization rate for 250 mGy-irradiated T. chilonis females on 300 Gy irradiated H. armigera eggs was significantly greater than that on normal eggs for all ratios except 1:10 and 1:60 (Table 3). As in the tests shown in Table 1, this also suggests that it may be possible to increase the opportunity of T. chilonis parasitoids by using 300 Gy irradiated H. armigera eggs when ratios of T. chilonis female adults to H. armigera eggs of 1:30 or 1:90 are used. Irradiation of both the parasitoids (with 250 mGy) and their hosts (with 300 Gy) positively influenced the parasitization rate at all hostÁparasitoid ratios (Table 4), suggesting that low dose radiation may be stimulating the reproductive potential of T. chilonis females while they are in the pupal stage. Tests involving parasitoids exposed to low dose radiation showed that they also produced significantly higher parasitiza- tion rates than from control host eggs when provided with F1 sterile eggs resulting

Downloaded By: [Hendrichs, Jorge] At: 15:55 6 November 2009 from NàSß (Table 5), just as in the previous tests with normal, non-irradiated T. chilonis females (Table 2). This occurred at all hostÁparasitoid ratios above 1:10. When parasitoids exposed to low dose radiation were provided with normal host eggs vs. eggs from F1 sterile moths, especially from the NàSß cross, they generally exhibited a higher parasitization rate than the normal, non-irradiated parasitoids

Table 3. The parasitization rate (%) of 250 mGy low dose irradiated Trichogramma chilonis on normal and 300 Gy irradiated Helicoverpa armigera eggs.

No. T. chilonis female adults/H. armigera eggs

Type of H. armigera eggs 1:10 1:30 1:60 1:90

Control 72.094.3a 55.293.9a 47.795.6a 27.291.2a 300Gy 69.693.2a 78.391.8b 57.295.2a 39.991.4b

Note: Means (9SE) followed by different letters in the same column are significantly different (t-test with significance level 0.05). 240 E. Wang et al.

Table 4. The parasitization rate (%) of 250 mGy low dose irradiated Trichogramma chilonis on normal and 300 Gy irradiated Helicoverpa armigera eggs.

No. T. chilonis female adults/H. armigera eggs

Type of H. armigera eggs Type of T. chilonis 1:10 1:30 1:60 1:90

Control Control 51.697.3a 18.091.6a 9.590.8a 6.190.3a 250 mGy 72.094.2b 55.293.9b 47.795.6b 27.291.0b 300 Gy Control 52.894.2a 43.294.6a 31.591.8a 22.790.3a 250 mGy 69.693.2b 78.391.8b 57.295.2b 39.991.4b

Note: Means (9SE) followed by different letters in the same column for a given host egg type are significantly different (t-test with significance level 0.05).

(Table 6). This was most evident at the hostÁparasitoid ratio of 1:30 and 1:90 for all crosses.

Discussion Control of H. armigera by augmentative releases of T. chilonis combined with release of 250 Gy irradiated moths to produce sterile F1 progeny might be a promising strategy to cope with this pest, which has been a serious problem in recent years in China. This methodology must prove cost-effective as well as sustainable and environmentally-friendly. The studies we report indicate that irradiated H. armigera eggs can be used successfully as hosts for T. chilonis in the insectary, and eggs from F1 sterile moths may indeed prove useful as supplemental hosts to maintain and increase the feral population of T. chilonis. These findings are in keeping with the uses of nuclear techniques in biological control proposed by Wang and Wang (2003). We found that irradiation can be useful in three ways: (1) by decreasing loss of suitability of H. armigera eggs as hosts for T. chilonis beyond the first day after oviposition by exposing them to 300 Gy of radiation to inhibit embryonic development; (2) by producing F1 sterile eggs from H. armigera moths exposed to

Downloaded By: [Hendrichs, Jorge] At: 15:55 6 November 2009 Table 5. The parasitization rate (%) of 250 mGy low dose irradiated Trichogramma chilonis on normal Helicoverpa armigera eggs and eggs resulting from 250 Gy irradiated H. armigera adults mated with normal adults.

No. T. chilonis female adults/H. armigera eggs

Cross 1:10 1:30 1:60 1:90

NàNß 50.895.1a 21.691.0a 19.591.6a 16.490.4ab NàSß 47.293.5a 41.793.2b 24.691.8b 17.890.6b SàNß 40.493.3a 41.394.9b 22.190.8ab 15.890.8a SàSß 44.092.5a 36.593.3b 20.190.0a 15.290.4a F 1.424 7.601 2.951 3.876 df 3 3 3 3 P 0.273 0.002 0.064 0.029

Note: Means (9SE) followed by different letters in the same column are significantly different (Multiple range tests: Duncan test with significance level 0.05). Biocontrol Science and Technology 241

Table 6. The parasitization rate (%) of normal and 250 mGy low dose irradiated Trichogramma chilonis on eggs resulting from 250 Gy irradiated Helicoverpa armigera adults mated with normal adults.

No. T. chilonis female adults/H. armigera eggs

Cross Type of T. chilonis 1:10 1:30 1:60 1:90

NàNß Control 49.293.2a 17.691.3a 8.790.4a 6.290.4a 250 mGy 50.895.1a 21.691.0b 19.591.6b 16.490.4b NàSß Control 48.093.2a 24.391.2a 21.593.47a 10.990.4a 250 mGy 47.293.5a 41.793.2b 24.691.8a 17.890.6b SàNß Control 41.694.7a 19.390.2a 13.390.4a 12.690.7a 250 mGy 40.493.3a 41.394.9b 22.190.8b 15.890.8b SàSß Control 46.094.2a 18.790.4a 12.690.3a 9.290.4a 250 mGy 44.092.5a 36.593.3b 20.191.0b 15.1290.4b

Note: Means (9SE) followed by different letters in the same column for a given cross are significantly different (t-test with significance level 0.05).

250 Gy radiation, and by showing that their eggs are highly suitable for oviposition by T. chilonis females; and (3) by stimulating the reproductive potential of T. chilonis females exposed to a very low dose (250 mGy) radiation while they are in the pupal stage. In this study, only the parasitization rates of T. chilonis on H. armigera eggs were analyzed. There is a critical need to evaluate the fitness of parasitoids emerging from irradiated eggs as well as the sex allocation of offspring from irradiated female parasitoids. There also is a critical need to field test these findings. These results support the possibility that radiation can be an important tool in developing an improved rearing and release capability and thereby an improved pest management system for use of Trichogramma chilonis and F1 steriles against Helicoverpa armigera.

Acknowledgements Thanks are due to Dr Patrick Greany and Mr Wang Huesong for their help and support. Technical and financial assistances were provided, in part, by contract no. 10778 of the

Downloaded By: [Hendrichs, Jorge] At: 15:55 6 November 2009 International Atomic Energy Agency, Joint FAO/IAEA Division of Nuclear Techniques in Food and Agriculture, Vienna, Austria.

References Abdel-Salam, K.A., and Mahmoud, S.M. (1995), ‘Stimulating Effects of Low Levels of 60Co Gamma Ray on the Silkworm, Bombyx mori (L.), Anzeiger fur Schadlingskunde, Pflanzenschutz’, Umweltschutz, 68, 147Á150. Brower, J.H. (1982), ‘Parasitization of Irradiated Eggs and Eggs from Irradiated Adults of the Indian Meal Moth (Lepidoptera: Pyralidae)byTrichogramma pretiosum (Hymenoptera: Trichogrammatidae)’, Journal of Economic Entomology, 75, 939Á944. Cossentine, J.E., and Jensen, L.B.M. (2000), ‘Releases of Trichogramma platneri (Hymenop- tera: Trichogrammatidae) in Apple Orchards under a Sterile Codling Moth Release Program’, Biological Control, 18, 179Á186. Harwalkar, M.R., Rananavare, H.D., and Rahailkar, G.W. (1987), ‘Development of Trichogramma brasiliensis (Hymenoptera: Trichogrammatidae) on Eggs of Radiation Sterilized Females of Potato Tuberworm, Phthorimaea operculella (Lepidoptera: Gelechii- dae)’, Entomophaga, 32, 159Á162. 242 E. Wang et al.

Liu, S., and Cheng, G. (1983), ‘The Embryonic Development of the Sugarcane Yellow Stem Borer and the Parasitism of Trichogramma on Sterile Eggs’, Journal of South China Agricultural College (in Chinese), 4, 77Á87. Luckey, T.D. (1991), Radiation Hormesis, Boca Raton, FL: CRC Press. Pak, G.A. (1986), ‘Behavioral Variation among Strains of Trichogramma spp. A Review of the Literature on Host Selection’, Journal of Applied Entomology, 101, 55Á64. Saour, G. (2004), ‘Parasitization of Potato Tuber Moth Eggs (Lepidoptera: Gelechiidae) from Irradiated Adults by Trichogramma (Hymenoptera: Trichogrammatidae) and Control of Moth Population with Combined Releases of Sterile Insect and Egg Parasitoid’, Journal of Applied Entomology, 128, 681Á686. SPSS Inc. (2004), User’s Manual, Chicago, IL: Author. Tunc¸bilek, A.S., and Ayvaz, A. (2003), ‘Influences of Host Age, Sex Ratio, Population Density, and Photoperiod on Parasitism by Trichogramma evanescens Westwood (Hymenoptera: Trichogrammatidae)’, Journal of Pest Science, 76, 176Á180. Tunc¸bilek, A.S., Canpolat, U., and Ayvaz, A. (2009), Effects of Gamma Radiation on Suitability of Stored Cereal Pest Eggs and the Reproductive Capability of the Egg Parasitoid Trichogramma evanescens (Trichogrammatidae: Hymenoptera), Biocontrol Science and Technology (this volume). Wang, E.D., and Wang, H.S. (2003), A Review of Nuclear Technique Application in Production and Augmentation of Natural Enemies for Control Insect Pest, Journal of Nuclear Agricultural Sciences (in Chinese), 17, 319Á322. Yusifov, N.I., Kuzin, A.M., Agaev, F. A., and Alieva, S.G. (1990), ‘The Effect of Low Level Ionizing Radiation on Embryogenesis of Silkworm, Bombyx mori L.’, Radiation and Environmental Biophysics, 29, 323Á327. Downloaded By: [Hendrichs, Jorge] At: 15:55 6 November 2009 Biocontrol Science and Technology, Vol. 19, S1, 2009, 243Á259

RESEARCH ARTICLE Effects of host density, host age, temperature and gamma irradiation on the mass production of Nesolynx thymus (Hymenoptera: Eulophidae), an endoparasitoid of Uzi fly, Exorista sorbillans (Diptera: Tachinidae) Md Mahbub Hasan*, Md Rayhan Uddin, Md Ataur Rahman Khan, and Aminuzzaman Md Saleh Reza

Department of Zoology, Rajshahi University, Rajshahi 6205, Bangladesh

The Uzifly, Exorista sorbillans Weidemann, is an endoparasitoid of the the silkworm moth, Bombyx mori L., and can impact commercial sericulture. Effects of Uzifly age, density, temperature and gamma radiation (60Co) on the mass- production of the hyperparasitoid Nysolynx thymus (Girault) were investigated. There was a direct relationship between progeny production and an increase in number of host puparia presented, and a significant decline in the number of progeny with increased host-age. Maximum progeny production was obtained by maintaining a 1:5 to 1:8 parasitoid-to-host ratio for all the age groups. Two to 4- day-old host puparia were most suitable for obtaining the maximum progeny production of N. thymus for all the host-densities. The intrinsic rate of increase 1 (rm day ) increased with the increased host density for all the host-age groups. The values of net reproductive rate (R0) and gross reproductive rate (GRR) increased for all the density levels and host age groups. On the other hand, the values for the doubling time (D) and finite capacity of increase (l) gradually decreased with increased host density for all the host age groups. Host density and host-age significantly influenced the sex-ratio of progeny in N. thymus. Higher proportions of females were observed at higher host density levels and for younger host age groups. The progeny production and sex ratio of the parasitoid varied significantly with temperature. The maximum mean number of progeny was recorded at 258C, while the minimum was at 308C. The trend of progeny production at different temperatures was on the order 2520308C. The highest values for the net reproductive rate (R0) and GRR for the progeny production

Downloaded By: [Hendrichs, Jorge] At: 15:55 6 November 2009 were recorded at 258C compared to 20 and 308C. Both the values for doubling time of capacity (D) and finite capacity (l) increased with the trend of 25, 30 and 1 208C. The highest value for the intrinsic rate of increase (rm day ) of progeny production was recorded at 258C, while the lowest value was at 208C. The sex- ratios were always female-biased at all the temperatures. Temperature had a significant effect on the longevity of adult N. thymus. The longevity of the adults decreased with an increase of temperature for both sexes. The highest rate of parasitism was observed at 208C followed by 25 and 308C. More than 95% parasitism was observed at all temperatures. Gamma irradiation significantly increased the progeny production of N. thymus when reared either on early or late irradiated host puparia, particularly in the parental generation, but irradiated early host pupae were more suitable for mass production of N. thymus than the irradiated late pupae. The sex ratio of parasitoids developing from gamma irradiated host pupae varied significantly. Higher proportions of females were observed for all the dose and host-age groups.

*Corresponding author. Email: [email protected]

First Published Online 17 April 2009 ISSN 0958-3157 print/ISSN 1360-0478 online # 2009 Taylor & Francis DOI: 10.1080/09583150902790319 http://www.informaworld.com 244 M.M. Hasan et al.

Keywords: host density; host age; temperature, gamma irradiation; mass production; parasitoid; Nesolynx thymus; Exorista sorbillans; Bombyx mori

Introduction The number of parasitoids inoculated into a system, the synchrony of the parasitoid and its host, and the generation time of the parasitoid relative to that of the host are key criteria affecting the likely success of the parasitoid as a classical biological control agent (Barlow, Goldson, and McHeill 1994). Nesolynx thymus (Girault) is a gregarious pupal parasitoid of the Uzifly, Exorista sorbillans Weidemann, a tachinid endoparasitoid of silkworm, Bombyx mori, and is globally known as Uzi. The Uzifly causes economic injury to the cocoon crop in silkworm cultivating areas of India, except those above 400 m above mean sea level (AMSL) in the foothills of the Himalayas (Darjeeling). Among the several hymenopteran parasitoids of the Uzifly, N. thymus was found to have the best characteristics as a potential control agent (Kumar, Kishore, Jayaprakas, and Sengupta 1991). The physiological interactions between parasitoids and their hosts are not only complex but each association appears to be unique (Vinson and Iwantsch 1980). However, for optimal mass-production of any natural enemy, it is essential to standardize the host’s age. Hymenopteran parasitoids are known to increase progeny production in response to rising host density (Legner 1967). The number of hosts parasitized per unit of time depends upon on the ability of the individual parasitoids to locate and parasitize a varying number of hosts. A number of parasitoids respond by increasing the number of hosts that each individual destroys (functional response) or respond to increased host density by increasing their own numbers (numerical response) (Solomon 1949). Fecundity, the total number of laid eggs, and fertility, the number of viable progeny, are variable features of an insect, influenced by a plethora of intrinsic and extrinsic factors (Panagiotis, Eliopoulos, and Stathas 2005). The evaluation of a natural enemy as a biological control agent requires a thorough study of the main effects and possible interactions of such factors on these characteristics (Jervis and Copland 1996). In the case of endoparasitoids, however, fecundity is relatively difficult to measure, because one or more eggs are laid inside the bodies of multiple Downloaded By: [Hendrichs, Jorge] At: 15:55 6 November 2009 hosts. Moreover, fertility is a more reliable criterion for evaluation, representing the net number of progeny after elimination of individuals that fail to complete development (Jervis and Copland 1996). Thus, it is pragmatic to study fertility rather than fecundity in endoparasitoids. Various developmental processes in insects are influenced by physical factors such as temperature (Beck 1980, 1983; Saunders 1982; Denlinger 1985; Ratte 1985). Progeny production and adult longevity are two of the major factors influencing the abundance of an insect and its population dynamics. Both have been studied in relation to temperature for many species of insects (Andrewartha and Birch 1954; Syme 1975, 1977; Harrison, King, and Ouzts 1985; Tingle and Copland 1988). Nuclear techniques, such as the use of ionizing radiation, could play a prime role in augmentative biological control, especially in facilitating the mass rearing of insects (Greany and Carpenter 2000). Several potential uses of nuclear techniques have been identified by researchers (Ramadan and Wong 1989; Sivinski and Smittle 1990; Greany and Carpenter 2000), including: (1) improvements in rearing media, Biocontrol Science and Technology 245

(2) provision of sterilized natural prey to be used as food during shipment, to ameliorate concerns relating to the incidental presence of ‘hitchhiking’ pests, (3) provision of supplemental food or hosts in the field, to increase the initial survival and buildup of released natural enemies, and (4) reproductive sterilization of weed- feeding insects that are candidates for biological control, for use in open field trials. The objective of the present study was to determine the effect of host age, density and temperature on the mass-production of N. thymus prior to successful implementation of biological control. The present study also was designed to determine the effects of gamma irradiation on the host for mass-production of N. thymus to determine the potential value of nuclear techniques in improved biological control.

Materials and methods Effect of host density and age Adults of E. sorbillans and N. thymus were collected from the sericultural area in the northern region of Bangladesh. They were maintained as stock cultures in the Department of Zoology, Rajshahi University. To determine the effect of host-age and density on the progeny production of N. thymus, a single male-female pair of 1-day- old parasitoids was provided with a single host puparium of either 1, 2, 3, 4, 5 or 6 days of age in separate glass vials (4 mL). Similarly, cohorts with 1:1, 1:2, 1:3, 1:4, 1:5, 1:6, 1:7, 1:8, 1:9 or 1:10 parasitoid to host ratios were maintained separately with 1Á2-day-old mated female parasitoids to observe the effect of host density on the progeny production of N. thymus. There were 10 observations for each ratio. The adult parasitoids were fed on 50% aqueous honey solution. The gender of progeny that emerged from the different ages and densities of parasitized puparia was determined. The rates of parasitism also were recorded.

Effect of temperature on adult parasitoid Both the Uzifly host pupae and N. thymus were collected from the stock culture. A single pair of 1-day-old parasitoids was provided with ten 2-day-old host pupae in

Downloaded By: [Hendrichs, Jorge] At: 15:55 6 November 2009 glass vials. The adult parasitoids were fed on 50% aqueous honey solution. They were kept separately in an incubator set at 20, 25 or 308C to observe the effect of temperature on progeny production. Parasitoid progeny were counted and separated by sex. The longevity of adult parasitoids and the rate of parasitism also were recorded. Each temperature experiment consisted of 20 observations. Multivariate statistics were used to determine the significance levels of the parameters considered in these experiments.

Effect of irradiated host In this investigation, both the Uzi pupae and adult N. thymus were collected from the stock culture. To assess the potential value of nuclear techniques in improving host suitability, two cohorts of early (2Á4-day-old) and late (6Á7-day-old) host puparia were selected for irradiation. The host pupae were irradiated with 0 (control), 0.5, 1, 2, 4 or 8 Gy for early pupae and 0 (control), 10, 30, 50, 70 or 90 Gy for late pupae. 246 M.M. Hasan et al.

The assessment of radiation doses against the host pupae was selected based on earlier work (Hasan and Khan 1998; Jahan, Rahman, Hasan, and Islam 1998). Single pairs of 1-day-old parasitoids were introduced into glass vials containing irradiated host pupae separately according to dose and pupal age. The adult parasitoids were fed on 50% aqueous honey solution. The progeny of parasitoids emerging from the irradiated host puparia were counted. There were 10 observations for each dose and age. The experiments were carried out for three successive generations. All the experiments were carried out at 28 8C and 65% RH.

Gamma irradiation techniques The radiation source was a deep-therapy unit of 60Co at the Bangladesh Atomic Energy Commission, Dhaka. The dose rate was approximately 0.78 Gy/min.

Statistical analyses

The intrinsic rate of natural increase (rm) expressed as the number progeny per individual per day was calculated employing the formula (Birch 1948):

Xv rmx e lxmx 1; x1

where, v is the age class, lx and mx are the proportion of surviving females at age x and the number of progeny produced per female at the age interval x, respectively. With a stable age distribution and under given factors, the intrinsic rate of natural increase is a useful comparative statistic of population growth potential (Southwood 1978).

In addition, gross reproductive rate, GRR/amx or total number of progeny produced per female during its lifetime (Price 1997), net reproductive rate, R0 / y ax0lxmx or number of progeny produced per female (Krebs 1994), finite capacity of rm increase, l/e or number of times the population will multiply itself per unit of time (Krebs 1994), mean generation time, TIn R0/rm (measured in days) and doubling time, DIn 2/rm (number of days required for the population to double its Downloaded By: [Hendrichs, Jorge] At: 15:55 6 November 2009 numbers) were calculated for factors (Southwood 1978). Multivariate statistics were used to determine the significance levels of the parameters considered in these experiments. All the statistical procedures were carried out using the software packages including Minitab (Version 9.0) and Excel (XP Microsoft Office).

Results Effect of host density and age All the parameters including host density, age, sex and their possible interactions showed significant (F5,5861 1119.69; F9,5861 651.70; F4,5861 504.17; F1,5861  3200.00, PB0.001) effects on the progeny production of N. thymus for all the host-age groups. There was a significant and positive relationship between progeny production and an increase in number of host puparia (Figure 1), and a significant (PB0.001) decline in the number of progeny with an increase in host age. Two to 4-day-old host puparia Biocontrol Science and Technology 247

Age-1 Age-2 Age-3 Age-4 Age-5 Age-6 350 300 250 200 150 100 Progeny (No.) 50 0 012345678910 Host density (No.)

Figure 1. Host-density dependent progeny production in N. thymus reared on different ages of host pupae (age in days).

were found to be the most suitable for obtaining the maximum progeny production of N. thymus for all the host-densities and parasitoid sex ratios (Figure 1). Maximum progeny production was found to be between 100 and 300 for 4-day-old host puparia (Figure 1). Maximum progeny production was obtained by providing 2Á4-day-old host puparia to all the pairs of parasitoids, while minimum progeny production was obtained by providing 5Á6-day-old host puparia (Figure 1). 1 The intrinsic rate of increase (rm day ) increased with the increased host density 1 for all the host-age groups (Figure 2). The results also show that the rm day values were relatively higher for the progeny production when the parasitoids were provided with 2Á3-day-old host puparia (Figure 2). This value was found to be greater for all the densities and age groups. The net reproductive rate (R0) and gross reproductive rate (GRR) were found to be greater for all the density levels and host age groups. On the other hand, the values for the doubling time (D) and finite capacity of increase (l) gradually decreased with increased host density for all the host age groups. The present findings show that the host density and age significantly (F9,5861  651.70 and F5,5861 1119.69; PB0.001) influenced the sex-ratio of progeny in Downloaded By: [Hendrichs, Jorge] At: 15:55 6 November 2009 Age 1 Age 2 Age 3 Age 4 Age 5 Age 6

0.85

) 0.8 -1 0.75 day m 0.7 0.65 0.6

Intricsic rate (r 0.55 0.5 012345678910 Host density (No.)

1 Figure 2. Host-density and age-dependent rm day values for the progeny production of N. thymus (age in days). 248 M.M. Hasan et al.

Table 1. Host-age and density effects upon male and female progeny production by N. thymus.

Host age (days)

1-day-old 2-day-old 3-day-old 4-day-old 5-day-old 6-day-old

Host Mean no. Mean no. Mean no. Mean no. Mean no. Mean no. Density1 ß:à ß:à ß:à ß:à ß:à ß:à

1:1 3.6:79.2 3.9:90.6 4.4:64.6 24.9:77.1 5.8:61.6 2.5:42.0 1:2 5.4:113.4 5.5:135.1 5.8:93.6 12.0:133.9 7.8:81.9 4.4:53.6 1:3 8.2:139.3 8.2:165.9 6.9:113.6 19.5:166.1 13.7:97.0 4.8:64.0 1:4 9.4:168.5 12.4:194.4 9.9:145.8 19.9:167.0 16.9:125.1 5.7:69.7 1:5 8.9:150.0 16.2:163.5 22.5:164.1 15.4:181.2 22.6:158.2 7.2:72.3 1:6 10.3:159.1 11.8:162.2 22.9:174.2 22.6:232.9 15.1:167.8 7.4:81.0 1:7 12.2:162.2 11.1:168.9 19.0:191.7 18.3:243.7 12.6:154.2 6.9:77.1 1:8 16.7:171.6 14.3:190.1 15.0:206.6 25.0:280.2 14.3:152.5 7.4:82.7 1:9 14.7:168.7 15.3:192.2 26.7:216.4 28.7:275.1 13.6:120.9 7.0:86.9 1:10 14.9:184.7 14.9:184.7 16.5:202.9 32.2:283.3 11.3:134.9 9.1:82.3

1Ratio parasitoids:host pupae.

N. thymus. Higher proportions of females were observed for all the host density levels and host-age groups (Table 1). It is interesting to note that the proportion of females gradually decreased as the host density increased for more or less all the age groups. The results show that the highest proportions of females were recorded on 2-day-old hosts while the lowest on 5-day-old hosts (Table 1). It also shows that all the female- biased sex ratios were significantly different from a 1:1 ratio. The data presented in Figure 3 indicate that the rate of parasitism clearly decreased with the increased host density and age; 100% parasitism occurred when the parasitoid was released against 1Á3-day-old age Uzi pupae. However, the results show that the rate of parasitism declined substantially with 6-day-old host pupae, where only 55% parasitism was observed at the host density number of 10. The results demonstrated that progeny production of the parasitoid varied significantly Downloaded By: [Hendrichs, Jorge] At: 15:55 6 November 2009 (F2,114 294.26; PB0.001) with temperature (Figure 4). The maximum number of

100

80 Host age 60 1 2 40 3 4 Parasitism (%) 20 5 6 0 12345678910 Host density (No.)

Figure 3. Host-density and age-dependent parasitism (%) for N. thymus (age in days). Biocontrol Science and Technology 249

250 230 210 190 170 150 130 Progeny (No.) 110 90 70 50 20 25 30 Temperature (°C)

Figure 4. Effect of temperature on the progeny production of parasitoid N. thymus (vertical bars indicate the standard error).

progeny (over 200) of N. thymus was recorded at 258C, while the minimum was at 308C (Figure 4). The highest values for the net reproductive rate (R0) and gross reproductive rate (GRR) for progeny production were recorded at 258C (Figure 5). The lowest values for these parameters were recorded at 208C. Both the values for doubling time of capacity (D) and finite capacity (l) increased at 25, 30 and 208C, respectively. The 1 highest value for the intrinsic rate of increase (rm day ) of progeny production was recorded at 258C, while the lowest value was at 208C (Figure 5). The sex-ratio of

N. thymus also varied significantly (F1,114 360.00; PB0.001) at different rearing temperatures. The significant variation was observed in the temperaturesex-ratio interaction (F2,114 258.34; PB0.001) (Figure 6). The sex ratios were female-biased for all the temperatures (Figure 6, Table 1). The highest percentage of female progeny

GRR rm

Downloaded By: [Hendrichs, Jorge] At: 15:55 6 November 2009 250 0.8 230

0.79 ) 1

210 _ 190 0.78 day m 170 0.77 150 GRR 130 0.76 110 0.75

90 Intrinsic rate (r 0.74 70 50 0.73 20 25 30 Temperature (°C)

1 Figure 5. Effect of temperature on the GRR and rm day values for the progeny production of parasitoid N. thymus. 250 M.M. Hasan et al.

94.80 94.60 94.40 94.20 94.00

Female (%) 93.80 93.60 93.40 20 25 30 Temperature (°C)

Figure 6. Effect of temperature on the sex ratio of the progeny of parasitoid N. thymus.

was obtained at 20 and 258C, while it decreased at 308C. The results of the test also indicate that the percentages for female-biased sex-ratios were more than 93.8 for all the temperatures (Figure 6). Temperature had a significant (PB0.001) effect on the longevity of adult N. thymus. The longevity of the adults decreased with increase of temperature for both sexes (Figure 7). The maximum longevity of adults was 11 and 18 days at 208C for males and females, respectively (Figure 7). The highest rate of parasitism was observed at 208C followed by 25 and 308C, and the average rate was more than 95% at all temperatures tested (Figure 8).

Effect of gamma irradiation on host Gamma irradiation of hosts increased the progeny production of N. thymus reared

either on early (2Á4-day-old) or late (6Á7-day-old) host puparia (early age, F1,676  3834.63, late age, F1,676  2399.47; PB0.001) The progeny production of N. thymus increased with increased doses of gamma radiation for both the early and late host puparia (F4.09; PB0.001) (Figures 9 and 10). A maximum of 272 progeny was produced at 8 Gy for a single pair of parasitoids reared on the early host puparia, while a minimum of 199 progeny was produced on the control batch (Figure 9). The

Downloaded By: [Hendrichs, Jorge] At: 15:55 6 November 2009 Male Female 20 18 16 14 12 10 8 Period days 6 4 2 0 20 25 30 Temperature (°C)

Figure 7. Effect of temperature on the longevity of adults of parasitoid N. thymus. Biocontrol Science and Technology 251

100

95

90

Parasitism (%) 85

80 20 25 30 Temperature (°C)

Figure 8. Effect of temperature on the parasitism percent of parasitoid N. thymus (line bars indicate the standard error).

present findings also show that the irradiated early host pupae were more suitable for mass production of N. thymus than the irradiated late pupae (Figures 9 and 10). There was a trend of significantly decreased progeny production in two successive generations compared to the control batch for both the host age groups (early age, F27.06, PB0.00001; late age, F6.02, PB0.01) (Figures 9 and 10). 1 The values R0, GRR and rm day for the progeny production increased with the increased gamma irradiation dose levels for both the host-age groups. However, the

F1 and F2 generations for early host age group did not follow the same trends in which the declining trend was observed at the moderate dose levels ranging from 2 to 4Gy. The sex ratios of N. thymus developing either from early- or late-irradiated Uzi pupae varied significantly, depending upon dose and developmental stage of the host

350 Parental F F

Downloaded By: [Hendrichs, Jorge] At: 15:55 6 November 2009 1 2 300

250

200

150

Progeny (No.) 100

50

0 0 0.5 1 2 4 8 Doses (Gy)

Figure 9. Progeny production of N. thymus developing from irradiated early (2Á4-day-old) host Uzi pupae (vertical bars indicate standard errors). 252 M.M. Hasan et al.

250 Parental F1 F2

200

150

100 Progeny (No.)

50

0 010 30 50 70 90 Doses (Gy)

Figure 10. Progeny production of N. thymus developing from irradiated late (6Á7-day-old) host Uzi pupae (vertical bars indicate the standard error).

pupae. The proportions of female progeny appeared to be a bit bimodal, with somewhat more females being produced at the lowest and highest irradiation doses

for all generations (parental, F1 and F2), but with a slightly lower proportion of female progeny at the intermediate doses (Figures 11 and 12). The maximum female proportion was produced at 8 and 50 Gy dose levels from the early- and late-Uzi pupae, respectively.

Discussion The number of hosts parasitized per unit of time depends upon the ability of the individual parasitoids to locate and parasitize varying numbers of hosts. The present

20 Parent F1 F2 18

Downloaded By: [Hendrichs, Jorge] At: 15:55 6 November 2009 16 14 12

10 8

Female:Male ratio 6 4

2 0 0 0.5 1 2 4 8 Doses (Gy)

Figure 11. Sex-ratio of N. thymus developing from irradiated early (2Á4-day-old) host Uzi pupae (vertical bars indicate the standard error). Biocontrol Science and Technology 253

20 Parent F F 18 1 2 16 14 12 10 8

Female:Male ratio 6 4 2 0 0 1030507090 Doses (Gy)

Figure 12. Sex-ratio of N. thymus developing from irradiated late (6Á7-day-old) host Uzi pupae (vertical bars indicate the standard error).

findings suggest that maximum progeny production could be obtained by providing irradiated 2Á4-day-old Uzi host to N. thymus. The results also showed that progeny production of N. thymus gradually increased with an increase in host density. These results are in agreement with Kishore, Sharma, Sharan, Sinhadeo, and Thangavelu (2001), who reported that the progeny production in N. thymus increased with the density of host puparia. Jalali, Singh, and Ballal (1987) also observed host density- dependent progeny production for Cotesia marginiventris (Cresson). Ulleyet (1949) in Gyptus sp., Puri and Sangwan (1973), and Utida (1950) observed the same trend while working with Neocatalaccus mamezoophagous and Bracon gelechiae Ashmead, respectively. The progeny production of the parasitoid was clearly affected by the host age, i.e., early stage (2Á4-day-old) host puparia were more suitable for mass production of the parasitoid compared to late stage puparia. These results corroborate the findings of Medeiros, Romalho, Lemos, and Zanuncio (2000) who reported that the reproduc- Downloaded By: [Hendrichs, Jorge] At: 15:55 6 November 2009 tive potential of Podisus nigrispinus (Dallas), a predator of cotton leafworm, decreased with an increase of age of the adults. Similar results were observed by Morgan, Smittle, and Patterson (1986), who determined that exposing 2-day-old Musca domestica L. pupae to Spalangia endius Walker females at a parasitoid to host density of 1:5 produced the greatest number of progeny. The present results also correlate well with those reported by Barclay (1986) while establishing a host- parasitoid dynamics model. Host density and host age significantly influenced the sex-ratio of progeny of N. thymus. The results are in agreement with the findings of Meunier and Bernstein (2002) that parasitoid sex-ratios change as a function of host and parasitoid densities and these changes could influence the dynamics of host-parasitoid systems. These studies have implications for augmentative biological control programs based upon mass rearing of N. thymus. It would appear to be most efficient from a production standpoint to use 2Á4-day-old parasitized host pupae for inundative release of the parasitoid. The study suggests that, although the functional response limits the 254 M.M. Hasan et al.

potential for N. thymus to generate density-dependent aggregation of parasitism, it may still be a promising candidate for biological control of E. sorbillans as well as other dipterans due to its aggregative response to host density. Additional studies are needed, however, to investigate the impact of this parasitoid on dipteran pest populations in natural conditions. It is evident from the results that there was significant (PB0.001) variation in progeny production of N. thymus when they were reared at different temperatures. Temperature-dependent progeny production in parasitoids has been reported by several researchers (Urbaneja, Llacer, Garrido, and Jacas 2001; Roy, Brodeur, and Cloutier 2003; Daane, Bentley, and Weber 2004). Hinton (1981) reported that there are few examples of insects whose reproductive potential is not significantly affected by temperature within the temperature range studied. Ratte (1985) mentioned that total egg production in insects reached a maximum at a temperature slightly lower than the optimum. Minkenberg (1989, 1990) found that the reproductive potential of Dacnusa sibirica Telenga, a parasitoid of Liriomyza spp., decreased from 225 to 48 eggs with increasing temperatures from 15 to 258C, whereas the fecundity of another leaf miner parasitoid, Diglyphus isaea (Walker), did not significantly change within the same temperature range (209Á293 eggs). The present results also agree with Taylor (1981) who reported that the optimum temperature for the development of Acyrthosiphon pisum (Harris) was 258C. He added that this temperature is consistent with that selected in the thermal gradient by non-parasitized aphids, i.e., 24.98C (temperature at DT0). He also reported that the maximum rate of adult development of Aphidius rapae (Curtis) occurred at 278C. A remarkably low number of progeny at 158C also was reported by Ahmad (1936) (2Á5 progeny/day/female). Pawson and Petersen (1990) observed a temperature preference for the oviposition behaviour of five species of pteromalid wasps using housefly pupae as the host. They reported that temperature had a significant effect on oviposition behaviour of Musidifurax raptor Girault and Sanders, Pachycrepoideus vindemmiae (Rondani), and Urolepis rufipes (Ashmead). Ryoo, Hong, and Yoo (1991) studied the reproductive potential of Lariophagus distinguendus (F.) in relation to temperature. Jahan and Islam (1998) measured the effects of different constant temperatures, viz. 18, 22, 26 and 308C on the development of Metaphycus helvolus (Compere), a Downloaded By: [Hendrichs, Jorge] At: 15:55 6 November 2009 parasitoid of L. They found that the reproductive potential of the parasitoid displayed a linear increase at all the constant temperatures ranging from 18 to 308C. 1 In the present experiment, analyses of rm day , R0 and GRR indicate that N. thymus was best adapted to temperatures of 258C (Figure 5). The parasitoid’s performance in terms of progeny production was poorest at 208C. The low value of l for N. thymus at 258C suggests that this temperature is near the lower limit of positive 1 population growth. Results also reveal that the intrinsic rate of increase (rm day ) did not vary greatly between 20 and 308C, suggesting that performance may not be seriously compromised at these temperatures. Roy et al. (2003) mentioned that the intrinsic rate of natural increase is useful to estimate the population growth potential of insects and mites, which may help predict the outcome of pest-natural enemy interactions. They also added that the developmental rate of spider mite prey 1 response to temperature has a major influence on the temperatureÁrm day relationship. The present results agree with the findings of Ren, Stansly, and Liv Biocontrol Science and Technology 255

(2002), who suggested that the magnitude of intrinsic rate of natural increase was greatly influenced by constant temperatures. Temperature had a significant (PB0.001) effect on the longevity of adult N. thymus. The longevity of the adults decreased with increase of temperatures for both the sexes. These findings are consistent with previous results for a number of parasitoids (Urbaneja et al. 2001; Bazzocchi, Lanzoni, Burgio, and Fiacconi 2003). There were female-biased sex-ratios at all the temperatures studied. Ren et al. (2002) reported similar results while working with the whitefly predator Nephaspis oculatus (Blatchley). The female-biased sex-ratios of parasitoids influenced by temperature have also been reported by several researchers (Murai 2000; Urbaneja et al. 2001; Chabi-Olaye, Fiaboe, and Schulthess 2004). Nevertheless, our findings indicate that N. thymus exhibits sufficient environmental plasticity to be a useful biological control agent against E. sorbillans as well as other dipteran pests under a wide range of temperatures. Nuclear techniques could also play an important role in augmentative biological control, not only in promoting mass rearing, but in several additional ways. It has been reported that ionizing radiation may be used to a great advantage to improve conventional in vivo rearing strategies for many parasitoids (Greany and Carpenter 1999). It has also been mentioned that the approach for prolonging the acceptability of house fly pupae would be to expose pupae to gamma irradiation and store them at cool temperatures (Morgan et al. 1986). Gamma irradiation of the host enhanced the progeny production of the parasitoid in their studies. This is consistent with the previous results obtained for tephritid fruit fly parasitoids by Hill (1997), and Ramadan and Wong (1989). Sivinski and Smittle (1990) reported that gamma radiation inhibited development (maturation) of the Caribbean fruit flies, Anastre- pha suspensa (Lowe), which are attacked by the parasitoid, Diachasmimorpha longicaudata (Ashmead). Ramadan and Wong (1989) exposed pupae of the oriental fruit fly, Dacus dorsalis (Hendel) to gamma radiation prior to eclosion of the parasitoid, D. longicaudata. Sivinski and Smittle (1990) found that gamma radiation from a 137cesium source prevented adult eclosion of non-parasitized Caribbean fruit flies, but it did not prevent the larvae from serving as viable hosts for D. longicaudata. This allowed the investigators to safely release A. suspensa puparia from larvae exposed to parasitoids without fear of releasing fertile flies into the area. Downloaded By: [Hendrichs, Jorge] At: 15:55 6 November 2009 The results of the present findings are also similar to the earlier studies of Morgan et al. (1986). They noted that the development of pupae of Musca domestica could be inhibited using gamma radiation (500 Gy) prior to mass culture for the parasitoid Spalangia endius Walker. Similar results also were obtained by Roth, Fincher, and Summerlin (1991), while working with irradiated horn fly pupae as hosts for hymenopteran parasitoids. Irradiated house fly pupae could be held successfully for an extended period (about 10 weeks) prior to parasitization. Carpenter, Bloem, and Hofmeyr (2004) found that the level of acceptability of eggs laid by irradiated false codling moth, Cryptophlebia leucotreta (Meyrick) (as hosts for Trichogrammatidae cryptophlebiae Nagaraja) was favourable for combined use of sterile insect techniques and augmentative releases of parasitoids. Vargas et al. (2004) mentioned that the effects of the sterile melon flies and the parasitoids were directed to the adult and larval stages, respectively, and would seem to be compatible from an IPM perspective, when multiple strategies are desirable. Irradiation of host larvae prior to parasitization is used in mass-rearing programmes in Florida, Mexico and 256 M.M. Hasan et al.

Guatemala to prevent mixed lots of other parasitoid spp. and fertile flies (Sivinski and Smittle 1990; Sivinski, Vulince, Menezes, and Aluja 1998). It should be noted that irradiation of hosts is not always helpful in parasitoid rearing. Menezes et al. (1998) studied the development of the parasitoid Coptera haywardi (Oglobin) in irradiated and unirradiated host pupae of Caribbean fruit fly and the Mediterranean fruit fly, Ceratitis capitata (Weidemann) and found that irradiated host pupae were not acceptable for this parasitoid, for unknown reasons. Another application for ionizing radiation that has promise is to inhibit the cellular and/or humoral (biochemical) defense reactions of host insects that might otherwise serve as optimal factitious hosts for beneficial insects. This approach was tested as a means of inhibiting encapsulation of the parasitoid Microplitis croceipes Cresson in a candidate factitious host, Galleria mellonella L. (Ferkovich unpub- lished, cf. Greany and Carpenter 1999). Recent studies by Genchev, Milcheva- Dimitrova, and Kozhuharova (2007) also showed that 65 Gy of gamma radiation enabled the otherwise marginally suitable factitious host Galleria mellonella L. to be used as a highly suitable host for the parasitoid Venturia canescens Grav. Although not relevant to pupal parasitoids, it has been reported that gamma radiation inhibits the behavioural resistance (e.g., defensive attack) of hosts so that they can be made more suitable for attack by parasitoids that may otherwise be injured by their hosts (Greany and Carpenter 1999). The foregoing discussion leads to the conclusion that ionizing radiation offers a reliable means to achieve developmental arrest of insect hosts for use in in vivo rearing prior to mass production of the parasitoid N. thymus. These findings will be further tested in an area-wide demonstration site at a sericultural farming area in the northern region of Bangladesh, where both N. thymus and sterile Uziflies will be released. It remains to be determined whether sterile flies and augmentative parasitoid releases will be cost-effective and sustainable in area wide IPM systems in the northern region of Bangladesh.

Acknowledgements This publication derives from a research project funded by the International Atomic Energy Agency, Vienna, under a research contract No. IAEA/BGD-10776. We would like to thank the Bangladesh Atomic Energy Agency for irradiation facilities and to Dr M.W. Gates, Downloaded By: [Hendrichs, Jorge] At: 15:55 6 November 2009 Department of Entomology, Smithsonian Institute, USA, for identifying the parasitoid. Thanks are due to Dr J. Hendrichs of FAO/IAEA, Vienna and Dr J.E. Carpenter of ARS- USDA, Tifton, GA, USA, for advice on experimental design and Dr Patrick D. Greany of PDG Consulting, Tallahassee, FL, USA, for critically reviewing the manuscript. We would also like to extend our thanks to the Chairman, Department of Zoology, Rajshahi University for providing the laboratory facilities.

References Ahmad, T. (1936), ‘The Infuence of Ecological Factors on the Mediterranean Flour Moth, Ephestia ku¨hniella, and its Parasite Nemeritis canescens’, Journal of Ecology,5, 67Á93. Andrewartha, H.G., and Birch, L.C. (1954), The Distribution and Abundance of Animals, Chicago, IL: University of Chicago Press, p. 782. Barclay, H.J. (1986), ‘Host-parasitoid Dynamics: Effect of the Position of Density Dependence’, Ecological Modeling, 32, 291Á299. Barlow, N.D., Goldson, S.L., and McHeill, M.A. (1994), ‘A Prospective Model for the Phenology of M. hyperodae (Hymenoptera: Braconidae), a Potential Biological Control Biocontrol Science and Technology 257

Agent of Argentine Stem Weevil in New Zealand’, Biocontrol Science and Technology,4, 375Á386. Bazzocchi, G.G., Lanzoni, A., Burgio, G., and Fiacconi, M.R. (2003), ‘Effects of Temperature and Host on the Pre-Imaginal Development of the Parasitoid Diglyphus isaea (Hymenop- tera: Eulophidae)’, Biological Control, 26, 74Á82. Beck, S.D. (1980), Insect Photoperiodism, (2nd ed.), New York: Academic Press. Beck, S.D. (1983), ‘Thermal and Thermoperiodic Effects on Larval Development and Diapause in the European Corn Borer Ostrinia nubilalis’, Journal of Insect Physiology, 29, 107Á112. Birch, L. (1948), ‘The Intrinsic Rate of Natural Increase of an Insect Population’, Journal of Animal Ecology, 17, 15Á26. Carpenter, J.E., Bloem, S., and Hofmeyr, J.H. (2004), ‘Acceptability and Suitability of Eggs of False Codling Moth (Lepidoptera: Tortricidae) from Irradiated Parents to Parasitism by Trichogrammatoidea cryptophlebiae (Hymenoptera: Trichogrammatidae)’, Biological Con- trol, 30, 351Á359. Chabi-Olaye, A., Fiaboe, M.K., and Schulthess, F. (2004), ‘Host Suitability and Thermal Requirements of Lathromeris ovicida Risbec (Hymenoptera: Trichogrammatidae), an Egg Parasitoid of Cereal Stemborers in Africa’, Biological Control, 30, 617Á623. Daane, K.M., Bentley, W.J., and Weber, E.A. (2004), ‘Vine Mealybug: A Formidable Pest Spreads Throughout California Vineyards’, Practical Winery Vineyard Magazine,3,35Á40. Denlinger, D.L. (1985), ‘Hormonal Control of Diapause’,inComprehensive Insect Physiology, Biochemistry and Pharmacology, eds. G.A. Kerkut and L.I. Gilbert Vol. VIII, Oxford: Pergamon Press, pp. 353Á412,. Genchev, N. Milcheva-Dimitrova, and Kozhuharova, M.V. (2007), ‘Use of Gamma Radiation for Suppression of Hemocytic Immune Response in Larvae of Galleria Mellonella (Lepidoptera) against Venturia canescens (Hymenoptera)’, Journal of Balkan Ecology, 10, 411Á419. Greany, P.D., and Carpenter, J.E. (1999), ‘Use of Nuclear Techniques in Biological Control of Insects and Weeds’, Nuclear News, 42, 32Á34. Greany, P.D., and Carpenter, J.E. (2000), ‘Use of Nuclear Techniques in Biological Control’,in Proceedings of Area-wide Control of Fruit Flies and other Pests, ed. K.H. Tan, Penang, Malaysia: Penerbit Universiti Sains Malaysia, pp. 221Á227,. Harrison, W.W., King, E.G., and Ouzts, J.D. (1985), ‘Development of Trichogramma exiguum and T. pretiosum at Five Temperature Regimes’, Environmental Entomology, 14, 118Á121. Hasan, M., and Khan, A.R. (1998), ‘Gamma Irradiation of the Uzi-Fly, Exorista sorbillans Wiedmann, an Endoparasitoid of the Silkworm, Bombyx mori L.’, International Pest Control, 40, 199Á201. Hill, G. (1997), The Potential for the Integration of Nuclear Techniques in Arthropod Biological Control. Report for IAEA Consultants Group on Use of Nuclear Techniques in Biological Downloaded By: [Hendrichs, Jorge] At: 15:55 6 November 2009 Control; held April 14Á18, Vienna. Hinton, H.E. (1981), Biology of Insect Eggs, Vol. I, Oxford, UK: Pergamon. Jahan, M., and Islam, K.S. (1998), ‘Effect of Temperature on the Development of Metaphycus helvolus (Compere), on Encyrtid Parasitoid of Brown Soft Scale’, Bangladesh Journal of Entomology,8,89Á96. Jahan, M.S., Rahman, S.M., Hasan, M.M., and Islam, M.S. (1998), ‘Effect of Gamma- Irradiation on Adult Longevity and Reproductive Potential of the Uzi Exorista bombycis (Louis) an Endoparasitoid of the Silkworm, Bombyx mori L.’, Sericologia, 38, 261Á266. Jalali, S.K., Singh, S.P., and Ballal, C.R. (1987), ‘Studies on Host Age Preference and Biology of Exotic Parasite, Cotesia marginiventris (Cresson) (Hymenoptera: Braconidae)’, Entomon, 12, 59Á62. Jervis, M.A., and Copland, M.J.W. (1996), ‘The Life Cycle’,inInsect Natural Enemies Á Practical Approaches to Their Study and Evaluation, eds. M.A. Jervis and N.A.C. Kidd, London: Chapman & Hall, pp. 63Á161. Kishore, R., Sharma, S.P., Sharan, S.K., Sinhadeo, S.N., and Thangavelu, K. (2001), ‘Optimal Conditions of Host-Parasitoid Density Ratio and Age for the Multiplication of Nesolynx thymus (Girault), a Parasitoid of the Uzifly, Blepharipa zebina Walker’, Sericologia, 41, 449Á454. 258 M.M. Hasan et al.

Krebs, C.J. (1994), Ecology: The Experimental Analysis of Distribution and Abundance (4th ed.), New York: Harper Collins College Publishers. Kumar, P., Kishore, R., Jayaprakas, C.A., and Sengupta, K. (1991), ‘Parasitoids of Uzi fly, Exorista sorbillans Wiedemann (Diptera: Tachinidae). XIII: Studies on the Efficiency of Nesolynx thymus (Girault) at the Field Level’, Indian Journal of Sericulture, 30, 161Á163. Legner, E.F. (1967), ‘Behaviour Changes in the Reproduction of Spalangia cameroni, S. endius, Muscidifurax raptor and Nasonia vitripennis (Hymenoptera: Pteromalidae) at Increasing Fly Host Densities’, Annals Entomological Society of America, 60, 819Á826. Medeiros, R.S., Romalho, F.S., Lemos, W.P., and Zanuncio, J.C. (2000), ‘Age-Dependent Fecundity and Life-Fertility Tables for Podisus nigrispinus (Dallas) (Heteroptera, Pentato- midae)’, Journal of Applied Entomology, 124, 319Á324. Menezes, E., Sivinski, J., Holler, T., Aluja, M., Jeronimo, F., and Ramirez, E. (1998), ‘Development of Coptera haywardi (Hymenoptera: Diapriidae) In Irradiated and Uni- rradiated Pupae of the Caribean Fruit Fly and Mediterranean Fruit Fly (Diptera: Tephritidae)’, Florida Entomologist, 81, 567Á570. Meunier, J., and Bernstein, C. (2002), ‘The Influence of Local Mate Competition on Host- Parasitoid dynamics’, Ecological Modeling, 152, 77Á88. Minkenberg, O.P.J.M. (1989), ‘Temperature Effects on the Life History of the Culophid Wasp Diglyphus isaea, an Ectoparasitoid of Leafminers (Liriomyza spp.) on Tomatoes’, Annual Applied Biology, 115, 381Á397. Minkenberg, O.P.J.M. (1990), ‘Reproduction of Dacnusa sibirica (Hymenoptera: Braconidae), an Endoparasitoid of the Leafminer Liriomyza bryoniae (Diptera: Agromyzidae), at Constant Temperatures’, Environmental Entomology, 19, 625Á629. Morgan, P.B., Smittle, B.J., and Patterson, R.S. (1986), ‘Use of Irradiated Pupae to Mass Culture the Micro-Hymenopteous Pupal Parasitoid Spalangia endius Walker (Hymenop- tera: Pteromalidae). I. Musca domestica L. (Diptera: Muscidae)’, Journal of Entomological Science, 21, 222Á226. Murai, T. (2000), ‘Effect of Temperature on Development and Reproduction of the Onion Thrips, Thrips tabaci Lindeman (Thysanoptera: Thripidae), on Pollen and Honey Solution’, Applied Entomology and Zoology, 35, 499Á504. Panagiotis, A., Eliopoulos, A., and Stathas, G.J. (2005), ‘Effects of Temperature, Host Instar, and Adult Feeding on Progeny Production by the Endoparasitoid Venturia canescens (Gravenhorst) (Hymenoptera: Ichneumonidae)’, Environmental Entomology, 34, 14Á21. Pawson, B.M., and Petersen, T.J. (1990), ‘Temperature Preference and Effect of Photoperiod on Oviposition Behaviour of Five Pteromalid Wasps (Hymenoptera: Pteromalidae) using House Fly (Diptera: Muscidae) Pupae as Hosts’, Environmental Entomology, 19, 1452Á1456. Price, P.W. (1997), Insect Ecology (3rd ed.), New York: John Wiley. Puri, S.N., and Sangwan, H.S. (1972), ‘Economics of Mass Rearing of Bracon gelechiae Ashmead’, Indian Journal of Entomology, 34, 232Á239. Downloaded By: [Hendrichs, Jorge] At: 15:55 6 November 2009 Ramadan, M.M., and Wong, T.T.Y. (1989), ‘Effect of Gamma Radiation on Biosteres longicaudatus (Ashmead) (Hymenotpera: Braconidae), a Larval Parasitoid of Dacus dorsalis (Hendel) (Diptera: Tephritidae)’, Proceedings of the Hawaiian Entomological Society, 29, 111Á113. Ratte, H.T. (1985), ‘Temperature and Insect Development’,inEnvironmental Physiology and Biochemistry of Insects, ed. K.H. Hoffman, Berlin: Springer-Verlag, pp. 33Á66. Ren, S., Stansly, P.A., and Liv, T. (2002), ‘Life History of the Whitefly Predator Nephaspis oculatus (Coleoptera: ) at Six Constant Temperatures’, Biological Control, 23, 262Á268. Roth, J.P., Fincher, G.T., and Summerlin, J.W. (1991), ‘Suitability of Irradiated or Freeze- Killed Hymenopteran Parasitoids’, Journal of Economic Entomology, 84, 94Á98. Roy, M., Brodeur, J., and Cloutier, C. (2003), ‘Effect of Temperature on Intrinsic Rates of 1 Natural Increase (rmDAY ) of a Coccinellid and its Spider Mite Prey’, Biological Control, 45, 57Á72. Ryoo, M.I., Hong, Y.S., and Yoo, C.K. (1991), ‘Relationship between Temperature and Development of Lariophagus distinguendus (Hymenoptera: Pteromalidae), an Ectoparasi- toid of Sitophilus oryzae (Coleoptera: Curculionidae)’, Journal of Economic Entomology, 84, 825Á829. Biocontrol Science and Technology 259

Saunders, D.S. (1982), Insect Clocks (2nd ed.), Oxford: Pergamon Press. Sivinski, J., and Smittle, B. (1990), ‘Effects of Gamma Radiation on the Development of the Caribbean Fruit Fly (Anastrepha suspensa) and the Subsequent Development of its Parasite Diachasmimorpha longicaudata’, Entomologia Experimentalis et Applicata, 55, 295Á297. Siviniski, J., Vulince, K., Menezes, E., and Aluja, M. (1998), ‘The Bionomics of Coptera haywardi (Oglobin) (Hymenoptera: Diapriidae) and other Pupal Parasitoids of Tephritid Fruit Flies (Diptera)’, Biological Control, 11, 193Á202. Solomon, M.E. (1949), ‘The Natural Control of Animal Populations,’ Journal of Animal Ecology, 18, 1Á35. Southwood, R. (1978), Ecological Methods with Particular Reference to the Study of Insect Populations, London: Chapman & Hall, p. 524. Syme, P.D. (1975), ‘The Effects of Flowers on the Longevity and Fecundity of Two Native Parasites of the European Pine Shoot Moth in Ontario’, Environmental Entomology,4, 337Á346. Syme, P.D. (1977), ‘Observations on the Longevity and Fecundity of Orgilus obscurator (Hymenoptera: Braconidae) and the Effect of Certain Foods on Longevity’, Canadian Entomologist, 109, 995Á1000. Taylor, F. (1981), ‘Ecology and Evolution of Physiological Time in Insects’, American Naturalist, 117, 1Á23. Tingle, C.C.D., and Copland, M.J.W. (1988), ‘Effects of Temperature and Host Plant on the Regulation of Glass House Mealybug (Hemiptera: Pseudococcidae) Population by Introduced Parasitoids (Hymenoptera: )’, Bulletin of Entomological Research, 78, 135Á142. Ulleyet, G.C. (1949), ‘Distribution of Progeny by Chelonus taxanus Crass’, Canadian Entomologist, 81, 25Á44. Urbaneja, A., Llacer, E., Garrido, A., and Jacas, J.A. (2001), ‘Effect of Temperature on the Life History of Cirrospilus sp. Near lyncus (Hymenoptera: Eulophidae), a Parasitoid of Phyllocnistis citrella (Lepidoptera: Gracillariidae)’, Biological Control, 21, 293Á299. Utida, S. (1950), ‘On the Equilibrium State of the Interacting Population of an Insect and its Parasite’, Ecology, 31, 165Á175. Vargas, R.I., Long, J., Miller, N.W., Delate, K., Jackson., C.G., Uchida, G.K., Bautista, R.C., and Harris, E.J. (2004), ‘Releases of Psyttalia fletcheri (Hymenoptera: Braconidae) and Sterile Flies to Suppress Melon Fly (Diptera: Tephritidae) in Hawaii’, Journal of Economic Entomology, 97, 1531Á1539. Vinson, S.B., and Iwantsch, G.F. (1980), ‘Host Suitability for Insect Parasitoids’, Annual Review of Entomology, 25, 397Á419. Downloaded By: [Hendrichs, Jorge] At: 15:55 6 November 2009 Biocontrol Science and Technology, Vol. 19, S1, 2009, 261Á270

Use of irradiated Musca domestica pupae to optimize mass rearing and commercial shipment of the parasitoid Spalangia endius (Hymenoptera: Pteromalidae) Miguel C. Zapatera*, Carlos E. Andiarenaa, Gladys Perez Camargoa, and Norberto Bartolonib

aCa´tedra de Gene´tica, Facultad de Agronomı´a, Universidad de Buenos Aires, Av. San Martı´n 4453 (1417), Buenos Aires, Argentina; bCa´tedra de Me´todos Cuantitativos Aplicados, Facultad de Agronomı´a, Universidad de Buenos Aires, Av. San Martı´n 4453 (1417), Buenos Aires, Argentina

This paper examines the potential for using irradiated Musca domestica pupae as suitable hosts of the parasitoid Spalangia endius for its use in biological control programs. Prior to being exposed to parasitoids, M. domestica pupae were gamma irradiated at 500 Gy and maintained for up to 2 months in anoxia at 68C. The parasitization percentage, estimated by parasitoid emergence, decreased 25% after 26.5 days, 50% after 53.2 days, and 58% after 60 days. This was compared to a control group of S. endius parasitoids reared on cold-stored non-irradiated pupae whose emergence percentage decreased by 25% after 7.7 days, 50% after 15.5 days, and 72% after 22 days. Fecundity and adult longevity of parasitoids emerging from irradiated pupae were evaluated as indicators of fitness. There were no significant differences in fitness between parasitoids raised on irradiated, cold- stored pupae and the standard, live pupae presently being used in biocontrol programs. If this procedure is implemented for the mass rearing process of S. endius, it could allow the production of surplus stocks of pupae, improved efficiency, reduced rearing costs, and allow commercial shipments of non- parasitized host pupae. Keywords: gamma radiation; Musca domestica; Spalangia endius; mass-rearing; parasitism; host stockpiling; pupal storage Downloaded By: [Hendrichs, Jorge] At: 15:57 6 November 2009 Introduction The house fly, Musca domestica L. (Diptera: Muscidae), is an important pest worldwide and breeds in places where food, warmth, and moisture are present. This situation is common in places with intensive animal production, around manure and produce composting areas, and near dumps and industrial landfill sites. Urbanization close to these facilities has resulted in a gradual lowering of the tolerance threshold for nuisance flies. Unfortunately, the close proximity of humans and animals to these facilities also often makes it difficult to apply chemicals for fly control. In addition, because of frequent applications, which are necessary for effective control, insecticide resistance has become a serious problem (Meyer, Georghiou, and Hawley 1987; Scott, Roush, and Rutz 1989). Contamination and intoxication are also frequent problems on many farms. As a result, biological control using pupal parasitoids within the

*Corresponding author. Email: [email protected]

ISSN 0958-3157 print/ISSN 1360-0478 online # 2009 Taylor & Francis DOI: 10.1080/09583150802439819 http://www.informaworld.com 262 M.C. Zapater et al.

framework of an integrated management system can provide an economical and efficient alternative (Zapater, Mart´nez-Rey,ı and Mazzoli 1994). Spalangia spp. (Hymenoptera: Pteromalidae) are pupal parasitoids of certain fly species distributed around the world (Boucek 1963). Natural levels of parasitism due to these wasps are generally low. Meyer, Mullens, Cyr, and Stokes (1990) reported that 4% and 6.2% of stable flies and house flies, respectively, were being parasitized by Spalangia spp. in California dairies. Natural levels of parasitism of house fly pupae in installations of caged chickens have been reported at 0.5% (Rutz and Axtell 1981) and 7.6% (Rueda and Axtell 1985). Parasitism levels of stable flies in feedlots are only around 0.1% (Smith, Hall, and Thomas 1987). However, inundative releases of Spalangia endius Walker against M. domestica have been shown to increase parasitism levels to 80Á90% with considerable control; in some instances parasitism rates even reached 100% (Morgan, Weidhass, and Patterson 1981). In caged-layer poultry houses, Zapater (1997) demonstrated that weekly releases of four S. endius per chicken combined with adequate manure management reduced fly populations 13.2 times compared to those facilities without parasitoid releases. Once the use of house fly parasitoids was considered to be innocuous to the environment, seven species were released on Easter Island, which has a fragile ecosystem (Ripa 1980, 1986). Spalangia endius and other muscid fly parasitoids are presently commercia- lized in North America (Hunter 1994), Colombia (Vergara-Ru´zı 1996), and Argentina (Zapater et al. 1994). Techniques for mass rearing S. endius have been reported by Morgan and Patterson (1978), Morgan, LaBrecque, and Patterson (1978) and Morgan (1981). One of the most important issues for a commercial insectary is the ability to guarantee customers regular (e.g. weekly) shipments of parasitoids. In order to deal with normal variations in daily pupal production and occasional urgent or increased demands for parasitoids, more host pupae are generally produced than are needed. Because the optimal age of house fly pupae for parasitism by S. endius is 24Á72-h-old, it is not possible to stockpile host pupae for any length of time or purchase non- parasitized pupae from another insectary. The short period of time that host pupae are suitable for parasitism further complicates insectaryoperations where different species/ strains of parasitoids are being reared in different facilities to avoid contamination problems. Another issue for commercial insectaries is that not all of the pupae that are Downloaded By: [Hendrichs, Jorge] At: 15:57 6 November 2009 exposed to parasitoids are parasitized. Fortunately, adult flies emerge from non- parasitized pupae before the parasitoids begin to emerge. Therefore, to insure that only parasitized host material is shipped and released, exposed pupae must be held for an additional 6 days to allow adult flies and empty puparia to be eliminated. A potential strategy to address the above mentioned problems is to try to prolong the suitability of house fly pupae for parasitism, and prevent the emergence of adult flies from non-parasitized pupae using a combination of gamma radiation, anoxia, and cold. Morgan, Smittle, and Patterson (1986) showed that no house fly adults emerged from pupae irradiated (gamma radiation) at a dose of 500 Gy, and that pupae irradiated at this dose were still good hosts for S. endius. They were also able to extend the suitability of pupae for parasitism to 8 weeks by storing them at 4.48C and adequate humidity. In this paper, we extend this research and further describe the effect on S. endius fitness of irradiating house fly pupae and placing them in cold storage in anoxia for up to 2 months before using them as host material under mass rearing conditions. Biocontrol Science and Technology 263

Materials and methods Insect strains Musca domestica strain A 20-year-old house fly laboratory stock originating from the Institute for Pesticide Research, Wageningen, The Nertherlands, was provided to the University of Buenos Aires in 1999. In August 2000, the colony was invigorated by out-crossing virgin laboratory females with wild males collected from a poultry house in Mendoza, Argentina. The resulting offspring from these crosses has been used as the parental stock in all succeeding experiments. A colony of around 30,000 adults is permanently maintained. The average number of pupae in 10 mL is 325924.

Spalangia endius strain A colony of S. endius was established in March 2000 with wild insects collected from three different areas in Argentina: Mi Granja, Co´rdoba province; La Plata, Buenos Aires province; and Pergamino, Buenos Aires province. Parasitoids were collected from poultry facilities by the placement and retrieval of mesh bags containing laboratory-reared house fly pupae. Pupae were removed from the bags and held in Plexiglas cages for fly emergence. After adult flies and empty puparia had been eliminated, the remaining pupae were held until parasitoids emerged. Parasitoids are presently maintained on house fly pupae under mass rearing conditions at a weekly production level of about 100,000 adults.

Irradiation Irradiation of the house fly pupae was conducted at ‘IONICS‘, Ingenieros 2475, El Talar, Buenos Aires, Argentina, a commercial irradiation facility, using a Cobalt60 irradiator with an activity level of (1942.510 13 Bq (525,000 Ci). Because of the high dose rate, special procedures adjusting the exposure distance were developed by the staff at ‘IONICS‘ to ensure that an effective dose of only 500 Gy was delivered at a dose rate of 20 Gy/min. The dose was calculated such that no fly emergence was observed from material irradiated at 500 Gy, while increasing fly emergence was Downloaded By: [Hendrichs, Jorge] At: 15:57 6 November 2009 detected at doses of 400 and 300 Gy similar to that reported by Morgan et al. (1986).

Experiment I: storage potential of irradiated pupae Two hundred mL of 48-h-old (912 h) M. domestica pupae were placed in each one of 54 plastic bags. The bags were hermetically sealed, which caused anoxia to develop due to respiration of the pupae, and then irradiated with 500 Gy of gamma radiation. The bags of pupae were placed in a cooler before and after transportation to the irradiation facility. Later, the bags were maintained permanently at 690.58Cina refrigerator. Every day for the first eight days (days 0Á7) and every third day for days 9Á63 (total of 27 sampling days), one sample of 50 pupae was extracted from each of two bags and the bags discarded. The pupae were placed in small plastic Petri dishes (20.5 cm, diameterhigh) and introduced into the parasitization cage containing 25Á30 pairs of parasitoids. The cage was 406040 cm (longwidehigh), made of Plexiglas with a fine mesh screening on two sides for ventilation. An approximate 264 M.C. Zapater et al.

ratio of one female parasitoid for every four pupae was maintained in the cage and parasitoids were replaced every 2 days. The parasitization cage was kept in an environmental chamber maintained at 25918C, 14 h L:10 h D, and 70Á85% RH. After 2 days, the pupae were removed and placed in small plastic tubes (13 cm, diameterlong) covered with mesh to allow for parasitoid emergence. The tubes with parasitized pupae were maintained as above. A total of nine repetitions were conducted. In addition, two bags of non-irradiated pupae were prepared and exposed to parasitoids to compare the percentage of non-irradiated pupae parasitized on day 0 with that of the irradiated pupae. Percent parasitism was calculated as the number of emerged parasitoids, divided by the number of exposed pupae, 100.

Experiment II: storage potential of non-irradiated pupae Similar to Experiment I, plastic bags were prepared consisting of 200 mL of 48-h-old (912 h) M. domestica pupae, but in this case they were not irradiated. The bags were again maintained at 690.58C in a refrigerator until needed. Every day during 22 days beginning on day 0 (23 treatments), one sample of 50 pupae was extracted from each of two bags and placed into Petri dishes and the bags discarded. The Petri dishes were then introduced into similar parasitization cages as in Experiment I and exposed to S. endius for 2 days. As before, parasitized pupae were placed in small plastic tubes to allow parasitoid emergence for determination of the percent parasitism. Fourteen repetitions were performed consisting of 46 bags of pupae prepared for each repetition. In addition to the samples taken each day to assess the suitability of the pupae for parasitism, a second sample of 50 pupae was taken from each bag (one sample from each of two bags for days 0Á22) to assess fly emergence. Each sample was placed in a separate Petri dish. The Petri dishes were then placed within a Plexiglas box 1010 cm and 6 cm high that had a 3 cm hole in the top covered with fine mesh for ventilation and maintained at 25918C, 14 h L:10 h D, and 70Á85% RH. After 30 days, the numbers of fully emerged flies, half emerged flies, and dead pupae were counted.

Experiment III: fitness of parasitoids reared on irradiated pupae held in cold storage Downloaded By: [Hendrichs, Jorge] At: 15:57 6 November 2009 in anoxia Two simple tests were carried out to evaluate the fecundity and longevity of parasitoid females using house fly pupae that had been irradiated and then held in cold storage and anoxia for 1, 20 and 40 days as in Experiment I. Pupae from the normal colony production that were 48 h old (912 h) and had not been subjected to irradiation and cold storage were used for the control treatment.

Fecundity Fecundity was evaluated by placing 400 pupae from each of the four treatments into separate 3 cm Petri dishes and then keeping the dishes in the parasitization cage containing approximately one female parasitoid for every four pupae. The dishes were removed after 48 h and placed in separate 10106 cm Plexiglas boxes kept in an environmental chamber 25918C, 14 h L:10 h D, 70Á85% RH to allow parasitoid emergence. Newly emerged parasitoids (B24 h old) were removed and placed in Biocontrol Science and Technology 265

additional 10106 cm boxes with excess 24-h-old pupae to allow the parasitoids to host feed and mate for 48 h. Ten females from each of the four treatments were then selected and placed in individual 13 cm plastic tubes containing 25 normal colony pupae (48-h-old). The pupae were removed and 25 new pupae added every 24 h for 4 days. Exposed pupae were held separately in plastic tubes in the environmental chamber to allow parasitoid emergence and to determine the number of progeny produced per female per day. A total of three replicates were performed.

Longevity Longevity was calculated by selecting 15 newly emerged females from each treatment and placing them in individual 13 cm tubes containing five normal colony pupae (24-h-old). The host pupae were removed every 4 days and replaced with new pupae until the female died. Dead insects were counted daily and recorded. Three repetitions were done for each treatment.

Statistical analyses Data were analyzed through linear regression analysis (Neter, Kutner, Nachtsheim, and Wasserman 1996). Least square estimates were obtained for the relationship among response variables and predictive variables, and a measure of how much variance in the response variable was explained by the independent variables (R2). In Experiment II, we also employed a curvilinear regression (exponential). In the analysis of Experiment III, we employed a non-parametric procedure: the KruskalÁ Wallis test ANOVA by ranks (Conover 1980).

Results Experiment I: storage potential of irradiated pupae The effect of storing house fly pupae in anoxia in the cold for increasing lengths of time on percent parasitism by S. endius is presented in Figure 1. Results indicated that there was a decrease in the parasitism rate as the length of pupal storage time Downloaded By: [Hendrichs, Jorge] At: 15:57 6 November 2009

Figure 1. Mean percent parasitism by Spalangia endius on irradiated Musca domestica pupae that had been held in cold storage (690.58C) and anoxia for up to 63 days. N9 replications. 266 M.C. Zapater et al.

Figure 2. Mean percent parasitism by Spalangia endius on non-irradiated Musca domestica pupae that had been held in cold storage (690.58C) and anoxia for up to 22 days. N14 replications. increased following the formula Y53.93 Á 0.50X. The R2 value for this equation was 0.51. In accordance with the equation, the rate of parasitism percentage decreased 25% after 26.5 days, 50% after 53.2 days and 58% near the end of the experiment 60 days later. This experiment was carried out under simulated mass rearing conditions, which probably accounts for the resulting variation in parasitism rates.

Experiment II: storage potential of non-irradiated pupae In Experiment II, using non-irradiated pupae, the rate of parasitism (based on S. endius emergence) decreased much more rapidly than in irradiated pupae (Figure 2). The initial parasitism values were similar for both experiments, but the parasitism rate began decreasing much more rapidly with time, showing a linear regression between the average % parasitism and the number of days pupae were stored by: Y 54.35 Á 1.76X. This function had an R2 0.54. In accordance with the regression formula, rates of parasitism decreased 25% after 7.7 days, 50% after 15.5 days and 72% at the end of the experiment at 22 days. In the second part of Experiment II, emergence (full emergence and half Downloaded By: [Hendrichs, Jorge] At: 15:57 6 November 2009 emergence) of adult flies from non-irradiated pupae remained high following 2Á3 days of storage, but decreased rapidly after the following curvilinear equation Y1/ 0.01e7.030.42X (Figure 3). A 25% reduction in adult fly emergence was observed after 4 days, 50% after 6.2 days, 75% after 12.5 days and a 99% reduction in fly emergence was seen after 16.5 days of pupal storage. The correlation between S. endius emergence and fly emergence was calculated and resulted in a coefficient of linear correlation, r0.91. The high correlation suggests that S. endius prefers to parasitize live or recently dead fly pupae.

Experiment III: fitness of parasitoids reared on irradiated pupae held in cold storage and anoxia Fecundity The results of the fecundity experiment comparing the number of progeny produced per female reared from normal (control) pupae versus females reared from irradiated Biocontrol Science and Technology 267

Figure 3. Mean percent adult emergence (pupae that either fully or partially emerged as adults) of non-irradiated Musca domestica pupae that had been held in cold storage and anoxia for up to 22 days. N 14 replications.

pupae held in cold storage for 1, 20 or 40 days are presented in Figure 4. A KruskalÁ Wallis ANOVA test for ranks was employed and no significant differences in fecundity were found among females from the different treatments (df3, N179, H1.2792, P0.1692). Females produced an average of 12Á14 offspring on day 3, which decreased to 5Á6onday6.

Longevity The longevity of adult female parasitoids emerging from normal (control) pupae and pupae that were stored in anoxia and cold for 1, 20 or 40 days was monitored for 16 days and cumulative daily averages plotted in Figure 5. A KruskalÁWallis ANOVA test for ranks was applied and no significant differences among the four treatments Downloaded By: [Hendrichs, Jorge] At: 15:57 6 November 2009

Figure 4. Mean daily adult progeny production (fecundity) of Spalangia endius females emerged from Musca domestica pupae that had been irradiated and refrigerated under anoxia (conserved) for 1, 20, and 40 days and from non-irradiated/non-refrigerated (control) pupae. Newly emerged females were allowed to host feed and mate for 2 days and then were exposed to 25 fresh pupae each day for 4 days. N3 replications. 268 M.C. Zapater et al.

Figure 5. Mean cumulative percent mortality of Spalangia endius females emerged from Musca domestica pupae that had been irradiated and refrigerated under anoxia (conserved) for 1, 20 and 40 days and from non-irradiated/non-refrigerated (control) pupae. Each female was provided five normal pupae that were replaced every 4 days until the female died. N3 replications. were discovered (df3, N179, H0.7899, PB0.8519). Fifty percent of the females had died by day 8 and 100% by day 16.

Discussion Our tests confirmed that 500 Gy irradiation of house fly pupae prevents adult emergence. Results from our experiments also indicated that the combined treatment of irradiation, anoxia and refrigeration can extend the suitability of house fly pupae for parasitism by S. endius to 30 days or more. For example, when irradiated pupae were used, the rate of parasitism based on progeny production decreased by 50% from about 60 to 30% after 53.2 days of pupal storage; when non-irradiated pupae were used, 50% fewer parasitoids were produced after only 15.5 days. And although the percentage of parasitized host pupae that produced viable adult parasiotids gradually declined with increased storage time, the parasitoids that were produced in this manner from pupae stored up to 40 days were of good quality and lived as long Downloaded By: [Hendrichs, Jorge] At: 15:57 6 November 2009 and produced as many offspring as parasitoids reared from normal pupae. Thus, from a commercial standpoint, it should be possible to guarantee customers a specified number of quality adult parasitoids by appropriately adjusting the number of parasitized pupae that are sent depending on how long the pupae had been stored. The use of irradiated host material for mass rearing parasitoids has a number of advantages. First, in the case of S. endius rearing, developing house fly pupae produce a significant amount of biological heat. As a result, pupae must be well spread out on trays when they are exposed to the parasitoids. However, if the pupae are irradiated, development is stopped and more pupae could be placed per unit area. Similar concerns about the negative impact of metabolic heat, once parasitized pupae have been packaged for shipment, would also be minimized. Second, because not all of the exposed pupae are parasitized, current protocols require that the pupae must be held for four days to eliminate any flies that emerge before they can be shipped to customers. This costs both time and space. If irradiated pupae were used, parasitized pupae could be shipped sooner, which would free-up holding space and Biocontrol Science and Technology 269

create more flexibility in the system for orders to be prepared, shipments to arrive, and parasitoids to be delivered to the field. The use of irradiated host material has already become standard practice in the mass rearing of fruit fly parasitoids in support of sterile insect release programs (Sivinski and Smittle 1990; Cancino, Ru´ız, Go´mez, and Toledo 2002; see also articles by Cancino and Hepdurgun, this issue). The ability to store/stockpile pupae for a period of time has a number of additional advantages. It would allow facility managers to better plan for the exact number of parasitoids needed and avoid the tendency to overproduce both host pupae and parasitoids. With proper management, a rotating stockpile of host pupae could be maintained such that when parasitoid demand was low, excess host material produced during that time could be put into storage. If parasitoid demand increased unexpectedly or an opportunity arose to develop new clients, pupae could be brought out of storage to quickly meet the need. The ability to store pupae could also reduce work, for example, during weekends and holidays. Contamination of parasitoid strains can be a significant problem when rearing multiple species. If colony host pupae are parasitized by an unintended species prior to collection of the pupae for exposure to the intended parasitoid species (in this case S. endius), it is easy to contaminate stocks. However, if the host pupae are collected and then irradiated before exposure to a particular parasitoid, the window of opportunity for contamination by a competing parasitoid species is greatly reduced and much easier to manage. The use of gamma irradiation would also make it possible for insectaries to trade/sell irradiated host pupae amongst themselves instead of parasitized pupae, which occasionally occurs when demand exceeds production or problems with a colony develop. This would allow them to continue to provide their customers with the same strain they normally provide, rather than having to supplement an order with a strain from another facility. The large commercial irradiator used in this study had a greater degree of dose error at low doses than most traditional Gammacell irradiators. Morgan et al. (1986) showed that a dose of 500 Gy was optimal for preventing fly emergence with 3-day- old pupae. However, for commercial mass-rearing, non-optimal doses between 250 and 750 Gy could probably be occasionally tolerated. Unfortunately, the primary limitation to the wide scale application of this technology is the availability of an irradiation source. Alternatives to irradiation, such as X-ray machines or linear Downloaded By: [Hendrichs, Jorge] At: 15:57 6 November 2009 accelerators, that would be more accessible, at more reasonable prices, are needed.

Acknowledgements Special thanks to Ionics S.A. for the pupal irradiation and technical advice. The authors would also like to thank K. Bloem for helpful edits to earlier drafts of this manuscript. This work was funded by the Joint FAO/IAEA Division of Nuclear Techniques in Food and Agriculture Section, Vienna, Austria, under research contract No 10844/RO.

References Boucek, Z. (1963), ‘A Taxonomic Study in Spalangia (Hymenoptera: Chalcidoidea)’, Acta Entomologicae Museum Pragae, 35, 429Á512. Conover, W.J. (1980), Practical Nonparametric Statistics. Texas Tech University, New York: John Wiley and Sons. Cancino, J., Ruiz, L., Go´mez, Y., and Toledo, J. (2002), ‘Irradiacio´n de larvas de Anastrepha ludens (Loew) (Diptera:Tephritidae) para inhibir la emergencia de moscas en la cr´ıa del 270 M.C. Zapater et al.

parasitoide Diachasmimorpha longicaudata (Ashmead) (Hymenoptera: Braconidae)’, Folia Entomolo´gica Mexicana, 41, 195Á208. Hunter, C.D. (1994), Suppliers of Beneficial Organisms in North America, Sacramento, CA: California Environmental Protection Agency, pp. 30. Meyer, J.A., Georghiou, G.P., and Hawley, M.K. (1987), ‘House Fly Resistance to Permethrin in Southern California Dairies’, Journal of Economic Entomology, 66, 711Á713. Meyer, J.A., Mullens, B.A., Cyr, T.L., and Stokes, C. (1990), ‘Commercially and Naturally Occurring Fly Parasitoids (Hymenoptera: Pteromalidae) as Biological Control Agents of Stable Flies and House Flies on California Dairies’, Journal of Economic Entomology, 83, 799Á806. Morgan, P.B. (1981), ‘Mass Production of Spalangia endius Walker for Augmentative and/or Inoculative Releases’,inStatus of Biological Control of Filth Flies, ed. R.S. Patterson, P.G. Koehler, P.B. Morgan and R.L. Harris, Proc. Workshop, Feb. 4Á5, 1981, Gainsville, FL: USDA, ARS, pp. 185Á188. Morgan, P.B., and Patterson, R.S. (1978), ‘Culturing Microhymenopteran Pupal Parasitoids of Muscoid Flies’,inFacilities for Insect Research and Production, eds., N.C. Leppla and T.R. Ashley, USDA Tech. Bull., pp. 32Á33. Morgan, P.B., LaBrecque, G.C., and Patterson, R.S. (1978), ‘Mass Culturing the Micro- hymenopteran Parasite Spalangia endius Walker’, Journal of Medical Entomology, 14, 671Á 673. Morgan, P.B., Weidhass, D., and Patterson, R.S. (1981), ‘Programmed Releases of Spalangia endius and Muscidifurax raptor Against Estimated Populations of Musca domestica’, Journal of Medical Entomology, 18, 158Á166. Morgan, P.B., Smittle, B.J., and Patterson, R.S. (1986), ‘Use of Irradiated Pupae to Mass Culture the Microhymenopterous Pupal Parasitoid Spalangia endius Walker (Hymenoptera: Pteromalidae). I. Musca domestica L. (Diptera:Muscidae)’, Journal of Entomological Sciences, 21, 222Á227. Neter, J., Kutner, M., Nachtsheim, C., and Wasserman, W. (1996), Applied Linear Statistical Models, New York: WCB-McGraw-Hill. Ripa, R. (1980), ‘Biological Control of Muscoid Flies on Easter Islands’,inBiocontrol of Artropods Affecting Livestock and Poultry, eds. D. Rutz and R. Patterson, Boulder, CO: Westiew Press, pp. 111Á119. Ripa, R. (1986), ‘Survey and Use of Biological Control Agents on Easter Islands and in Chile’, in Biological Control of Muscoid Flies, eds. R. Patterson and D. Rutz, USA: Miscellaneous publication of the Entomological Society of America, Vol. 61, pp. 39Á44. Rueda, L.M., and Axtell, R. (1985), ‘Comparison of Hymenopterous Parasites of House Fly, Musca domestica, Pupae in Different Production Systems’, Environmental Entomology, 14, 217Á222. Rutz, D., and Axtell, R. (1981), ‘House Fly Control in Broiler-Breeder Poultry Houses by

Downloaded By: [Hendrichs, Jorge] At: 15:57 6 November 2009 Pupal Parasites: Indigenous Parasite Species and Releases of Muscidifurax raptor’, Environmental Entomology, 10, 343Á345. Scott, J.G., Roush, R.T., and Rutz, D.A. (1989), ‘Insecticide Resistance of House Flies from New York Dairies (Diptera: Muscidae)’, Journal of Agricultural Entomology,6,53Á64. Sivinski, J., and Smittle, B. (1990), ‘Effect of Gamma Radiation on the Development of the Caribbean Fruit Fly Anastrepha suspensa, and the Subsequent Development of its Parasite Diachasmimorpha longicaudata (Ashmead)’, Entomologia Experimentalis et Applicata, 55, 295Á297. Smith, J., Hall, R., and Thomas, G. (1987), ‘Field Parasitism of the Stable Fly’, Annals of the Entomological Society of America, 80, 391Á397. Vergara-Ruiz, R. (1996), ‘Manejo integrado de moscas comunes en explotaciones pecuarias en Colombia’,inEl Control Biolo´gico en Ame´rica Latina, ed. M.C. Zapater, Buenos Aires: IOBC-NTRS, pp. 115Á124. Zapater, M.C. (1997), ‘Control biolo´gico de moscas en ‘‘El Peligro’’ 07Á08/97’, Capia Informa, 169, 3Á5. Zapater, M.C., Martinez-Rey, C.E., and Mazzoli, J.A. (1994), ‘Control de moscas en establecimientos agropecuarios de cr´ıa intensiva: primer logro del control biolo´gico inundativo en la Argentina’, Gaceta Agrono´mica, 82, 360Á375. Biocontrol Science and Technology, Vol. 19, S1, 2009, 271Á275

RESEARCH ARTICLE Synergism between biological control and sterile insect technique: Can commercial mass production of biocontrol agents and sterile insects be integrated within the same industrial entity? Shimon Steinberg* and Jean-Pierre Cayol

Bio-Fly Sde Eliyahu, Bet Shean Valley 10810, Israel

The integration of commercial facilities for mass production of beneficial arthropods (Bio-Bee) and sterile insects (Bio-Fly) within the same industrial entity in Israel has proven successful. The synergism between the two companies has resulted in the integration of nuclear techniques and the use of biocontrol agents in area-wide integrated pest management programmes. Keywords: SIT; biological control; private sector; host irradiation; Ceratitis capitata; mass rearing

Introduction In the last few decades, Israel has been a pioneer and a leader in the development and use of biological control techniques for the environment-friendly management of insect pests in agriculture. For the past 25 years, Bio-Bee Sde Eliyahu Ltd. has been mass producing and using arthropod natural enemies for biological pest control. Predatory mites, pirate bugs, parasitic wasps and ladybeetles are used in commercial/ augmentative biological control to combat spider mites, whiteflies, thrips, aphids, leafminers and mealybugs in protected as well as open field crops. Responding to a demand from the plant protection organisations and the fruit industry in Israel, the Hashemite Kingdom of Jordan and the Palestinian Territories, Bio-Fly, a daughter company of Bio-Bee, was established in 2004 to mass rear the

Downloaded By: [Hendrichs, Jorge] At: 15:59 6 November 2009 Mediterranean fruit fly Ceratitis capitata (Wiedemann) (Diptera: Tephritidae) for Sterile Insect Technique (SIT) purposes (Bassi, Steinberg, and Cayol 2007). The use of natural enemies and the SIT both rely on the mass production and augmentative release of arthropods in the field to control a given target pest. As the establishment of Bio-Fly benefited from the experience of Bio-Bee in mass rearing procedures, the expertise and know-how that is continuously gained at Bio-Fly’s mass rearing facility now offers unique opportunities to synergise between SIT and the biological control industry, along three major lines.

The use of SIT ‘by-products’ for the production of natural enemies One of the limiting factors in mass production of beneficial arthropods such as predatory insects and mites as well as parasitoids, is the cost of mass producing

*Corresponding author. Email: [email protected]

First Published Online 26 June 2009 ISSN 0958-3157 print/ISSN 1360-0478 online # 2009 Taylor & Francis DOI: 10.1080/09583150902884971 http://www.informaworld.com 272 S. Steinberg and J.-P. Cayol

another arthropod species to be used as prey (for predatory insects and mites) or as a host (for parasitoids). Very often, the resulting commercial price of beneficial arthropods is restricting the use of commercial/augmentative biocontrol to high added-value crops. The cheaper the overall production cost of beneficial arthropods, the more likely they can be utilized by the agricultural community on a large scale in crops that bear marginal profit. The availability of large numbers of Mediterranean fruit fly eggs, larvae and pupae, as ‘by-products’ of Bio-Fly’s SIT activities, has provided several opportunities to improve the mass production of predators and parasitoids by Bio-Bee.

Predator mass rearing The minute pirate bug Orius laevigatus (Fieber) (Heteroptera: Anthocoridae) is a highly effective predator of Western flower thrips Frankliniella occidentalis (pergande) in protected vegetables, mainly sweet pepper, and is considered to be of key importance in the commercial augmentative biocontrol programmes in protected sweet pepper throughout Europe. Frozen eggs of the Mediterranean flour moth Ephestia kuehniella Zeller represent the most common diet for mass rearing of O. laevigatus, as well as for other Orius species and other predators such as Mirid bugs, ladybeetles and lacewings (e.g., Schmidt, Richards, Nadel, and Ferguson 1995; Cocuzza et al. 1997; Yano, Watanabe, and Yara 2002). The Ephestia eggs are considered to be the most expensive component of all other input materials used for industrial mass rearing of O. laevigatus. Hence, cheaper alternative food sources are needed. A Mediterranean fruit fly larvae-based diet proved to be as efficient as Ephestia eggs or as immature stages of the greenhouse whitefly Trialeurodes vaporariorum (Westwood) (Homoptera: Aleurodidae) for the production of the Mirid bug Macrolophus caliginosus Wagner (Heteroptera: Miridae) (Nannini, Ruiu, and Floris 2008; Nannini, Foddi, Murgia, Pisci, and Sanna 2008). A thorough study jointly conducted by Bio-Bee/Bio-Fly has shown that specially-treated Mediterranean fruit fly developmental stages were comparable to Ephestia eggs with respect to their effect on O. laevigatus over-all fitness, i.e., developmental time, fecundity and field performance. Downloaded By: [Hendrichs, Jorge] At: 15:59 6 November 2009 Parasitoid mass rearing Spalangia cameroni Perkins (Hymenoptera: Pteromalidae) is a solitary pupal parasitoid of flies, especially those living in manure, the house fly Musca domestica Linnaeus (Diptera: Muscidae) being one of its most important hosts. As generalist parasitoids, Spalangia spp. are known to attack Mediterranean fruit fly pupae (Stibick 2004). Furthermore, Geden and Kaufman (2007) have shown that progeny production of S. cameroni from house fly pupae killed by gamma radiation was not significantly different from production on live hosts. It was therefore decided to investigate whether gamma irradiated male pupae of the VIENNA-8 genetic sexing strain of the Mediterranean fruit fly (Franz 2005), could serve as an appropriate host for mass rearing of S. cameroni. By using irradiated Mediterranean fruit fly male pupae, it was confirmed that no adult flies emerged from parasitized or not- parasitized (yet irradiated) hosts. This allows the commercial production of S. cameroni at no risk from an agricultural perspective. As a result, the production Biocontrol Science and Technology 273

of S. cameroni is taking place on gamma irradiated young male pupae of the Mediterranean fruit fly. In further R&D efforts, fitness parameters of the S. cameroni colony such as rate of parasitism, progeny production, offspring sex ratio and long-term genetic deterioration will be compared between different hosts such as Mediterranean fruit fly irradiated male pupae, house fly pupae and pupae of the green-bottle fly Lucilia caesar Linnaeus (Diptera: Calliphoridae). In case of future fitness decrease (though not yet detected) in S. cameroni, which would be caused by rearing on a relatively small factitious host, other Pteromalid wasps such as Muscidifurax spp. could be evaluated.

Integration of nuclear techniques into biological control procedures Kaspi and Parrella (2003) have shown that release of sterile (gamma-irradiated) serpentine leafminer Liriomyza trifolii (Burgess) (Diptera: Agromyzidae) can significantly reduce the reproductive capacity of a wild population. Their study indicated that sterilization of L. trifolii flies is feasible and that sterile males are of high quality and competitive with normal males. In a subsequent study, Kaspi and Parrella (2008) demonstrated a synergistic interaction between releases of the leafminer larval parasitoid Diglyphus isaea (Walker) (Hymenoptera: Eulophidae), a well-known natural enemy of leafminers world-wide, and sterile males of L. trifolii. Bio-Bee produces the celery miner fly, Liriomyza bryoniae (Kaltenbach) (Diptera: Agromyzidae) and its highly effective parasitoid D. isaea. Hence, a similar evaluation of this type of synergistic interaction will be carried out. Bio-Bee is a leading producer of predatory mites world-wide. Based upon the successes mentioned above and the expertise gained by both teams with regards to gamma irradiation, use of sub-sterilizing irradiation doses and traditional selection to induce mutations for developing strains of predatory mites which are better adapted to specific conditions such as tomato plants that are known to be less favourable to natural enemies due to presence of various glandular hairs (or trichomes) on the plant is being considered. In the long-run, with the support and expertise of Bio-Bee in the production of natural enemies, Bio-Fly is considering the production of tephritid fruit fly Downloaded By: [Hendrichs, Jorge] At: 15:59 6 November 2009 parasitoids using Mediterranean fruit fly as a host and combine augmentative releases of parasitoids and sterile males of the Mediterranean fruit fly. Potential candidates for biological control of the Mediterranean fruit fly are the following parasitic wasps: Diachasmimorpha kraussii (Fullaway) (Hymenoptera: Braconidae), Fopius arisanus (Sonan) (Hymenoptera: Braconidae), F. ceratitivorus (Wharton) (Hymenoptera: Braconidae) and Aganaspis daci (Weld) (Hymenoptera: Eucoilidae). These species are currently being tested at the Israel Cohen Biological Control Institute of the Citrus Growers Board (Yael Argov, personal communication).

Synergism in the implementation of area-wide integrated pest management programmes Traditionally, most of the natural enemies produced by Bio-Bee are being used in protected environments such as greenhouses. When applied in open fields, given the close biocontrol agent-pest-host-plant relationship, the release strategy for the 274 S. Steinberg and J.-P. Cayol

natural enemies is traditionally based on a field-by-field approach, i.e., strictly limited to the fields planted with the crop(s) to be protected. Through its involvement in SIT field operations, Bio-Fly has acquired and integrated its expertise on area- wide integrated pest management (AW-IPM) (Hendrichs, Kenmore, Robinson, and Vreysen 2007) into the ‘culture’ of Bio-Bee Sde Eliyahu Ltd. Natural enemies and sterile insects, both being non-chemical alternatives for pest control, are, by essence, complementary technologies that involve augmentative releases. Following several years of damage caused by the Mediterranean fruit fly and following the recent ban of malathion, the major producers of table grapes in Israel are planning an AW-IPM programme with an SIT component to control the Mediterranean fruit fly. Another major pest of table grapes in Israel is the vine mealybug Planococcus ficus (Signoret) (Hemiptera: Pseudococcidae). The vine mealybug will be controlled biologically by seasonal inoculative releases of the lady beetle Cryptolaemus montrouzieri Mulsant (Coleoptera: Coccinelidae) and parasitic wasp Anagyrus near pseudococci (Hymenoptera: Encyrtidae), as part of an overall AW-IPM programme integrating biological control and SIT against the two major pests.

Conclusion The experience of Bio-Bee/Bio-Fly illustrates that commercial mass production of biocontrol agents and sterile insects can be integrated within the same industrial entity, benefiting from their mutual experience. In the early days of the establishment of Bio-Fly, the expertise of Bio-Bee in mass rearing arthropods and automated control of artificial rearing conditions was essential to the establishment of Bio-Fly and critical for its success in the first months. As illustrated above, the integration of nuclear techniques within the biological control operations of Bio-Bee has been done in a progressive manner, from using sterilised by-products provided by Bio-Fly to developing specific uses for nuclear techniques in biological control. This progressive approach was important as it resulted in the sustainable adoption (and adaptation) of the nuclear techniques to the specific problems faced by the biological control company and the successful integration of biological control and SIT within a single AW-IPM programme in table grapes. To date, there is no doubt that the contribution Downloaded By: [Hendrichs, Jorge] At: 15:59 6 November 2009 of nuclear techniques to several critical aspects of the biological control led by Bio- Bee is invaluable and is here to stay. The fact that two technical and commercial entities, one that mass produces and releases arthropod natural enemies for biological pest control and the other that mass produces and applies tephritid sterile males on an area-wide basis for SIT purposes, exist under the same managerial roof, offers a unique opportunity to synergise between the two. Bio-Bee/Bio-Fly provides an ideal environment to turn this synergy into a reality.

References Bassi, Y., Steinberg, S., and Cayol, J.P. (2007), ‘Private Sector Investment in Mediterranean Fruit Fly Mass-production and SIT Operations Á The ‘Sheep‘ of the Private Sector among the ‘Wolves‘ of the Public Good?’,inArea-wide Control of Insect Pests, from Research to Field Implementation, eds. M.J.B. Vreysen, A.S. Robinson, and J. Hendrichs, The Netherlands: Springer, AA Dordrecht, pp. 457Á471. Biocontrol Science and Technology 275

Cocuzza, G.E., DeClercq, P., van de Veire, M., DeCock, A., Degheele, D., and Vacante, V. (1997), ‘Reproduction of Orius laevigatus and Orius albidipennis on Pollen and Ephestia kuehniella Eggs’, Entomologia Experimentalis et Applicata, 82, 101Á104. Franz, G. (2005), ‘Genetic Sexing Strains in Mediterranean Fruit Fly, an Example for other Species Amenable to Large-scale Rearing for the Sterile Insect Technique’,inSterile Insect Technique: Principles and Practice in Area-wide Integrated Pest Management, eds. V.A. Dyck, J. Hendrichs, and A.S. Robinson, The Netherlands: Springer, AA Dordrecht, pp. 427Á451. Geden, C.J., and Kaufman, P.E. (2007), ‘Development of Spalangia cameroni and Muscidi- furax raptor (Hymenoptera: Pteromalidae) on Live House Fly (Diptera: Muscidae) Pupae and Pupae Killed by Heat Shock, Irradiation, and Cold’, Environmental Entomology, 36, 34Á39. Hendrichs, J., Kenmore, P., Robinson, A.S., and Vreysen, M.J.B. (2007), ‘Area-wide Integrated Pest Management (AW-IPM): Principles, Practice and Prospects’,inArea-wide Control of Insect Pests, from Research to Field Implementation, eds. M.J.B. Vreysen, A.S. Robinson, and J. Hendrichs, The Netherlands: Springer, AA Dordrecht, pp. 3Á33. Kaspi, R., and Parrella, M.P. (2003), ‘The Feasibility of Using the Sterile Insect Technique against Liriomyza trifolii (Diptera: Agromyzidae) Infesting Greenhouse Chrysanthemum’, Annals of Applied Biology, 143, 25Á34. Kaspi, R., and Parrella, M.P. (2008), ‘Synergistic Interaction between Parasitoids and Sterile Insects’. Integrated Control in Protected Crops. Temperate Climate, IOBC/WPRS Bulletin, 32, 99Á102. Nannini, M., Ruiu, L., and Floris, I. (2008), ‘Ceratitis capitata Larvae as an Alternative Food Source for Macrolophus caliginosus’. Integrated Control in Protected Crops. Temperate Climate, IOBC/WPRS Bulletin, 32, 147Á150. Nannini, M., Foddi, F., Murgia, G., Pisci, R., and Sanna, F. (2008), ‘A Novel Use of Ceratitis capitata for Biological Control Programs’. Integrated Control in Protected Crops. Temperate Climate, IOBC/WPRS Bulletin, 32, 151Á154. Schmidt, J.M., Richards, P.C., Nadel, H., and Ferguson, G. (1995), ‘A Rearing Method for the Production of Large Numbers of the Insidious Flower Bug, Orius insidiosus (Say) (Hemiptera: Anthocoridae)’, Canadian Entomologist, 127, 445Á447. Stibick, J.N.L. (2004), Natural Enemies of True Fruit Flies (Tephritidae). USDA APHIS PPQ 86 pp. Yano, E., Watanabe, K., and Yara, K. (2002), ‘Life History Parameters of Orius sauteri (Poppius) (Het., Anthocoridae) reared on Ephestia kuehniella Eggs and Minimum Amount of the Diet for Rearing Individuals’, Journal of Applied Entomology, 126, 389Á394. Downloaded By: [Hendrichs, Jorge] At: 15:59 6 November 2009 Biocontrol Science and Technology, Vol. 19, S1, 2009, 277Á290

RESEARCH ARTICLE Enhancing biological control of sugarcane shoot borer, Chilo infuscatellus (Lepidoptera: Pyralidae), through use of radiation to improve laboratory rearing and field augmentation of egg and larval parasitoids Bilquis Fatima*, Nazir Ahmad, Raza Muhammad Memon, Moula Bux, and Qadeer Ahmad

Nuclear Institute of Agriculture, Tando Jam-70060, Pakistan

Feasibility studies were conducted to evaluate the use of gamma radiation to improve the production and field performance of the egg parasitoid Trichogramma chilonis Ishii and to improve production of the gregarious larval endoparasitoid Cotesia flavipes Cameron for biological control of the sugarcane shoot borer, Chilo infuscatellus Snellen. Sitotroga cerealella (Olivier) eggs were used as hosts for T. chilonis and the suitability of non-irradiated host eggs decreased as the age of the eggs increased, with no success in parasitization of eggs older than 4 days of age. However, irradiation of host eggs using 20Á25 Gray (Gy) decreased the age effect and significantly more 2-, 3-, 4- and 6-day-old irradiated eggs were successfully parasitized than non-irradiated eggs. Radiation doses of 20 and 25 Gy were most effective for economical production of T. chilonis. Irradiation of host eggs did not affect the hatch percentage up to a dose of 15 Gy. Hatchability was significantly reduced at higher doses, with negligible hatching at 50 Gy. Irradiation also skewed the sex ratio of S. cerealella in favor of males at higher doses. Radiation at 60Á80 Gy improved the suitability of C. infuscatellus larvae for parasitism by C. flavipes, allowing normally unsuitable fourth and fifth instar larvae of C. infuscatellus to be successfully parasitized. The sex ratio of parasitoids reared on irradiated larvae was skewed in favor of females. Irradiation also slowed immature development of C. flavipes and the combination of irradiation and low temperature (108C) proved effective for prolonged storage of the parasitoids. Pupae of C. flavipes irradiated at

Downloaded By: [Hendrichs, Jorge] At: 15:59 6 November 2009 20 Gy could be stored for 2 months at 108 C without apparent loss of quality and deferred emergence by 29Á30 days. Provisioning with irradiated supplemental hosts in the field increased overall pest suppression. The infestation by C. infuscatellus was higher in the control sugarcane block, where it remained above the economic threshold level (10% infestation) from April to October. High temperatures and low relative humidity during May to July reduced increases of the parasitoid populations. These findings were used to facilitate the area-wide application of biocontrol agents in a 40,000 hectare area to suppress sugarcane borers to sub- economic levels (B10% infestation). Sugarcane borer damage was 5.9% in treated blocks vs. 19.2% in untreated control blocks. Keywords: radiation; augmentative biological control; sugarcane shoot borer; Chilo infuscatellus; Trichogramma chilonis; Cotesia flavipes; parasitoids; rearing; area-wide; biological management; endoparasitoid; supplemental hosts

*Corresponding author: Email: [email protected]

ISSN 0958-3157 print/ISSN 1360-0478 online # 2009 Taylor & Francis DOI: 10.1080/09583150902793438 http://www.informaworld.com 278 B. Fatima et al.

Introduction Interest has increased in using biological control, which has long been recognized as an important tool in suppressing insect pests. Beneficial insects have been successfully deployed in a variety of augmentation and conservation strategies (Nordlund 1984; Mohyuddin 1991; Ashraf, Fatima, Hussain, and Ahmad 1999). For example, the use of Trichogramma wasps as biocontrol agents is a recognized alternative to use of insecticides and has been applied successfully for the management of many insect pests (Browning and Melton 1987; Hassan, Kohler, and Rost 1988; Losey, Fleischer, Calvin, Harkness, and Leahy 1995; Mannion, Carpenter, and Gross 1995). Similarly, the larval parasitoid Cotesia flavipes Cameron, in conjunction with Trichogramma chilonis Ishii, is used to control the Chilo infuscatellus Snellen population in India (Suasa-Ard and Charernsom 1999; Saikia and Nath 2002; Tanwar and Ashok 2002). Augmentation usually involves periodic releases of beneficial insects and environmental management, such as providing food or hosts during times of low host density. However, economical production of natural enemies is a prerequisite for augmentative biological control programs. Considerable technological advances have been made in mass rearing of parasitoids and predators for augmentative biological control (Leppla, Bloem, and Luck 2002; Cohen 2003). Nuclear techniques may play an important role in augmentative biological control, not only in facilitating mass rearing, but also by inhibiting reproduction by reproductively sterilizing insect hosts, by provisioning non-reproductive supplemental hosts for use in field insectaries, by expediting the safe transport of parasitoids in irradiated hosts, by improving host suitability for mass rearing, and by microbially sterilizing artificial rearing media (Greany and Carpenter 2000). Irradiation has improved the suitability of certain lepidopterous hosts for parasitization (Mannion, Carpenter, and Gross 1994; Carpenter, Mannion, and Hidrayani 1995; Carpenter 1996). Marston and Ertle (1969), tested the acceptability of irradiated moth eggs to Trichogramma minutum Riley, and rep- orted that irradiated eggs were as suitable as control eggs for parasitoid deve- lopment. Eggs of the Mediterranean flour moth, Ephestia kuehniella Zeller, killed

Downloaded By: [Hendrichs, Jorge] At: 15:59 6 November 2009 with ultraviolet irradiation were suitable for mass rearing of Trichogramma spp. (Voegele, Daumal, Brun, and Onillon 1974). Lewis and Young (1972) reported that adult males of Helicoverpa (Heliothis) zea (Boddie) sterilized with 320 Gy of gamma irradiation paired with untreated females, produced sterile eggs that were as suitable as fertile eggs for attack and development of Trichogramma evanescens Westwood. The studies reported herein were designed specifically to evaluate the feasibility of: (1) using radiation-sterilized host eggs for efficient and economical production and prolonged storage and release of the egg parasitoid, T. chilonis, a polyphagous parasitoid of many lepidopteran pests, (2) using gamma radiation to improve production of the larval parasitoid C. flavipes to enhance control of sugarcane borers in the field, and (3) providing irradiated supplemental hosts for use as a field insectary to enhance the initial survival and population build-up of T. chilonis to achieve more cost-effective management of C. infuscatellus. These findings were used to implement an area-wide control effort in a large area (ca. 40,000 ha). Biocontrol Science and Technology 279

Materials and methods Tests on host egg suitability as affected by radiation Eggs of the Angumoiis grain moth, Sitotroga cerealella (OIivier), were obtained from moths reared in the laboratory at 25928C in 2.5 L glass jars. For this purpose, large numbers of 1Á2-day-old adults were collected from stock cultures and placed in inverted 1 L plastic jars with screen bottoms. Eggs that fell through the screen bottoms were collected by sifting them through a 70 mesh screen. Newly laid S. cerealella eggs were irradiated with doses ranging from 5 to 50 Gy with 60Co source at the Nuclear Institute of Agriculture, Tando Jam (Theratronic T-780 with an emission rate of 175 cGy/min and GWXJ Á 80 with an emission rate of 225 cGy/ min) and were kept in separate Petri dishes at 100 eggs/dish. The whole experiment was replicated four times and the hatch percentage in all the doses was recorded. Egg cards were prepared by sprinkling ca. 2000 eggs on cards coated with Stickum†, which were then allowed to air dry prior to exposure to parasitoids. The eggs were exposed to radiation after 1 h at different doses from 5 to 50 Gy with five Gy intervals followed by their exposure to the parasitoid T. chilonis in conical flasks using eggs of different ages (1Á7 days). Twenty pairs of the parasitoid were exposed to each card for 24 h. Egg cards were removed from the flasks after 24 h and rate of parasitism on each card was recorded. Parasitization per card was recorded, as indicated by darkening of the host eggs. The effect of egg age and radiation dose on parasitization was determined by counting the number of parasitoids produced vs. number of host eggs exposed.

Seasonal population fluctuation of T. chilonis as a function of temperature and humidity Seasonal population fluctuations of T. chilonis were determined in sugarcane fields by exposing two fresh egg cards of the factitious host S. cerealella per 0.404 ha, for 24 h at weekly intervals. The cards were returned to the laboratory and parasitization was determined by noting the characteristic darkening. The maximum/minimum temperatures and relative humidity were recorded in the field. The experiment continued throughout the crop-growing season. Downloaded By: [Hendrichs, Jorge] At: 15:59 6 November 2009

Providing supplemental irradiated host eggs to parasitoids for use in a field insectary Tests were conducted to determine whether provision of irradiated supplemental hosts could influence the initial survival and establishment of released parasitoids in the field. Three sugarcane blocks were selected at the same site. Releases of T. chilonis were made in the first block at monthly intervals and no supplemental hosts were provided to the parasitoids in the field. In the second block, fresh eggs of S. cerealella irradiated at 25 Gy were attached to sugarcane leaves at the time of parasitoid releases. Ten cards (ca. 2000 eggs/card) per 0.404 ha were attached to the leaves of the sugarcane crop in conjunction with five cards of parasitoids. All cards were spaced at uniform distances. The cards with supplemental irradiated eggs were provided twice during the first and second parasitoid releases. The third block was kept as a control and no plant protection measures were adopted. To record the establishment of T. chilonis parasitoids, fresh S. cerealella egg cards were placed for 280 B. Fatima et al.

24 h at two cards per 0.404 ha in each block at 2-week intervals. The cards were brought to the laboratory and the percent parasitism was recorded. The borer infestation was recorded at 2-week intervals.

Parasitization of irradiated C. infuscatellus larvae by C. flavipes The impact of irradiation on the larvae of C. infuscatellus was evaluated as a means of enhancing parasitism by C. flavipes. Larvae of C. infuscatellus were collected from the field and reared on natural diet to make a homogenous stock. Different host instars (second, third, fourth and fifth) were irradiated at 20Á120 Gy and then exposed to C. flavipes. Parasitization was recorded for each instar and compared with untreated larvae of each stage of development. The numbers of cocoons recovered and the percentage of adult parasitoid emergence were recorded from each instar.

Use of irradiation and low temperatures for prolonged storage of C. flavipes The sugarcane-adapted strain of C. flavipes used for these experiments originated in Thailand; a laboratory strain was established at the Nuclear Institute of Agriculture, Tando Jam and was subsequently field colonized. Third, fourth and fifth instars (100 larvae for each replicate) of C. infuscatellus were exposed to the C. flavipes parasitoids for 24 h at ambient temperature and humidity (24928C, 5Á60% R.H). After two and eight days of parasitization, C. flavipes were given different doses of gamma rays ranging from 5 to 50 Gy to irradiate the eggs and mature larvae of the parasitoids, respectively. Mature cocoons also were given the same treatment. The irradiated, parasitized larvae along with diet were kept in the incubators set at 10, 15, 20 and 258C for prolonged storage. The larvae were examined regularly to record emergence of the parasitoids for cocoon formation and pupation. Cocoons containing pupae were confined in glass tubes plugged with cotton and kept in the same incubator where parasitoid larvae completed their development. Upon parasitoid emergence, pupal survival and duration was recorded.

Evaluation of supplemental irradiated host eggs for use as a field insectary Downloaded By: [Hendrichs, Jorge] At: 15:59 6 November 2009 This study was conducted at the Nuclear Institute of Agriculture Experimental Farm. Three 1.62-ha sugarcane blocks were selected and each block was ca. 100 m from the next block Sugarcane variety BL4 (Saccharum sp. hybrid) was used in all tests. The first block was treated with cards of the egg parasitoid, T. chilonis, at monthly intervals from February to October and with no supplemental hosts. The second block was treated with egg parasitoids and supplemental irradiated hosts. For this purpose, newly laid S. cerealella eggs irradiated at 25 Gy were glued on paper cards with ca. 2000 eggs/card and the cards were attached to sugarcane leaves in addition to the parasitoid cards. Ten cards per 0.404 ha were attached to the leaves of the sugarcane crop in conjunction with five cards with ca. 2000 parasitized eggs at uniform distances. Supplemental irradiated cards were provided twice, with the first and second release of the parasitoids. The third block was kept as a control and no plant protection measures were adopted. Each month, the establishment of the parasitoids was monitored by placing two fresh S. cerealella egg cards per 0.404 ha in each block for 24 h. The cards were brought to the laboratory and the parasitism Biocontrol Science and Technology 281

percentage was recorded at 2-week intervals by examination of the appearance of the eggs; darkening connoted parasitization as the parasitized eggs became dark after 3 days whereas, un-parasitized remained yellow. The borer infestation also was recorded at 2-week intervals on an internodal basis (total internodes and number of infested internodes were counted).

Scale-up to area-wide management T. chilonis parasitoids that were mass-reared on irradiated S. cerealella eggs were released at monthly intervals from February to September over a target area of ca. 40,000 ha during the 2002Á2003 seasons. The infestation of sugarcane stem borers was recorded at monthly intervals from the entire treated area. Two sugarcane blocks were selected at each of two different 1.62 ha sites. Releases of T. chilonis were made in the first block at monthly intervals throughout the growing season, i.e., February to September, and the second block was left untreated. To evaluate parasitoid efficacy, newly laid sentinel irradiated S. cerealella eggs were placed in the field, recovered, and evaluated as above. Feral eggs also were collected from both the blocks at the same time and the parasitization percentage was recorded. The sugarcane stem borer infestation was also recorded during the same 2-week evaluation intervals.

Statistical analyses Statistical analyses were conducted using Statistix† Version 8.1, Analytical Software, Inc., Tallahassee, FL, USA.

Results Part I: rearing improvement studies Effect of radiation on host egg hatchability and suitability for parasitization by T. chilonis Radiation significantly reduced the hatch percentage of the irradiated, non-parasitized host eggs on a dose-dependent basis and it was negligible at 50 Gy (Table 1). Emergence Downloaded By: [Hendrichs, Jorge] At: 15:59 6 November 2009 of adult S. cerealella was the same as controls up to a dosage of 25 Gy, and 0 when exposed to 50 Gy.The sex ratio of moths emerging from the irradiated eggs was skewed in favor of males at higher doses, which may be attributed to the fact that females are more radiosensitive than males and some females may have died in the embryonic stage. Parasitoids preferred newly laid eggs and host suitability decreased as host age increased (Table 2). Radiation of host eggs decreased the age effect and significantly (P50.05) more eggs were parasitized on 2-, 3- and 4-day-old irradiated host eggs as compared to the normal eggs. Irradiated eggs were successfully parasitized up to 6 days old, while control eggs older than 4 days were unsuitable as hosts.

Parasitization of irradiated C. infuscatellus larvae by C. flavipes Fourth and fifth instar non-irradiated host larvae were not successfully parasitized, suggesting immunity in the host. However, fourth and fifth instar host larvae were successfully parasitized at high rates when treated with at least 60 Gy of radiation 282 B. Fatima et al.

Table 1. Effect of gamma radiation on the eggs of S. cerealella.

Sex ratio (%)

Dose Pupae Adult (Gy) Hatch% recovered% emergence% Male Female

5 88.693.43A 64.493.64BC 65.495.85AB 57.293.11CDE 42.891.92B 10 80.096.28B 62.293.67C 64.695.85AB 55.094.30DE 45.092.0AB 15 81.691.94B 66.693.43AB 63.092.23AB 64.293.56BCDE 35.092.86C 20 76.497.60BC 51.493.2D 66.292.38A 73.695.98BC 26.492.30D 25 72.492.88CD 50.492.96D 65.693.13A 72.895.31BCD 27.293.70D 30 68.094.58D 40.494.03E 59.694.98BC 75.097.1BC 25.093.39D 35 40.492.96E 27.691.81F 68.895.76A 73.694.03BC 26.492.40D 40 19.695.07F 12.492.40G 56.895.11CD 100.090.00A 0.090.0E 45 3.691.14G 2.090.70H 50.095.11D 100.090.00B 0.090.0E 50 0.690.54G 0.690.89H 0.0090.00E 0.0090.00F 0.090.0E Control 90.491.94A 68.493.84A 68.693.20A 53.292.68E 46.894.76A

Means (9SE) in the same column followed by the same letter are not significantly (PB0.05) different by LSD analysis.

(Table 3). These studies indicated that irradiation reduced immunity in the fourth and fifth instars of C. infuscatellus for parasitization by this parasitoid. Parasitiza- tion varied in the larvae when irradiated at different doses and comparatively higher numbers of fourth and fifth instar larvae were successfully parasitized when treated with 60 or 80 Gy. More parasitoid cocoons were recovered from the irradiated fourth and fifth instars than from non-irradiated third instars, which may be due to the greater size of fourth and fifth instars compared to third instars.

Use of irradiation and low temperatures for prolonged storage of C. flavipes The immature development of C. flavipes varied by radiation dose (Table 4) and significant variations were recorded in developmental rates at different radiation doses. The egg-larval period was significantly longer when parasitoids were irradiated at 20 Gy than the other tested doses. The pupal period of the parasitoid

Downloaded By: [Hendrichs, Jorge] At: 15:59 6 November 2009 also varied significantly at different radiation doses, with those receiving 40 Gy exhibiting a significantly shorter pupal development period (Table 4). Overall, 20 Gy irradiation in conjunction with low temperature (108C) prolonged the storage of the parasitoids (Table 4). Studies indicated that the C. flavipes larvae irradiated at 20 Gy could be stored for 2 months at 108C without apparent damage to quality, as indicated by analysis of F1 adult parasitization rates. In addition, parasitoid cocoons irradiated at 20 Gy could be stored at 108C to delay adult emergence by 29Á30 days with no reduction in parasitoid viability.

Part II: field studies on area-wide management through augmentation with parasitoids Seasonal population fluctuation of T. chilonis in sugarcane fields Parasitization was low during the month of January when temperatures were lower and thereafter increased and reached its first peak in April (Table 5). However, hot weather during May through July suppressed the parasitoids, which did not recover Table 2. Effect of irradiation on suitability of host eggs for parasitization of T. chilonis.

Parasitization potential (%) in host eggs at different age (days)

Dose (Gy) 1 2 3 4 5 6 7 Technology and Science Biocontrol

5 19.291.30B 16.492.40BC 11.292.49CD 6.492.30C 0.690.89C 0.090.0C 0.090.0B 10 23.292.77A 21.093.39A 17.693.84A 11.092.23AB 0.090.0C 0.090.0C 0.090.0B 15 20.291.92B 17.093.39BC 12.692.70BC 11.892.16A 0.890.83C 0.090.0C 0.090.0B 20 23.691.14A 20.092.44AB 15.295.26AB 13.293.70A 9.091.5A 3.892.28A 0.090.0B 25 24.492.70A 19.893.03AB 13.293.76BC 9.292.58B 4.291.9B 1.491.14B 0.890.83A

Downloaded By: [Hendrichs, Jorge] At: 15:59 6 November 2009 6 November 15:59 At: Jorge] [Hendrichs, By: Downloaded 30 19.292.28B 16.492.40BC 13.293.11BC 9.491.14B 4.892.5B 1.890.83B 0.090.0B 35 17.891.64B 15.692.70C 14.693.04ABC 5.491.81C 0.890.83C 0.090.0C 0.090.0B 40 15.291.92C 13.493.97CD 8.491.94DE 1.491.67D 0.090.0C 0.090.0C 0.090.0B 45 13.492.51CD 7.692.28E 5.492.30E 0.290.44D 0.090.0C 0.090.0C 0.090.0B 50 11.691.34C 7.692.40E 1.491.14F 0.090.0D 0.090.0C 0.090.0C 0.090.0B Control 18.891.78B 11.292.38DE 7.292.77E 2.090.70D 0.090.0C 0.090.0C 0.090.0B

Means (9SE) in the same column followed by the same letter are not significantly (PB0.05) different by LSD analysis. 283 284

Table 3. Parasitization of irradiated Chilo infuscatellus larvae by C. flavipes.

Third instar Fourth instar Fifth instar

Mean no. of Mean no. of Mean no. of Emergence % Dose (Gy) cocoons/larva Emergence % cocoons/larva Emergence % cocoons/larva

20 29.492.07C 78.496.65A 0.090.0E 0.090.0D 0.090.0E 0.090.0D Fatima B. 40 31.892.5BC 74.696.22AB 0.090.0E 0.090.0D 0.090.0E 0.090.0D 60 38.492.30A 81.493.84A 46.495.45B 79.892.28A 50.893.83B 83.6979.0A Downloaded By: [Hendrichs, Jorge] At: 15:59 6 November 2009 6 November 15:59 At: Jorge] [Hendrichs, By: Downloaded 80 36.494.66AB 78.896.57A 53.894.32A 81.892.94A 55.693.13A 84.292.77A 100 31.694.98BC 71.493.20B 30.691.51C 70.695.17B 39.294.96C 69.897.49B al et 120 15.294.08D 57.892.58C 17.492.30D 58.893.49C 23.294.43D 62.293.83C . Control 30.894.81C 80.094.47A 0.090.0E 0.090.0D 0.090.0E 0.090.0D

Means (9SE) in the same column followed by the same letter are not significantly (PB0.05) different by LSD analysis. Biocontrol Science and Technology 285

Table 4. Effect of low temperature (108C) in conjunction with radiation on immature development of C. flavipes.

Mean over all F1 parasitism Dose Mean egg/larval Mean pupal developmental Mean pupal Mean no. of (Gy) period (days) period (days) period (days) survival (%) cocoons/larva

10 56.891.92BC 29.291.48A 86.092.23B 74.494.66B 28.291.92A 20 64.292.16A 31.492.30A 95.691.51A 83.693.97A 29.692.07A 30 58.491.51B 29.093.16A 87.493.04B 78.295.16AB 22.092.64B 40 53.494.50C 23.892.28B 78.094.69C 58.694.50C 13.092.44C 50 00.090.00D 00.090.00C 00.090.00D 00.090.00D 00.090.00D Control 55.293.76BC 28.692.70A 84.294.14B 74.896.26B 30.295.26A

Means (9SE) in the same column followed by the same letter are not significantly (PB0.05) different by LSD analysis.

until the more moderate months of August and September (Table 5). The relative humidity remained low from February to June.

Significance of providing supplemental irradiated hosts to the parasitoids for initial survival in the field Results from August and later months indicated that efficacy of the parasitoids was higher in the block where supplemental irradiated hosts were provided to the parasitoids, as evidenced by low infestation rates (Table 6). Temperature also played an important role in the establishment of the parasitoids in the field and the greatest number of parasitoids was observed in the months of April and September when the mean temperature ranged between 25 and 308C. Highest infestation rates were observed in the control block, where no plant protection measures were applied.

Table 5. Effect of temperature on the efficacy of T. chilonis in the sugarcane field.

Parasitism%

Normal Irradiated Max. Min. % Relative Downloaded By: [Hendrichs, Jorge] At: 15:59 6 November 2009 Months host eggs host eggs temperature (8C) temperature (8C) humidity

January 1.091.0E 1.491.14G 18.0 8.5 65.0 February 11.694.15D 14.494.21EF 29.8 11.8 69.0 March 30.297.33B 37.697.73B 37.5 15.3 62.0 April 39.698.47A 45.694.77A 40.0 22.7 67.0 May 23.695.27BC 29.696.80CD 40.8 28.1 64.0 June 21.696.69C 24.694.77D 39.2 27.2 65.0 July 26.095.95BC 31.293.27BC 36.6 26.5 72.0 August 38.498.04A 47.695.89A 36.1 25.8 73.0 September 43.297.08A 48.495.81A 33.0 24.0 72.0 October 28.095.43BC 36.693.50B 32.1 19.0 71.0 November 12.694.15D 17.094.94E 26.0 14.1 64.0 December 6.693.57DE 9.492.30F 26.0 11.0 67.0

Means (9SE) in the same column followed by the same letter are not significantly (PB0.05) different by LSD analysis. 286

Table 6. Effect of supplemental irradiated hosts to enhance the performance of T. chilonis in sugarcane fields.

Parasitization %

Block-1 released parasitoids9supple- Block-2 released parasitoids9non Block-3 untreated control mental host (%) supplemental host (%) (%)

Mean temp.

Month (8C) Parasitization Infestation Parasitization Infestation Parasitization Infestation Fatima B.

February 20.8 39.094.63C 0.890.83G 10.092.23E 0.090.00E 8.294.4A 0.890.83G

Downloaded By: [Hendrichs, Jorge] At: 15:59 6 November 2009 6 November 15:59 At: Jorge] [Hendrichs, By: Downloaded March 26.4 48.093.46B 4.692.7DE 37.894.2C 1.090.70E 2.090.70CD 7.891.92F

April 31.3 58.095.83A 6.092.2CDE 48.297.1AB 5.691.14D 1.891.09D 11.692.4E al et May 34.4 40.092.91C 7.291.78BC 34.297.46CD 7.092.91CD 2.091.22CD 14.294.43DE

June 33.2 37.093.39C 8.892.38B 30.094.35D 8.092.44BCD 3.290.83CD 16.893.96CD . July 31.5 41.094.69C 12.892.68A 32.094.48CD 10.291.92AB 0.890.83D 19.292.49BC August 31.2 57.095.24A 7.090.70BCD 45.095.33B 12.093.00A 2.891.30CD 21.692.70AB September 30.0 61.096.28A 3.691.14EF 52.895.76A 8.893.19BC 4.691.67BC 24.894.14A October 26.0 55.095.43A 1.290.83FG 00.090.00F 5.891.30D 7.292.58AB 12.293.11E

Means (9SE) in the same column followed by the same letter are not significantly (PB0.05) different by LSD analysis. Biocontrol Science and Technology 287

Both blocks receiving augmentative parasitoid releases, with or without supple- mental hosts, generally experienced reduced infestation rates throughout the season (Table 6). The infestation by C. infuscatellus in the control treatment was higher and remained above the economic threshold level (10%) from April to October.

Area-wide tests The infestation of sugarcane stem borers in the entire 40,000 ha area treated by provision of supplemental host eggs was successfully managed by the egg parasitoid, T. chilonis, resulting in damage levels below economic threshold level. The infestation of sugarcane borers on an internodal basis ranged from 0 to 9.7% with an overall mean of 4.8% in the treated areas. The mean infestation of the borers in the untreated areas was 21.1%, with a range of 16.4Á31% (Table 7). The percentage recovery of T. chilonis from feral and sentinel eggs were also higher in the treated area as compared to the un-treated fields. Moreover, the parasitoid recovery was significantly higher during the months of August and September followed by June and July when the environmental conditions were conducive. Pupal recovery of the parasitoids was much higher in the fields treated with supplemental, irradiated host eggs as compared to the untreated fields (Table 8).

Discussion Eggs of S. cerealella are widely used for the production of T. chilonis (Morrison, Stinner, and Ridgway 1976; Mannion et al. 1994) and for population suppression of sugarcane borers (Carpenter et al. 1995). However, for release programs, all of the eggs of the pest placed in the field should be parasitized or developmentally arrested; otherwise the eggs that hatch may serve to increase the population of S. cerealella. This problem can be avoided by using sterilized eggs. In our studies, gamma radiation was very effective in developmentally arresting host eggs without reducing their utility as hosts for parasitoids. Radiation doses of 20 and 25 Gy enhanced parasitization and reduced the host egg age effect for parasitization. Lewis and Young (1972) also reported the acceptance of irradiated host eggs by polyphagous Downloaded By: [Hendrichs, Jorge] At: 15:59 6 November 2009

Table 7. Chilo infuscatellus infestation in treated and untreated areas.

% Borer infestation

Location Village Treated fields Untreated fields

1 Moosa khatian 3.8 18.7 2 Abbri 4.7 20.6 3 Lakhi keti 4.6 17.5 4 Quba stop 4.0 25.4 5 Mehran sugar miils area 6.4 31.0 6 NIA expt. Farm 0.0 16.4 7 Tando Soomro 4.3 19.8 8 Chambar 9.7 22.5 Average 4.8 21.1 288 B. Fatima et al.

Table 8. Percent recovery of T. chilonis from feral and sentinel host eggs in treated and untreated sugarcane field.

Treated area Untreated

Month Feral eggs Sentinel eggs Feral eggs Sentinel eggs

January 00.090.0F 0.490.54H 0.090.0D 0.090.0F February 00.090.0F 1.091.0F 0.090.0D 1.091.0EF March 13.693.71E 18.496.38F 1.691.14D 3.491.14DE April 42.6910.31C 46.696.34C 5.492.40BC 7.491.81ABC May 36.494.92C 39.295.63D 4.291.92C 6.493.50BC June 41.693.04C 45.693.91C 4.691.51C 6.093.39BCD July 41.693.20C 46.295.26C 5.491.14BC 5.691.14CD August 52.695.49B 57.496.22B 7.091.58AB 8.692.30AB September 59.699.44A 66.695.72A 7.691.34A 9.492.88A October 28.096.67D 31.693.78E 6.492.60AB 7.692.96ABC November 12.493.43E 21.093.30H 1.090.70D 1.691.51EF December 00.490.54F 7.492.30G 0.090.0D 0.090.0F

Means (9SE) in the same column followed by the same letter are not significantly (PB0.05) different by LSD analysis.

egg parasitoids. Brower (1982) found that irradiated eggs of Indian meal moth, Plodia interpunctella (Hubner) were preferred over normal, untreated eggs. Thus, the weight of evidence from our studies and others indicates that irradiated eggs may be used effectively for laboratory mass rearing and field releases with no problems posed by the hatching of host eggs. Our studies also showed that the provisioning of host eggs in the field may be helpful. Parker and Pinnell (1972) reported similar results by provisioning hosts early in the season along with Trichogramma parasitoids. The irradiated eggs sustained the parasitoid population by providing additional host eggs with little risk of increasing pest populations. Temperature and relative humidity played a significant role in successful build-up of the parasitoids. Fewer parasitoids were observed in the field during May to July when the temperature remained high and relative humidity low. The parasitoid- Downloaded By: [Hendrichs, Jorge] At: 15:59 6 November 2009 releasing interval may need to be adjusted on the basis of prevailing environmental conditions for successful application of biological agents against sugarcane borers in the field. Doses of 60 and 80 Gy on host larvae were most effective to improve the mass rearing efficiency of C. flavipes by increasing their suitability past the normal third instar, enabling fourth instar host larvae to be used. In addition, more pupal cocoons were recovered from fourth instar hosts compared to third instars, which may be due to their greater size which might favor a gregarious endoparasitoid like C. flavipes. The sex ratio of the parasitoids reared on irradiated larvae also was skewed in favor of females. The immature development of C. flavipes was reduced in irradiated host larvae, which allowed them to be stored for longer periods before use. Host eggs and parasitoid larvae and pupae could be stored for 2 months at 108C after irradiation at 20 Gy with no apparent loss in host quality or parasitoid reproductive potential. Regarding the area-wide control trials using T. chilonis, provision of supplemental hosts to enhance the efficacy of the egg parasitoids proved effective for Biocontrol Science and Technology 289

the control of C. infuscatellus as the parasitization rate was higher and the C. infuscatellus infestation was lower in the blocks with supplemental hosts as compared to the other blocks. There are many factors which affect the establishment of parasitoids in the field. Ashraf, Fatima, and Ahmad (2001) reported that longevity of both male and female wasps of T. chilonis in the laboratory was significantly higher when reared at 22.5% RH. At low relative humidity, the parasitoids may be less active as observed in the field. Calvin, Knaff, Welch, Poston, and Etzinga (1984) also recorded significant effects of relative humidity on the developmental stages, longevity, fecundity and sex ratio of Trichogramma pretiosum Westwood. Similar results were observed in the present studies and the parasitization rate was higher during the months of April and August when the humidity was relatively high in the field. It has long been recognized that temperature is the single most important factor influencing the development of the immature stages and the adult maturation rates of the majority of insects. Temperature/development rate relationships could be useful in predicting phenological events in the field for ecological and pest management purposes, the optimization of mass rearing procedures under constant conditions, and construction of computer simulation models of population suppression. The present studies revealed that population establishment in the field had a linear relationship with the temperature. Establishment was slow during hot months and it increased in the months when temperature decreased. In conclusion, irradiation of host material proved useful for increasing the rearing efficiency of T. chilonis and C. flavipes, in addition to expediting the development of practical and cost-effective area-wide augmentative biological control programs in the field.

References Ashraf, M., Fatima, B., Hussain, T., and Ahmad, N. (1999), ‘Biological Control: An Essential Component of IPM Programme for Sugarcane Borers’, Symposium on Biological Control in the Tropics, MARDI Training Centre, Serdang, Selangor, Malaysia, 18Á19 March 1999. Ashraf, M., Fatima, B., and Ahmad, N. (2001), ‘Development of Parasitoid Rearing Systems to Enhance Augmentative Releases for the Management of Sugarcane Borers’, Proceedings of the 2nd FAO/IAEA Coordination Meeting on Evaluating the Use of Nuclear Techniques for the Colonization and Production of Natural Enemies of Agricultural Insect Pests, held at Downloaded By: [Hendrichs, Jorge] At: 15:59 6 November 2009 Tapachula, Chiapas, Mexico, 18Á22 June 2001, pp. 1Á13. Brower, J.H. (1982), ‘Parasitization of Irradiated Eggs and Eggs from Irradiated Adults of the Indian Meal Moth (Lepidoptera: Pyralidae) by Trichogramma pretiosum (Hymenoptera: Trichogrammatidae)’, Journal of Economic Entomology, 75, 939Á944. Browning, H.W., and Melton, C.W. (1987), ‘Indigenous and Exotic Trichogrammatids (Hymenoptera:Trichogrammatidae) Evaluated for Biological Control of Eoreuma loftini and Diatraea saccharalis (Lepidoptera:Pyralidae) Borers on Sugarcane’, Environmental Entomology, 16, 60Á364. Calvin, D.D., Knaff, M.C., Welch, S.M., Poston, F.L., and Elzinga, R.J. (1984), ‘Impact of Environmental Factors on Trichogramma pretiosum Reared on Corn Borer Eggs’, Environmental Entomology, 13, 274Á280. Carpenter, J.E. (1996), ‘Development and Integration of Alternative Management Strategies Using Inherited Sterility and Natural Enemies to Control Lepidopteran Pests’, Second Research Coordination Meeting on F1 Sterility, IAEA, Vienna, Austria, pp. 1Á12. Carpenter, J.E., Mannion, C.M., and Hidrayani, N. (1995), ‘Potential of Combining Inherited Sterility and Parasitoids for Managing Lepidopteran Pests’, Proceedings of FAO/IAEA First Coordination Meeting on Evaluation of Population Suppression by Irradiated Lepidoptera and Their Progeny, Jakarta, Indonesia, 24Á28 April 1995. IAEA-D4-RC-561, pp. 273Á283. 290 B. Fatima et al.

Cohen, A.C. (2003), Insect Diets: Science and Technology, Boca Raton, FL: CRC Press, 324 pp. Greany, P., and Carpenter, J.E. (2000), ‘Use of Nuclear Techniques in Biological Control’,in Area-Wide Control of Fruit Flies and Other Insect Pests, ed. K H. Tan, Penang: Penerbit Universiti Sains Malaysia, pp. 221Á227. Hassan, S.A., Kohler, E., and Rost, W.M. (1988), ‘Mass Production and Utilization of Trichogramma:10. Control of the Codling Moth Cydia pomonella and the Summer Fruit Tortrix Moth Adoxophyes orana (Lepidoptera:Tortricidae)’, Entomophaga, 33, 413Á420. Leppla, N.C., Bloem, K.A., and Luck, R.F. (2002), ‘Quality Control for Mass-Reared Arthropods’,inProceedings of the Eighth and Ninth Workshops of the International Organization for Biological Control Working Group on Quality Control of Mass-Reared Arthropods, 171 pp. Lewis, W.J., and Young, J.R. (1972), ‘Parasitism by Trichogramma evanescens of Eggs from Tepa-sterilized and Normal Heliothis zea’, Journal of Economic Entomology, 65, 705Á708. Losey, J.E., Fleischer, J., Calvin, D.D., Harkness, W.L., and Leahy, T. (1995), ‘Evaluation of Trichogramma nubilalis and Bacillus thuringiensis in Management of Ostrinia nubilalis (Lepidoptera:Pyralidae) in Sweet Corn’, Environmental Entomology, 24, 436Á445. Mannion, C.M., Carpenter, J.E., and Gross, H.R. (1994), ‘Potential for the Combined Use of Inherited Sterility and a Parasitoid, Archytas marmoratus (Diptera:Tachinidae), for Managing Helicoverpa zea (Lepidoptera:Noctiidae)’, Environmental Entomology, 23, 41Á46. Mannion, C.M., Carpenter, J.E., and Gross, H.R. (1995), ‘Integration of Inherited Sterility and a Parasitoid, Archytas marmoratus (Diptera:Tachinidae), for Managing Heliocoverpa zea (Lepidoptera:Noctiidae). Acceptability and Suitability of Hosts’, Environment Entomol- ogy, 24, 1679Á1684. Marston, N., and Ertle, L.R. (1969), ‘Host Age and Parasitism by Trichgramma minutum (Hymenoptera: Trichogramatidae)’, Annals of the Entomological Society of America, 62, 1476Á1482. Mohyuddin, A.I. (1991), ‘Utilization of Natural Enemies for the Control of Insect Pests of Sugarcane’, Insect Science and Application, 12, 19Á26. Morrison, R.K., Stinner, R.E., and Ridgway, R.L. (1976), ‘Mass Production of Trichogramma chilonis on Eggs of S. cerealella’, Southwestern Entomologist,1,74Á84. Nordlund, D.A. (1984), ‘Biological Control with Entomophagous Insects’, Journal Georgia Entomological Society, 19, 14Á27. Parker, F.D., and Pinnell, R.E. (1972), ‘Further Studies of the Biological Control of Pieris rapae Using Supplemental Host and Parasite Releases’, Environmental Entomology,1, 150Á157. Saikia, D.K., and Nath, R.K. (2002), ‘Larval Parasitoids of Sugarcane Early Shoot Borer, Chilo infuscatellus Snellen’, Insect Environment,8,90Á91. Suasa-Ard, W., and Charernsom, K. (1999), ‘Success of Cotesia flavipes (Cameron) for

Downloaded By: [Hendrichs, Jorge] At: 15:59 6 November 2009 Biological Control of Sugarcane Moth Borers in Thailand’, Proceedings of the XXIII ISSCT Congress, New Delhi, India, 22Á26 February, Vol. 2, pp. 559Á568. Tanwar, R.K., and Ashok, V. (2002), ‘Field Trials with Cotesia flavipes Cameron against Sugarcane Borers in Subtropical India’, Sugar Technology, 4, 153Á156. Voegele, J., Daumal, J., Brun, P., and Onillon, J. (1974), ‘The Effect of Cold Storage and UV Radiation Treatment of the Eggs of Ephestia kuehniella (Pyralidae) on the Fecundity of Trichogramma evanescens and T. brasiliensis (Hymenoptera: Trichogramatidae)’, Entomophaga, 19, 341Á348. Biocontrol Science and Technology, Vol. 19, S1, 2009, 291Á301

RESEARCH ARTICLE Impact of gamma radiation on the developmental characteristics of the gypsy moth, Lymantria dispar (Lepidoptera: Lymantriidae) preparatory to their use as supplemental hosts/prey for natural enemy enhancement Milan Zu´brik and Ju´lius Novotny´*

Forest Research Institute, Research Station Banska´Sˇ tiavnica, Lesnı´cka 11, 969 23 Banska´ Sˇ tiavnica, Slovak Republic

The developmental characteristics of irradiated and non-irradiated gypsy moth, Lymantria dispar L. eggs and larvae were compared. Gypsy moth eggs were irradiated a few days before hatching at 10, 20, 30, 40, 50, 60, 80, or 110 Gray (Gy) and first instar larvae at 50, 80, or 110 Gy of gamma radiation and tested for differences in their development by comparison with non-irradiated controls. Untreated larvae developed to the adult stage more rapidly than irradiated larvae treated either as eggs or as larvae and this was dose-dependent. Larval mortality and pupal developmental anomalies were dose-dependent. Pupal morphological abnormalities occurred in only 1.6% of controls, but in the 110 Gy group, they occurred in 93.2 and 95.1% of individuals treated as eggs or larvae, respectively. The tests showed that 50 Gy was optimal for irradiating gypsy moth eggs and larvae to achieve F1 sterility and extend larval development without excessive mortality. This may facilitate use of these sterile larvae as supplemental hosts in augmentative biological control programmes. Keywords: gypsy moth; Lymantria dispar; gamma radiation; larvae developmental characteristics; augmentative biological control; radiation hormesis

Introduction The gypsy moth (Lymantria dispar L.) is one of the most important forest insect pest species in Europe, Asia, and North America. Larvae of this moth defoliate large

Downloaded By: [Hendrichs, Jorge] At: 16:00 6 November 2009 areas of broadleaf stands annually. It was imported into Massachusetts in 1869 for possible silk production and it accidentally escaped captivity. As of 1994, the US Forest Service spent approximately $11 million annually on gypsy moth control (Campbell and Schlarbaum 1994). Currently, APHIS and its state cooperators spend nearly $10 million annually to prevent gypsy moths from establishing in the western portions of the United States (Vic Mastro, USDA APHIS, personal communica- tion). Control of the gypsy moth in the United States was reviewed by Liebhold and McManus (1999). Parasitoids, predators and pathogens are believed to play substantial roles in the dynamics of gypsy moth populations in Europe. Studies were performed on L. dispar natural enemies in Slovakia (Novotny´ 1989) and in Austria, the latter being carried out by the U.S. Department of Agriculture European Parasitoid Laboratory during an outbreak in the 1970s (Fuester, Drea, Gruber, Hozer, and Mercadier 1983). The

*Corresponding author. Email: [email protected]

ISSN 0958-3157 print/ISSN 1360-0478 online # 2009 Taylor & Francis DOI: 10.1080/09583150902812188 http://www.informaworld.com 292 M. Zu´brik and J. Novotny´

Commonwealth Institute for Biological Control discovered diverse complexes of parasitoids, predators and pathogens in Europe (Eichhorn 1996). There are several possibilities as to how populations of indigenous natural enemies build up (Maksimovicˇ and Sivcˇev 1984; Mills 1990). These studies indicate that increasing the efficacy of parasitoids and predators through direct augmenta- tive releases would be problematic. Similarly, pathogens might be useful for population suppression and they may reduce the host population density very quickly and effectively (Podgwaite, Reardon, Walton, and Witcosky 1992; Podgwaite, Dubois, Reardon, and Witcosky 1993; Shapiro, Robertson, Injac, Katagirik, and Bell 1984; Shapiro and Dougherty 1985). In recent years, dramatic collapses of gypsy moth populations in the United States are due, in large part, to the fungus Entomophaga maimaiga, collected in Japan and deliberately released near Boston between 1910 and 1911 (Hajek, Humber, and Elkinton 1995; Hajek, McManus, and Delalibera 2007). As a more cost-effective alternative to augmen- tative releases of natural enemies, it may instead be possible to release sterile, developmentally-compromised gypsy moth eggs and/or larvae produced and irradiated in the laboratory. These could serve as supplemental hosts/prey to achieve a more rapid build up of the natural enemies to preclude the pest population from achieving damaging levels. Traditional uses of radiation in pest management focus on use of the Sterilie Insect Technique (SIT). Since SIT was first used successfully for pest management (Knipling 1955), its effectiveness has been demonstrated for many types of pest insects. Strategies by which sterile insect releases might be used to control the gypsy moth (Maksimovicˇ 1971a,b; Mastro and Schwalbe 1988; Mastro 1993) normally involve introducing sterile adults into the pest’s habitat to flood the fertile population, with the goal of reducing the pest population by interfering with reproduction. A number of studies have been conducted in which pest insects that served as laboratory hosts for parasitoids were irradiated in order to arrest their development or to preclude release of fertile females when inadvertently released along with the parasitoids (reviewed by Greany and Carpenter 2000). However, use of radiation to produce innocuous, sterile pest immatures to supplement natural

Downloaded By: [Hendrichs, Jorge] At: 16:00 6 November 2009 populations for the purpose of building up natural enemy populations has not been pursued to date and this is the subject of the present research. Studies were therefore conducted on the effects of radiation on gypsy moth immatures preparatory to enabling this approach. Sterility is normally achieved by irradiation of the pupae or adults with adequate doses of gamma radiation. Since mature ovaries are not present in the eggs or larvae, the eggs and larvae are not usually irradiated. Despite this, it has been shown in research on Nezara viridula L. that fully-developed immatures may be irradiated and adult sterility achieved (Dyby and Sailer 1999). The aim of the present research was to apply a range of radiation doses to gypsy moth eggs and larvae to evaluate

effects on their developmental characteristics and those of the F1 generation. This in turn should establish the optimal rate/timing needed to produce supplemental hosts/ prey for natural enemy build up without fear of releasing reproductively competent pests. Biocontrol Science and Technology 293

Material and methods 1. Insect and bioassay We used a laboratory strain of the gypsy moth (Lymantria dispar L.) supplied by the USDA Á APHIS Methods Development Centre in Otis, MA (USA). Eggs and larvae in first instar were used as the starting stage for bioassay throughout this study. Irradiation was performed at the Entomology Unit of the FAO/IAEA Laboratories in Seibersdorf, Austria, using a 60Co Gammacell 220 (AE of Canada Ltd) with dose rates ranging between 50 and 60 Gy/min. Newly-laid egg masses were stored in the refrigerator at approximately 5Á78Cfor 120 days until hatching was possible. Eggs masses (10 masses per dose) were irradiated a few days before hatching, at eight different doses (10, 20, 30, 40, 50, 60, 80, or 110 Gy). Irradiated egg masses were placed into the Petri dishes until the larvae hatched. The number of hatched larvae and unhatched eggs were counted. Larvae (100 per dose) hatched from the irradiated eggs were moved into plastic cups (5 larvae per cup) containing the artificial diet, reared, and tested for differences in their development by comparison with non-irradiated controls. The larvae were maintained at 20Á268C and 60Á70% RH. An artificial wheat germ diet (Bell, Owens, Shapiro, and Tardif 1981), which is commonly used in gypsy moth rearing, was used. From the third instar, larvae were reared separately. Larvae were checked daily. The duration of each larval stage was determined by noting the time of moulting. Dead larvae were removed from rearing cups. Pupae were stored in boxes (6060100 cm high) until adults emerged. The condition and health of the pupae were determined and they were weighed 2 days after pupation. The number of successfully emerged imagos was recorded as well as the number of egg masses and the number of eggs that females laid. For a comparison of development, we reared a treatment of 100 non-irradiated larvae to serve as controls. Three replicates were performed. In the second experiment, L1 larvae were subjected to 50, 80, or 110 Gy and tested for differences in their development in comparison with non-irradiated controls. For the laboratory investigation, 100 larvae from each treatment were used. The larvae were maintained the same way as in the first experiment. Three replicates were performed. Downloaded By: [Hendrichs, Jorge] At: 16:00 6 November 2009 2. Statistical analyses Analyses of variance (ANOVA) were performed on larval duration, pupal weight and number of eggs produced by females to determine the effects of irradiation dose (Tukey HSD test for unequal number of replicates) using Statistica (StatSoft, Inc.). Significant differences were tested at P B 0.05. Standard errors (SE) of the means were calculated.

Results 1. Developmental time The developmental time of male gypsy moth larvae irradiated as eggs varied between 35.5 and 44.4 days (Table 1). The control group together with 10- and 20-Gy treatments developed significantly faster than the 110 Gy irradiation treatment, which had the longest development time. The lower irradiation doses (10Á20 Gy) 294

Table 1. Mean developmental time (9SE) in days for larvae of Lymantria dispar treated as eggs with increasing doses of gamma radiation.

Treatment (Gy) No. larvae reared L1 L2 L3 L4 L5 L6 Total (days)

Male larvae 0 300 10.090.1 a 5.290.1 b 5.790.2 b 5.390.1 b 9.390.2 a Á 35.5 10 300 10.690.2 ab 5.490.1 b 5.590.1 ab 4.790.1 a 9.490.1 a Á 35.6 20 300 10.590.2 ab 4.690.1 a 5.690.1 b 5.390.1 b 9.590.1 a Á 35.5

30 300 10.990.1 bc 5.290.2 ab 6.790.2 b 5.090.1 ab 9.590.2 a Á 37.3 Zu M. 40 300 11.290.1 bc 5.390.2 b 6.490.2 bc 5.290.1 ab 9.790.1 a Á 37.8

50 300 10.790.1 abc 5.390.1 ab 4.890.1 a 7.390.3 b 11.490.5 b Á 39.5 ´ 60 300 11.390.1 c 5.0901 ab 5.990.2 bc 6.390.2 c 11.090.3 b Á 39.5 Novotny J. and brik 80 300 11.590.1 c 5.290.1 ab 7.090.2 d 6.390.2 c 11.990.6 b Á 41.9

Downloaded By: [Hendrichs, Jorge] At: 16:00 6 November 2009 6 November 16:00 At: Jorge] [Hendrichs, By: Downloaded 110 300 12.790.2 d 5.590.1 ab 6.290.1 bcd 5.690.2 bc 14.490.5 c Á 44.4 Female larvae 0 300 10.190.2 a 5.090.1 bcd 5.490.1 ab 5.290.1 b 5.890.2 ab 10.190.2 a 41.6 10 300 11.090.2 bcd 5.590.2 cde 5.190.1 ab 4.490.1 a 5.590.2 ab 10.190.2 a 41.6 20 300 10.990.2 bc 4.490.2 ab 5.590.1 abc 5.490.1 bd 5.590.1 ab 9.890.1 a 41.5 ´ 30 300 10.690.2 ab 4.690.2 abc 5.390.3 abc 4.590.1 ac 5.190.1 a 10.090.2 a 39.9 40 300 10.990.2 bc 4.290.2 a 5.090.1 a 5.590.1 bd 5.190.1 a 9.890.1 a 40.5 50 300 10.490.1 ab 5.490.2 cde 4.990.2 a 6.190.2 d 7.190.5 c 10.590.3 ab 44.4 60 300 11.790.1 cd 5.890.2 e 5.990.3 c 5.390.2 b 5.590.3 ab 11.390.2 bc 45.5 80 300 11.190.1 bcd 5.190.1 abcde 5.890.3d 5.890.5 b 5.790.3 b 12.190.4 c 45.6 110 300 11.990.1 b 5.691.1 de 6.690.2 bc 5.390.1 bc 6.390.2 bc 12.390.2 c 48.0

Larvae were reared in the laboratory on artificial diet. Data for male and female larvae are presented separately. L1ÁL6larval instars. Means for male or female groups followed by different letters (within columns) are significantly different at P50.05 by Tukey HSD. Biocontrol Science and Technology 295

sometimes stimulated the growth rate (Table 1, L2, L3, L4). This may be an expression of a radiation-induced hormetic effect (Luckey 1991). Most of the males developed through five instars. In the first and fifth instars, larvae from the control treatment developed significantly faster than those of all other treatments in these stages (P B 0.01). Female developmental time varied between 41.5 and 48 days (Table 1). Most of the females developed through six instars. The control group needed the shortest time for development. The treatments differ also in the female larval stage. We recognized significant differences between treatments in all female larval stages, with a shorter development time in control, 10-, 20-, 30-, and 40-Gy treatments, which differ significantly from the 50-, 60-, and 110-Gy treatments. In experiments with larvae irradiated in the earliest larval stage (L1), very similar results were obtained (Table 2). Male development time varied between 35.6 and 40.5 days (Table 3). The control group developed significantly faster than the 110-Gy group, the irradiation treatment with the longest development time. We also determined the statistical differences (PB0.05) between the treatments in all male larval stages. The female developmental time varied between 41.6 and 50.2 days (Table 2). The control group again developed most rapidly. We observed significant differences between treatments in all female larval stages (Table 2).

2. Health condition Some irradiated larvae displayed serious health problems during development. Larval mortality appeared to be dose-dependent, reaching 15.7% among larvae irradiated as eggs receiving 110 Gy (Figure 1a). Similar results were obtained in treatments with larvae irradiated in early stage Á L1 (Figure 1b). Mortality reached 22% among larvae irradiated by 110 Gy. In the control group of larvae, mortality was only 4%. In both sets, differences were statistically significant (PB0.05). Dose-dependent effects of radiation on the health condition of pupae also were noted. Pupae developing from irradiated larvae or eggs showed a high incidence of morphological abnormalities. The thoracic segments of the pupae were undeveloped and the antennae also were damaged in different ways. Mortality of pupae, which

Downloaded By: [Hendrichs, Jorge] At: 16:00 6 November 2009 came from irradiated eggs, was relatively high and reached the highest level, ca. 91%, at 110 Gy (Figure 1d). Pupae deriving from irradiated larvae exhibited higher rates of mortality, reaching 97.5% at 110 Gy (Figure 1c). Differences were statistically significant (PB0.05). In the control group, morphological abnormalities reached only 1.6%, but in the 110 Gy treatment, it was 93% (from irradiated eggs) or 95% (from irradiated larvae), respectively. Above-mentioned abnormalities affected the ability of the adults to emerge. Abnormalities occurred in all treatment groups and were absent among the controls. The emergent adults were not able to mate successfully because of these morphological problems, the most common being wing damage and the defects of mobility. Pupal weights for irradiated males and females deriving from treated eggs and larvae were significantly (PB0.05) different as a function of radiation dose, with the lowest weights occurring for the highest dose treatments (Table 3). 296

Table 2. Mean developmental time (9SE) in days for larvae of Lymantria dispar treated as larvae in first instar with increasing doses of gamma radiation. Larvae were reared in the laboratory on artificial diet.

Treatment (Gy) Number of larvae followed L1 L2 L3 L4 L5 L6 Total (days) .Zu M.

Male larvae ´

0 300 10.190.1 a 5.290.1 a 5.790.2 ab 5.390.1 a 9.390.2 a Á 35.6 Novotny J. and brik 50 300 10.690.2 a 5.290.1 a 5.590.1 ab 6.190.2 b 11.990.4 b Á 39.3 80 300 12.990.1 b 6.290.2 b 5.390.1 a 6.690.2 b 9.590.6 a Á 40.5 Downloaded By: [Hendrichs, Jorge] At: 16:00 6 November 2009 6 November 16:00 At: Jorge] [Hendrichs, By: Downloaded 110 300 12.790.3 b 6.490.2 b 6.190.2 c 6.790.2 b 8.690.8 a Á 40.5 Female larvae 0 300 10.190.2 a 5.090.1 a 5.490.1 a 5.290.1 a 5.890.2 a 10.190.2 a 41.6 50 300 9.890.2 a 4.890.1 a 5.390.2 a 6.090.1 abc 7.890.4 b 10.890.3 ab 44.5 80 300 11.7 0.2 b 6.2 0.2 b 5.5 0.1 a 6.0 0.1 b 6.2 0.3 a 14.6 0.4 c 50.2 9 9 9 9 9 9 ´ 110 300 11.390.3 b 5.890.2 b 6.290.2 b 6.990.3 c 6.490.4 a 12.690.9 bc 49.2

Data for male and female larvae are presented separately. L1ÁL6larval instars. Means for male or female treatment groups followed by different letters (within columns) are significantly different at P 5 0.05 by Tukey HSD. Biocontrol Science and Technology 297

Table 3. Weight of pupae from irradiated and control groups of L. dispar fed on artificial diet in laboratory conditions.

Weight of pupae (g)

Male Female

Treatment (parents irradiated as eggs) N (X9SE) N (X9SE)

Control Á untreated 160 558.392.5 f 128 1476.8913.5 ef 10 Gy 176 579.693.2 g 108 1637.2913.4 g 20 Gy 123 466.096.7 d 158 1417.2917.2 e 30 Gy 132 473.496.1 e 136 1355.0912.2 d 40 Gy 112 464.493.0 d 151 1218.6911.2 c 50 Gy 156 570.397.7 g 106 1512.9916.0 f 60 Gy 144 459.994.3 c 113 1208.7919.1 c 80 Gy 121 434.996.1 b 134 1060.1925.0 b 110 Gy 146 362.993.2 a 107 903.6911.5 a Treatment (parents irradiated as larvae) Control Á untreated 160 558.392.5 c 128 1476.8913.5 d 50 Gy 123 533.994.7 c 12 1379.4916.4 c 80 Gy 140 416.898.8 b 22 1194.4919.7 b 110 Gy 127 324.898.2 a 20 1097.5918.3 a

Means in a treatment group followed by different letters are significantly different at P 5 0.05 by Tukey HSD. N, the total number of pupae.

100 100 90 (a)90 (b) 80 80 70 70 60 60 50 50 40 40 % mortality % mortality 30 30 20 20 10 10 0 0 0 0 80 10 20 30 40 50 60 10 20 30 40 50 60 80 110 110

Downloaded By: [Hendrichs, Jorge] At: 16:00 6 November 2009 Dose of radiation (in Gy) Dose of radiation (in Gy)

100 100 90 (c) 90 (d) 80 80 70 70 60 60 50 50 40 40 30 30 20 20

% unhatched pupae 10 % unhatched pupae 10 0 0 0 0 10 20 30 40 50 60 80 10 20 30 40 50 60 80 110 110 Dose of radiation (in Gy) Dose of radiation (in Gy)

Figure 1. Effect of dose of radiation on (a) percent mortality for gypsy moth larvae irradiated as eggs, (b) on percent mortality for gypsy moth larvae irradiated in early larval stage, (c) on percent uneclosed pupae when eggs of gypsy moth were treated with gamma radiation, and (d) on percent uneclosed pupae when larvae of gypsy moth in early stage were treated with gamma radiation (9SE). 298 M. Zu´brik and J. Novotny´

3. Impact of the irradiation dose on the hatching ability of the eggs and F1 eggs Egg hatch was influenced by the irradiation dose. In the control group, the proportion of unhatched eggs was 9.5%. In all other groups it was much higher Á up to 70.3% in the 110 Gy treatment (Table 4). Impact of the irradiation dose on the hatching ability of the F1 eggs also was observed (Table 5). In some cases, adults were not able to lay eggs and no eggs for calculation were available (Table 5). Mortality reached 98.6% in 60 Gy treatment (eggs from parents irradiated as eggs) and 100% in 80 Gy treatments (eggs from parents irradiated as larvae). Differences were statistically significant (PB0.05).

Discussion The findings of the present studies should be useful in developing improved capabilities for augmentative biological control of the gypsy moth. After its original introduction into the United States in 1869, classical biological control of the gypsy moth was applied using natural enemies from Europe (Doane and McManus 1981). Rather than relying upon natural build up of naturally-occurring and introduced predators and parasitoids, it may be possible to accelerate and increase levels of control by use of augmentative biological control approaches against gypsy moth, similar to use of Trichogramma spp. in greenhouses and under field conditions (Mills 1990; Smith 1993; Wallace and Smith 1995). Maksimovicˇ and Sivcˇev (1984) introduced into the gypsy moth population a large number of (fertile) eggs with the aim to increase the efficacy of the natural parasitoids and they found that the parasitism rates associated with the endopar- asitic wasps Cotesia melanoscela (Ratzeburg) and Apanteles liparidis Bouche were slightly increased. The approach they used, using viable gypsy moth eggs to augment the natural host population, is risky as there is a possibility of creating an artificial, unexpected outbreak of the gypsy moth in the area. To minimize this likelihood, our results suggest it might be possible to safely use irradiated gypsy moth eggs and/or larvae to augment the natural population. This is valid mostly at the higher doses of irradiation. The higher the irradiation dose used, the lower the risk of causing an unexpected artificial outbreak. The probability of irradiated eggs, larvae and pupae

Downloaded By: [Hendrichs, Jorge] At: 16:00 6 November 2009 reaching maturity is inversely related to the irradiation dose (Figure 1aÁc,

Table 4. Impact of the irradiation dose on the hatching ability of the eggs after irradiation.

Treatments N Number of unhatched eggs in % (X9SE)

Control Á untreated 30 9.590.3 a 10 Gy 30 9.690.1 a 20 Gy 30 11.090.1 b 30 Gy 30 14.390.1 c 40 Gy 30 15.590.1 d 50 Gy 30 21.691.4 e 60 Gy 30 30.190.6 f 80 Gy 30 50.891.5 g 110 Gy 30 70.391.5 h

Means followed by different letters are significantly different at P 5 0.05 by Tukey HSD. N, number of egg masses used for calculations. Biocontrol Science and Technology 299

Table 5. Impact of the irradiation dose on the hatching ability of the F1 eggs (average number of the eggs per egg mass; unhatched eggs Á in%).

Treatment (parents Average number of eggs per egg irradiated as eggs) N mass (X9SE) Unhatched eggs (%)

Control Á untreated 44 445.5930.4 a 9.890.6 a 10 Gy 10 434.9938.6 a 16.892.1 b 20 Gy 10 330.3929.2 a 17.591.6 b 30 Gy 10 291.4925.8 a 31.492.3 c 40 Gy 10 325.1931.9 a 57.795.6 c 50 Gy 10 458.1959.8 a 75.096.8 c 60 Gy 3 158.7923.1 b 98.691.3 c 80 Gy 0 No eggs available No eggs available 110 Gy 0 No eggs available No eggs available Average number of eggs (X9SE) (X9SE) Control Á untreated 44 445.5930.4 a 9.890.6 a 50 Gy 4 212.2918.0 a 91.695.1 b 80 Gy 3 48.6911.8 a 100.0090.0 b 110 Gy 0 No eggs available No eggs available

Means followed by different letters are significantly different at P 5 0.05 by Tukey HSD.

respectively). Hosts irradiated as eggs as well as larvae would not be able to create a new generation because of somatic damage, immobility or sterility, respectively. The gypsy moth embryo develops into a small caterpillar and over-winters in the egg (Doane and McManus 1981). In our experiments, eggs were irradiated shortly before hatch or as neonate (L1) larvae. Little difference was noted in use of either stage (see Figure 1, Tables 1 and 2). However, it is much more practicable to use irradiated eggs instead of larvae for field release. Extension of larval development occurred as a result of irradiating eggs (Table 1) and larvae (Table 2) and developmental was dose-dependent. This could have practical implications for augmentation. The parasitoids and predators should have more time to discover the treated larvae in the field, and parasitoids also would have Downloaded By: [Hendrichs, Jorge] At: 16:00 6 November 2009 more time for their own development. The very widespread and abundant parasitoid, Cotesia melanoscela Ratz. (Hymenoptera: Braconidae) is aided by slow host development because females are not very successful at parasitizing fourth and later instars for example (Hoch, Zu´brik, Novotny´, and Schopf 2001). The developmental time extension for the F1 generation also was noted by Mastro and Schwalbe (1988). However, larvae irradiated at higher doses exhibited high mortality (Figure 1a). A lot of larvae receiving high doses were weakened and died too early to allow parasitoids to complete their development. Similar problems occurred in production of sterile F1 gypsy moths (Reardon et al. 1987). Partial sterility was noted in the F1 generation (Table 4). A similar effect was observed by Mastro and Schwalbe (1988). If the parents are irradiated (usually males), F1 sterility is a well-known consequence of irradiation. It is widely used in integrated pest management and SIT (Bloem and Carpenter 2001). Irradiation of eggs and larvae generally does not cause outright sterility. Larvae do not have developed gonads and they generally are not be injured by radiation. However, our 300 M. Zu´brik and J. Novotny´

experiments showed that there are effects of irradiation of the eggs and larvae on fertility of resulting adults. There is a question whether this is a result of the injury of some somatic cells in the larvae that are responsible for the creating the future mating organs or if this is a result of the general effect of the irradiation on the condition of the larvae and resulting adults. In conclusion, these experiments showed that a dose of 50 Gy would be appropriate for irradiation of gypsy moth eggs and larvae intended to be used to supplement naturally-occurring parasitoid, predator and pathogen populations and allow for an increase in the efficacy of the natural enemy complex in the field without a fear that they could reproduce and increase the pest’s threat.

Acknowledgements This study was supported by IAEA (grant No. 10849). We thank Dr. A. Robinson (Entomology Unit, Seibersdorf, Austria), MUDr. A. Rakytska (Roosevelt Hospital, B. Bystrica, Slovakia) for permission to use an irradiation source, Mr Ilkanic for technical assistance and Catherine Kwech for correction of English.

References Bell, R.A., Owens, C.D., Shapiro, M. and Tardif, J.R. (1981), ‘Development of Mass Rearing Technology’,inU.S. Department of Agriculture (USDA) Forest Service Technical Bulletin 1584, ‘The Gypsy Moth: Research Toward Integrated Pest Management’, eds. C.C. Doane and M.L. McManus, Agricultural Technical Bulletin 1584, Washington, DC: USDA, pp. 599Á633. Bloem, S., and Carpenter, J.E. (2001), ‘Evaluation of Population Suppression by Irradiated Lepidoptera and Their Progeny’, Florida Entomologist, 84, 165Á171. Campbell, F.T., and Schlarbaum, S.E. (1994), Fading Forests: North American Trees and the Threat of Exotic Pests, New York: Natural Resources Defense Council. Doane, C.C., and McManus, M.L. (eds.) (1981), ‘The Gypsy Moth: Research Toward Integrated Pest Management’, Agricultural Technical Bulletin 1584, Washington, DC: USDA, 757 pp. Dyby, S.D., and Sailer, R.I. (1999), ‘Impact of Low-Level Radiation on Fertility and Fecundity of Nezara viridula (Hemiptera: Pentatomidae)’, Journal of Economic Entomology, 4, 945Á953. Eichhorn, O. (1996), ‘Experimental Studies upon the Parasitoid Complex of the Gypsy Moth (Lymantria dispar L.) (Lep., Lymantriidae) in Lower Host Population in Eastern Austria’, Downloaded By: [Hendrichs, Jorge] At: 16:00 6 November 2009 Journal of Applied Entomology, 120, 205Á212. Fuester, R.W., Drea, J.J., Gruber, F., Hozer, H., and Mercadier, G. (1983), ‘Larval Parasites and Other Natural Enemies of Lymantria dispar (Lepidoptera: Lymantriidae) in Burgen- land, Austria, and Wurzburg, Germany’, Environmental Entomology, 12, 724Á737. Greany, P.D., and Carpenter, J.E. (2000), ‘Use of Nuclear Techniques in Biological Control’,in Area-Wide Control of Fruit Flies and Other Insect Pests, ed. K.H. Tan, Penang: Penerbit Universiti Sains Malaysia. Hajek, A.E., Humber, R.A., and Elkinton, J.S. (1995), ‘The Mysterious Origin of Entomophaga maimaiga in North America’, American Entomologist, 41, 31Á42. Hajek, A.E., McManus, M.L., and Delalibera, I. (2007), ‘A Review of Introductions of Pathogens and Nematodes for Classical Biological Control of Insects and Mites’, Biological Control, 41, 1Á13. Hoch, G., Zu´brik, M., Novotny´, J., and Schopf, A. (2001), ‘The Natural Enemy Complex of the Gypsy Moth, Lymantria dispar (Lep., Lymantriidae) in Different Phases of its Population Dynamics in Eastern Austria and Slovakia Á A Comparative Study’, Journal of Applied. Entomology, 125, 217Á227. Knipling, E.F. (1955), ‘Possibilities of Insect Control or Eradication Through the Use of Sexually Sterile Males’, Journal of Economic Entomology, 48, 459Á462. Biocontrol Science and Technology 301

Liebhold, A., and McManus, M. (1999), ‘The Evolving Use of Insecticides in Gypsy Moth Management’, Journal of Forestry,3,20Á23. Luckey, T.D. (1991), Radiation Hormesis, Boca Raton, FL: CRC Press. Maksimovicˇ, M. (1971a), ‘Effect of Cobalt-60 Irradiation of Male Pupae of the Gypsy Moth, Lymantria dispar L., on Biological Functions of Male Moths’, Sterility Principle for Insect Control or Eradication, FAO Report No: AGE-IAEA/SM/138/10, p. 15Á22, 7 tab., 6 ref. Maksimovicˇ, M. (1971b), ‘Application Of The Sterile-Male Technique To The Gypsy Moth, Lymantria Dispar L: A Field Trial’, Application of Induced Sterility for Control of Lepidopterous Populations. FAO Report No: AGE-STI/PUB/281,p.75Á80. 3 tab., 1 graph, 5 ref. Maksimovicˇ, M., and Sivcev, I. (1984), ‘Further Studies on the Numerical Increase of Natural Enemies of the Gypsy Moth (Lymantria dispar L.) in Forests’, Zeitschrift fur Angewandte Entomologie, 98, 332Á343. Mastro, V.C. (1993), ‘Gypsy Moth F1 Sterility Programme: Current Status’,inRadiation Induced F1 Sterility in Lepidoptera for Area-Wide Control. Proceedings of the Final Research Co-ordination Meeting; Panel Proceedings Series (IAEA); Research Co-Ordina- tion Meeting on Radiation Induced F1 Sterility in Lepidoptera for Area-Wide Control, Phoenix, AZ (USA), 9Á13 Sep 1991/Joint FAO/IAEA Div. of Nuclear Techniques in Food and Agriculture, Vienna (Austria), 1993, pp. 125Á129. Mastro, V.C., and Schwalbe, C.P. (1988), ‘Modern Insect Control: Nuclear Techniques and Biotechnology’,inProceedings of an International Symposium on Modern Insect Control, International Atomic Energy Agency, Vienna, pp. 15Á40. Mills, N.J. (1990), ‘Biological Control, a Century of Pest Management’, Bulletin Entomological Research, 80, 359Á362. Novotny´, J. (1989), ‘Bioregulovanie Pocˇetnosti Mn´ısˇky Vlkohlavej’, Lesn´ıcke sˇtu´die, 46, 7. Podgwaite, J.D., Reardon, R.C., Walton, G.S., and Witcosky, J. (1992), ‘Efficacy of Aerially Applied Gypchek† against Gypsy Moth (Lepidoptera: Lymantiidae) in the Appalachian Highlands’, Journal of Entomological Science, 27, 337Á344. Podgwaite, J.D., Dubois, N.R., Reardon, R., and Witcosky, J. (1993), ‘Retarding Outbreak of Low-Density Gypsy Moth (Lepidoptera: Lymantriidae) Populations with Aerial Applica- tions of Gypchek† and Bacillus thuringiensis’, Journal of Economic Entomology, 86, 730Á 734. Reardon, R.C., McManus, M., Kolodny-Hirsch, D., Tichenor, R., Raupp, M., Schwalbe, C., Webb, R., and Meckley, P. (1987), ‘Development and Implementation of a Gypsy Moth Integrated Pest Management Program’, Journal of Arboriculture, 13, 209Á216. Shapiro, M., and Dougherty, E. (1985), ‘Selection of Active Strains of the Gypsy Moth Nuclearpolyhedrosis Virus’, in David G. Grimble and Franklin B. Lewis, coordinators. Symposium Proceedings: Microbial control of Spruce Budworms and Gypsy Moths; (1984) April 10Á12; Windsor Locks, CT. Gen. Tech. Rep. NE-100. Broomall, PA: U.S. Department

Downloaded By: [Hendrichs, Jorge] At: 16:00 6 November 2009 of Agriculture, Forest Service, Northeastern Forest Experiment Station, pp. 115Á122. Shapiro, M., Robertson, J.L., Injac, M.G., Katagirik, and Bell, A. (1984), ‘Comparative Infectivities of Gypsy Moth (Lepidoptera: Lymantriidae) Nucleopolyhedrosis Virus Isolates from North America, Europe, and Asia’, Journal of Economic Entomology, 77, 153Á156. Smith, S.M. (1993), ‘Insect Parasitoids: A Canadian Perspective on Their Use for Biological Control of Forest Insect Pests’, Phytoprotection, 74, 51Á67. Wallace, D.R., and Smith, S.M. (1995), ‘Inundative Releases’,inForest Insect Pests in Canada, Canadian Forest Service, Ottawa, pp. 397Á409. Biocontrol Science and Technology, Vol. 19, S1, 2009, 303Á315

RESEARCH ARTICLE

F1 sterile insect technique: A novel approach for risk assessment of Episimus unguiculus (Lepidoptera: Tortricidae), a candidate biological control agent of Schinus terebinthifolius in the continental USA Onour E. Moeria*, James P. Cudaa, William A. Overholtb, Stephanie Bloemc, and James E. Carpenterd

aDepartment of Entomology and Nematology, University of Florida, Gainesville, FL, USA; bBiological Control Research and Containment Laboratory, University of Florida, Fort Pierce, FL, USA; cPlant Epidemiology and Risk Analysis Laboratory, USDA-APHIS-PPQ-CPHST, Raleigh, NC, USA; dUSDA-ARS, Crop Protection and Management Research Unit, Tifton, GA, USA

Federal regulations mandate that researchers in the field of classical weed biological control follow the precautionary principle when proposing the release of an organism that can affect our environment. However, laboratory risk assessment experiments often predict a much broader host range than that which occurs in the field. Because open-field tests are prohibited in the area of introduction, the application of the F1 sterile insect technique (F1SIT) could be used to conduct field testing in the proposed release area in a safe and temporary manner. In this study, we determined the minimum dose of radiation required to field test the tortricid Episimus unguiculus (Clarke), a candidate for biological control of Brazilian peppertree, Schinus terebinthifolius Raddi (Anacardiaceae) in Florida. Male and female virgin E. unguiculus adults were treated with increasing doses of gamma radiation and either inbred or outcrossed to non-treated E. unguiculus adults. Pairs of adults were placed in oviposition cages and allowed to mate and oviposit. Data from fecundity and fertility counts were recorded. The dose at which treated females were 100% sterile was 200 Gy. The dose at which F1 females and males were 100% sterile was 225 Gy. As the dose of radiation increased, there was an increase in sterility, a decrease in fecundity for both treated female crosses, and a higher ratio of F1 males to females. The F1 sterile Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009 insect technique (F1SIT) could be suitably applied to other areas of pest management, including risk assessment of potential lepidopteran biological control agents of invasive, exotic weeds.

Keywords: F1 sterile insect technique; Episimus unguiculus; weed biological control; inherited sterility; Brazilian peppertree

Introduction Before a candidate weed biological control agent can be released into the environ- ment, the host specificity of the organism must be demonstrated. Host-range testing is a process of screening potential biological control agents to minimize the risk of damage to non-target plant species. Tests involve several different plant species

*Corresponding author. Email: [email protected]

First Published Online 22 April 2009 ISSN 0958-3157 print/ISSN 1360-0478 online # 2009 Taylor & Francis DOI: 10.1080/09583150902741932 http://www.informaworld.com 304 O.E. Moeri et al.

closely and distantly related to the family of the targeted host plant (Wapshere 1974). Related and sometimes unrelated plant species that are economically important, endangered, or native are of high priority and tested (McEvoy 1996; Schaffner 2001). Two phases in host-range testing, no-choice and choice tests are performed in the laboratory (Marohasy 1998; Withers, Barton-Browne, and Stanley 1999). No-choice tests involve larval development and oviposition tests on a single non-target species. Choice tests expose the insect to two or more plants at the same time and typically include the host plant (McEvoy 1996; Schaffner 2001). Although these tests are designed to predict field host range, caged laboratory tests often overestimate host specificity because of unnatural behavior exhibited by some candidate biological control agents as a result of being in a caged environment. This type of behavior may produce ‘false positives’, or acceptance of plants as hosts that would not normally be accepted by the potential biological control agent in nature (Marohasy 1998). Open field testing provides a more natural assessment of the ecological host range of candidate biological control agents. Typically, these tests are performed in the native range of the target weed. However, there are serious limitations to using this particular approach; not the least of which is the need to import non-native test plant species that would most likely be prohibited from entering the country of origin of the biological control agent. Other issues include seasonal availability of the test plants or potential biological control agent, and mortality of the agent by specialized predators and parasitoids in the wild. A new approach for risk assessment that could be adopted for some candidate biological control agents is field testing in the area of introduction. Open field testing can be done in a safe, temporary manner for potential lepidopteran biological con- trol agents by using the F1 sterile insect technique (Greany and Carpenter 2000). Advantages of this approach include the exposure of the biological control agent to the actual environmental conditions it would experience if approved for release, prediction of true field host range, and ability to reverse releases of the biological control agent without permanent establishment if non-target damage is detected (Bax et al. 2001; Carpenter, Bloem, and Bloem 2001a). The F1 sterile insect technique (F1SIT) is similar to the sterile insect technique (SIT), except it uses a lower dose of radiation providing partial sterility and a reduced number of progeny. Lepidoptera are highly radio- resistant and require a large dose of radiation to ensure sterility (LaChance 1985). With Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009 the use of F1SIT, however, a lower dose could be applied resulting in a more competitive insect due to decreased somatic damage (North 1975). Various studies have used this approach to demonstrate control of populations of pest Lepidoptera, including the codling moth, Cydia pomonella (L.) (Bloem, Bloem, Carpenter, and Calkins 1999a,b), false codling moth Thaumatotibia leucotreta (Meyrick) (Bloem, Carpenter, and Hofmeyr 2003), potato tuber moth Phthorimaea operculella (Zeller) (Makee and Saour 1997, 2003) and cactus moth Cactoblastis cactorum (Berg) (Carpenter et al. 2001a,b; Hight, Carpenter, Bloem, and Bloem 2005; Tate, Carpenter, and Bloem 2007). Because application of F1SIT has been demonstrated in pest management programs, this approach has potential for evaluating the risks of releasing exotic lepidopteran candidates for weed biological control (Dunn 1978; Cullen 1990; Greany and Carpenter 2000; Tate et al. 2007). An example where F1SIT could be used for risk assessment purposes is in classical biological control of Brazilian peppertree, Schinus terebinthifolius Raddi (Anacardiaceae) in the continental USA. Brazilian peppertree is native to Brazil, Biocontrol Science and Technology 305

Paraguay, and Argentina and was introduced into Florida as an ornamental in 1898 (Austin 1978; Ewel, Ojima, Karl, and DeBusk 1982). This non-native plant is considered one of Florida’s worst invasive terrestrial weeds (Austin 1978; Morton 1978; Schmitz 1994). Brazilian peppertree is distributed widely throughout central and southern Florida and listed by the Florida Exotic Pest Plant Council (FLEPPC) as a ‘Category 1’ invasive weed because it is drastically altering native plant communities. It also is a major problem in Hawaii (Yoshioka and Markin 1991; Hight, Cuda, and Medal 2002), California (Randall 2000) and more recently Texas (Gonzalez and Christoffersen 2006). The environmental damage from Brazilian peppertree results from its ability to produce dense monospecific stands that shade out native plants and its allelopathic effects on competing vegetation (Gogue, Hurst, and Bancroft 1974; Morgan and Overholt 2005; Donnelly, Green, and Walters 2008). Field surveys have been conducted in Brazil since the mid 1980s to identify potential biological control agents of Brazilian peppertree (Bennett, Crestana, Habeck, and Berti-Filho 1990; Cuda, Habeck, Hight, Medal, and Pedrosa-Macedo 2004). One of these was a tortricid leafroller Episimus unguiculus Clarke ( E. utilis Zimmerman) (Lepidoptera: Tortricidae). Razowski and Brown (2008) recently synonymised E. utilis with its senior synonym E. unguiculus. In 1954, the leafroller was released in Hawaii, however, it was not found to severely affect Brazilian peppertree (Bennett et al. 1990; Yoshioka and Markin 1991; Habeck, Bennett, and Balciunas 1994), which was probably due to biotic interference from introduced parasitoids and predators of agricultural pests (Krauss 1963; Martin et al. 2004). Although E. unguiculus was not effective in controlling Brazilian peppertree in Hawaii, the insect may be less prone to biotic mortality from introduced and native parasitoids and predators if it were released in Florida (Martin et al. 2004). The biology of E. unguiculus and methodology for its laboratory rearing were investigated by Martin et al. (2004). Larvae of E. unguiculus inflict damage by feeding on leaflets, which can eventually lead to defoliation and a reduction in plant growth. Recently, a simulated herbivory study was conducted that showed growth and reproduction of Brazilian peppertree are significantly reduced by sustained defoliation (Treadwell and Cuda 2007). Replicated no-choice and multiple choice tests with E. unguiculus were completed

Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009 as part of the host specificity testing process, and the results showed this insect exhibited a wide host range under confined laboratory conditions (J.P. Cuda, unpublished data). At least 12 non-target native and cultivated plant species representing eight genera in two plant families were unexpectedly accepted as developmental hosts (J.P. Cuda, unpublished data). However, California peppertree (Schinus molle L.), which is the most closely related congener of Brazilian peppertree, was not attacked. We believe that the broad fundamental host range exhibited by E. unguiculus in the laboratory is not indicative of the field host specificity of this insect in a natural environment (Sheppard, van Klinken, and Heard 2005). This is supported by field surveys conducted in Brazil (J.P. Cuda, personal observation) and more recently in Argentina, where McKay et al. (2009) found E. unguiculus associated only with Brazilian peppertree. More importantly, there are no reports of E. unguiculus attacking non-target plants in Hawaii in spite of the establishment of the insect for over 50 years. For cases such as these, where the results of laboratory tests lead to unexpected false positives with lepidopterans, F1SIT could be an 306 O.E. Moeri et al.

additional tool to confirm the field host specificity of the candidate biological control agent. The objectives of the current study were to determine the minimum dose of radiation that would sterilize the F1 generation of E. unguiculus and to verify the effects of radiation in E. unguiculus.

Materials and methods Colony rearing Colony rearing procedures were similar to those described by Martin et al. (2004). Pairs of adult E. unguiculus moths (24Á48 h old) were placed on individual Brazilian peppertree plants planted in 3.8 L (1 gal) pots (2022.5 cm, heightdiam.). Each plant was enclosed in a clear acrylic cylinder (4515 cm, heightdiam.) with six evenly spaced ventilation holes (6.5 cm diam.). The top of the cylinder was covered with a sheer polyester fabric (Jo-Ann Fabrics† #449-1676 white casa organza) and all six circular ventilation holes were each covered with a mesh, screen size of 150 150 mm (Green.tek† Inc., Edgerton, WI). The sheer polyester fabric was fastened to the top of the cylinder by a metal ring clamp (14.3Á21.6 cm) and further sealed with a rubber band to prevent small larvae from escaping. Two additional access holes in the cylinder (2.5 cm diam.) were plugged with #5 rubber stoppers. Each cylinder was provided with a Gatorade† feeder which consisted of a 15-mL glass vial with a 5-cm piece of dental wick soaked in Gatorade† as a nectar source for the adults (Cuda, Deloach, and Robbins 1990; Martin et al. 2004). When approximately 90% larval defoliation of the plant was observed, bouquets of five stems of Brazilian peppertree leaves (1Á5 days old, field collected in Ft. Pierce, FL) in water filled plastic vials (40 mL) were placed near the top of the plant in each cylinder as needed. Plants used for colony rearing were sprayed twice weekly with an organic insecticide consisting of 15 mL (1 tbsp) each of isopropyl alcohol (70%), insecticidal oil, and Ivory† liquid soap mixed in 3.8 L (one gallon) of water to protect the plants from damage by aphids and other soft-bodied pests. The rearing room was maintained at 25.894.08C and 40Á70% RH as recorded by a Fisher Scientific† Thermo-Hygro† digital maximumÁminimum temperature and relative humidity

Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009 recording instrument. Temperature and relative humidity recorded within the cylinder were 24.993.88Cand60Á80% RH, respectively. A photoperiod of 14 h L:10 h D was maintained by a programmable timer connected to sets of two 60-cm 20-W fluorescent bulbs (one standard and one Gro-Lux†) per shelf of colony plants. Colony rearing and experiments with E. unguiculus were conducted at the University of Florida, Department of Entomology and Nematology Containment Facility, Gainesville, FL.

Radiation biology study The procedures used to study the radiation biology were based on methodologies developed for the codling moth, C. pomonella (Bloem et al. 1999a,b), false codling moth, T. leucotreta (Bloem et al. 2003), and cactus moth, C. cactorum (Carpenter et al. 2001a,b). The E. unguiculus moths were collected from the colony at the fifth instar (red larval stage) or the pupal stage and placed individually into separate clear Biocontrol Science and Technology 307

plastic 30 mL (1 oz) diet cups with a 3.5-cm piece of moistened filter paper added to each cup to maintain humidity. Temperature and relative humidity in the experiment room were 24.694.58C and 50Á80% RH, respectively, with a photoperiod of 14 h L:10 h D. The cups were checked each morning at the same time and adults were removed upon emergence. Male and female virgin E. unguiculus adults (B24 h old) were collected and individually exposed to gamma radiation in plastic snap-cap vials (12 mL) within an aluminum-lined cardboard canister (8.8 cm height7.5 cm diam.). Doses of 0, 50, 100, 150, 200, 250, and 300 Gy were administered by using a Cesium- 137 Gammacell† 1000 irradiator with a dose of 12Á13 Gy/min (Florida Accelerator Services & Technology (FAST), Florida Department of Agriculture and Consumer Services, Division of Plant Industry, Gainesville, FL). The dose rate was determined by a dosimetry study using Far West† dosimetry film. Two canisters each containing 10 of the plastic vials were placed on top of each other within the irradiator. Dosimetry film was placed in three different positions within four different vials (one from the top and bottom of each canister). Results indicated that the canister containing the plastic vials should only be positioned at the bottom of the irradiator in order to minimize the variance between the levels of irradiation. Five treated (T) male or female moths were placed inside a triangular waxed paper oviposition chamber (301912 cm) with an equal number of either treated (T) or non-treated (N) adult moths of the opposite gender. Five replicates of three different crosses (treatments) (NT,TN, and TT) were completed for each dose of radiation. The oviposition chamber was then placed inside a 3.8 L (1 gal) plastic sealable freezer bag (Ziploc†) to maintain relative humidity and suspended on a string line to maximize the use of the limited amount of space in the containment laboratory. Temperature and relative humidity within the oviposition chamber were recorded as 24.193.88C and 60Á80% RH, respectively. Each oviposition chamber included a 2-cm piece of cotton dental wick soaked in Gatorade† as a nectar source and a small leaf disc of Pistacia vera L. (2.42.4 cm). Due to inconsistent oviposition in preliminary experiments with non-treated moths, small leaf discs of Brazilian peppertree were substituted with P. vera to stimulate oviposition. A phytochemical study of the leaves and bark of Schinus

Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009 terebinthifolius had previously found that its compounds show a greater similarity to compounds isolated from Pistacia species than to those isolated from other species of Schinus (Campello and Marsaioli 1975). It was later determined that the leaf material was not a factor in the preliminary results, yet the P. vera leaf discs were used throughout the rest of the experiments for consistency. The moths were allowed to mate and lay eggs for two intervals of 5 days to take into account the 7-day average lifespan for the adults (Martin et al. 2004). After the first 5-day period, they were transferred to a new oviposition chamber. At the end of the 10-day period, the females were collected, preserved in ethyl alcohol (80%), and subsequently dissected to determine their mating status (presence of spermatophores or inflated bursa copulatrix) (Ferro and Akre 1975). The egg sheets were then incubated for a period of 7 days at 24.694.58C, 50Á80% RH, and a photoperiod of 14 h L:10 h D, which corresponded to the developmental time of the egg stage (Martin et al. 2004). The total number of eggs laid (fecundity) and the number of eggs that hatched (fertility) were then counted for each egg sheet per radiation dose. 308 O.E. Moeri et al.

Inherited sterility study Based on the findings from the radiation biology study, five doses of radiation were chosen for evaluation of the cross NT (non-treated femaletreated male). Offspring of this cross would achieve inherited sterility. Radiation doses included 125, 150, 175, 200, 225 Gy, and a control (0 Gy). The protocol was the same as

previously described, with the exception that the F1 egg sheets were each placed on a Brazilian peppertree plant in a 3.8-L (1 gal) pot enclosed by a clear acrylic

cylinder (4515 cm, heightdiam.) in order to rear the F1 generation. Average temperature and relative humidity recorded within the cylinder were 25.293.68C and 70Á90% RH, respectively. When the larvae hatched, they were allowed to develop on the plant. At the fifth instar (red larval stage) or the pupal stage, the insects were collected and each individual was placed in a separate clear plastic diet cup 30 mL (1 oz) with a 3.5-cm piece of moistened filter paper to maintain

humidity. Upon emergence, each F1 female or male was outcrossed with a non- treated adult moth of the opposite sex. These F1 crosses were made as single pairs (1 female1 male). The protocol for the single pair crosses was the same as previously described for the radiation biology study. Ten crosses of F1 females and males were attempted for each dose, but due to virgin females (found not to be mated upon dissection) and/or limited emergence of adults, there was a range of five to 12 replications for each gender per dose. Temperature and relative humidity in the experiment room were 25.694.48C and 40Á70% RH, respectively, with a photoperiod of 14 h L:10 h D. The temperature and relative humidity recorded inside the oviposition chamber were slightly higher, averaging 26.194.98C and 50Á60% RH, respectively.

Statistical analyses Radiation biology study In order to determine the effect of radiation dose on fecundity, linear regressions using fecundity as the response variable (Y) and radiation dose as the treatment variable (X) were performed. A separate model was fitted for each of the treatments. Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009 Effect of radiation dose on fertility was determined by performing simple linear regressions of radiation dose predicting fertility for each of the crosses. In some cases, a polynomial model was indicated by scatter plots of the data. Alpha level for all of the regressions was P0.05. Regression analyses were performed using S-plus† 7.0 for Windows† (Insightful 2005).

Inherited sterility study

To determine the effects of radiation on the reproductive biology of F1 offspring of irradiated males, linear and nonlinear regressions of radiation dose administered on fecundity and fertility of the offspring were performed, fitting polynomial models

where appropriate. Sex ratio was recorded for F1 males and analyzed using a simple regression. Alpha level for each of the factors was P0.05. Regression analyses were performed using S-plus† 7.0 for Windows† (Insightful 2005). Biocontrol Science and Technology 309

Results Radiation biology study Effects of the radiation treatments on adults of E. unguiculus were dependent upon the dose of radiation and gender irradiated. In irradiated males, no significant changes in fecundity of mated females were observed as radiation dose increased (NT; F3.67; df1, 31; P0.05; R2 0.11), whereas in irradiated females, significantly fewer eggs were laid as dose increased (TN; y71.67Á0.15x; F11.62; df1,32; PB0.05; R20.27; TT; y81.77Á0.21x; F16.85; df1, 31; PB0.05; R20.35) (Figure 1). Fertility for treated females also decreased with increasing radiation dose (TN; y60.010.68x0.0017x2; F56.31; df2, 30; PB0.05; R20.79; TT; y63.62Á0.76x0.0020x2; F53.74; df2, 30; PB0.05; R2  0.78) and the same effect was observed for treated males crossed with non-treated females (NT; y64.43Á0.20x; F57.35; df1,31; PB0.05; R20.65) (Figure 2). Additionally, the dose of radiation at which treated females were found to be 100% sterile was 200 Gy (residual fertility of 0.1% at 150 Gy), whereas males irradiated at 200 Gy still had a residual fertility of 18%. Mating was confirmed in all adult female moths used in the experiments as determined by the presence of spermatophores or inflated bursa copulatrix (Ferro and Akre 1975).

Inherited sterility study

With respect to the fecundity for F1 females, there was no significant relationship between the dose of radiation administered to the treated male in the parental cross and fecundity observed in the F1 generation (F1 N; F0.07; df1, 42; P0.05; 2 2 R 0.002; NF1; F1.11; df2, 52; P0.05; R 0.04) (Figure 3). Percent egg Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009

Figure 1. Fecundity (mean number of eggs laid) per mated female of Episimus unguiculus adults for three crosses (TT,TN and NT) treated with increasing doses of gamma radiation. 310 O.E. Moeri et al.

Figure 2. Fertility (mean percentage of eggs that hatched) of Episimus unguiculus adults for three crosses (TT,TN and NT) treated with increasing doses of gamma radiation.

hatch for treatments with the F1 males and females both declined with an increased 2 dose of radiation (NF1 ; y55.16Á0.60x0.0016x ; F40.31; df2, 52; 2 PB0.05; R 0.61; F1N; y63.85Á0.32x; F58.28; df1, 43; PB0.05; 2 R 0.58) (Figure 4). For both the F1 female and male treatments, 100% sterility was achieved at 225 Gy, the minimum dose at which no viable offspring could survive. Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009

Figure 3. Fecundity (mean number of eggs laid) of F1 crosses (F1 N,NF1 )of Episimus unguiculus adults when increasing doses of radiation were administered to parental males. Biocontrol Science and Technology 311

Figure 4. Fertility (mean percentage of eggs that hatched) of F1 crosses (F1 N,N F1 )ofEpisimus unguiculus adults as a result of radiation administered to parental males.

Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009 Figure 5. Percentage of F1 Episimus unguiculus adult males as a result of radiation administered to Episimus unguiculus parental males.

In addition, an increasing ratio of F1 males to females was positively correlated with an increase in radiation dose, although the relationship was marginally significant (y54.980.12x; F7.52; df1, 4; P0.05; R20.65) (Figure 5).

Discussion

F1SIT has been used in several studies to control various pest Lepidoptera. The technique provides a safe, environmentally friendly approach to pest management. Early studies by Proverbs (1962), who first documented partial sterility in the codling moth, found that when males were partially sterilized and mated to wild females, the progeny number was reduced, mostly male, and highly sterile. Subsequent studies by North (1975) and LaChance (1985) comparing the use of partial sterility with 312 O.E. Moeri et al.

complete sterility in Lepidoptera determined that a partial sterilizing dose of radiation would increase competitiveness, possibly cause a delay in development, and lower sperm quality in the F1 generation. Recent laboratory and field studies have confirmed these effects in the codling moth (Bloem et al. 1999a,b) and false codling moth (Bloem et al. 2003), both members of the same family (Tortricidae) as E. unguiculus. Results of this study were similar to results found in previous studies. Higher doses of radiation resulted in an increase in sterility, a higher ratio of F1 males to females, and a declining trend in fecundity for both treated female crosses. Irradiated females of E. unguiculus were found to be 100% sterile at 200 Gy, which is similar to the female false codling moth but more radioresistant than the female codling moth in which complete sterility was observed at 100 Gy. When treated males of E. unguiculus were mated with non-treated females, sterility of the F1 generation was similar to that reported for other tortricids. In particular, the dose at which E. unguiculus was found to be 100% sterile was 225 Gy, whereas the dose for the codling moth was 250 Gy (Bloem et al. 1999a,b), and the range of partial sterility for the false codling moth was 150Á200 Gy (Bloem et al. 2003). We therefore determined that a radiation dose of 225 Gy will provide 100% sterility and ensure a reduced number of progeny which are more sterile than their parents and mostly male. Our results clearly show that F1SIT could be appropriately applied to other areas of pest management, including risk assessment of potential lepidopteran biological control agents of invasive, exotic weeds. A relevant example is the invasive Brazilian peppertree in Florida and E. unguiculus, an established biological control agent of Brazilian peppertree in Hawaii where no non-target impacts have been documented. Because laboratory host specificity tests showed that the fundamental host range of E. unguiculus is broader than expected (J.P. Cuda, unpublished data), the cage environment in laboratory screening tests potentially inhibited the normal behavior of the insects and enabled them to accept plants that they would not normally recognize in nature (Withers et al. 1999). Therefore, field testing in the proposed area of introduction would provide more accurate results. In this study, we found that a dose of 225 Gy can be applied to E. unguiculus adult male moths and upon mating with non-treated female moths; complete sterility in the F1 generation is assured. Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009 Based on fecundity results of females mated with irradiated male parents, no significant relationship was found between treatment dose and fecundity of females, therefore suggesting that the number of eggs laid would be similar to non-irradiated moths. However, fertility recorded in the irradiated male moth treatments (NT) would be greatly reduced. Normal oviposition behavior as well as larval damage and feeding would occur under natural conditions except that the F1 generation would be unable to reproduce. Performance of irradiated E. unguiculus males should be similar to non-treated males based on results of previous studies examining the effects of radiation on other tortricid moths (Bloem et al. 1999a,b, 2003). An additional safety factor of the technique is the fact that most of the F1 progeny will be males, therefore limiting the number of matings. Using F1SIT in addition to laboratory host-range testing can provide a tempo- rary and reversible way to test potential biological control agents in the proposed area of release as proposed by Bax et al. (2001) and Greany and Carpenter (2000). Further studies will be needed to address the performance of irradiated biological Biocontrol Science and Technology 313

control agents including oviposition, larval feeding preferences and survival, and host-finding behavior.

Acknowledgements We thank Dr Burrell Smittle, Carl Gillis, and Suzanne Fraser at The Florida Accelerator Services and Technology Gainesville, FL for their support and assistance in the irradiation of the E. unguiculus moths. We thank Judy Gillmore and students in the Weed Biological Control Laboratory, Entomology & Nematology Department, Institute of Food and Agricultural Sciences, University of Florida, for their support and maintenance of E. unguiculus colonies. We also thank Veronica Manrique for supplying Brazilian peppertree plants and Sean McCann for help with statistical analysis. Finally, we thank the South Florida Water Management District, the Florida Fish and Wildlife Conservation Commission (formerly the Department of Environmental Protection), and The University of Florida, IFAS, Center for Natural Resources for supporting this research.

References Austin, D.F. (1978), ‘Exotic Plants and their Effects in Southeastern Florida’, Environmental Conservation,5,25Á34. Bax, N., Carlton, J.T., Mathews-Amos, A., Haedrich, R.L., Howarth, F.G., Purcell, J.E., Rieser, A., and Gray, A. (2001), ‘The Control of Biological Invasions in the World’s Oceans’, Conservation Biology, 15, 1234Á1246. Bennett, F.D., Crestana, L., Habeck, D.H., and Berti-Filho, E. (1990), ‘Brazilian Peppertree Á Prospects for Biological Control’,inProceedings of VII International Symposium on Biological Control of Weeds, ed. E.S. Delfosse, Rome, Italy: Ministero dell’Agriculture e delle Foreste, pp. 293Á297. Bloem, S., Bloem, K.A., Carpenter, J.E., and Calkins, C.O. (1999a), ‘Inherited Sterility in Codling Moth (Lepidoptera: Tortricidae): Effect of Substerilizing Doses of Radiation on Field Competitiveness’, Environmental Entomology, 28, 669Á674. Bloem, S., Bloem, K.A., Carpenter, J.E., and Calkins, C.O. (1999b), ‘Inherited Sterility in Codling Moth (Lepidoptera: Tortricidae): Effect of Substerilizing Doses of Radiation on Insect Fecundity, Fertility, and Control’, Annals of the Entomological Society of America, 92, 222Á229. Bloem, S., Carpenter, J.E., and Hofmeyr, J.H. (2003), ‘Radiation Biology and Inherited Sterility in False Codling Moth (Lepidoptera: Tortricidae)’, Journal of Economic Entomol- ogy, 96, 1724Á1731. Campello, J.P., and Marsaioli, A.J. (1975), ‘Terebinthifolic Acid and Bauerenone: New Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009 Triterpenoid Ketones from Schinus terebinthifolius’, Phytochemistry, 14, 2300Á2302. Carpenter, J.E., Bloem, K.A., and Bloem, S. (2001a), ‘Applications of F1 Sterility for Research and Management of Cactoblastis cactorum (Lepidoptera: Pyralidae)’, Florida Entomologist, 84, 531Á536. Carpenter, J.E., Bloem, S., and Bloem, K.A. (2001b), ‘Inherited Sterility in Cactoblastis cactorum (Lepidoptera: Pyralidae)’, Florida Entomologist, 84, 537Á542. Cuda, J.P., Deloach, C.J., and Robbins, T.O. (1990), ‘Population Dynamics of Melipotis indomita (Lepidoptera: Noctuidae), an Indigenous Natural Enemy of Mesquite, Prosopis spp.’, Environmental Entomology, 19, 415Á422. Cuda, J.P., Habeck, D.H., Hight, S.D., Medal, J.C., and Pedrosa-Macedo, J.H. (2004), ‘Brazilian Peppertree, Schinus terebinthfolius: Sumac Family-Anacardiaceae’,inBiological Control of Invasive Plants in the United States, eds. E. Coombs, J. Clark, G. Piper and A. Cofrancesco, Corvallis, OR: Oregon State University Press, pp. 439Á441. Cullen, J.M. (1990), ‘Current Problems in Host-specificity Screening’, in Proceedings of the VII International Symposium on Biological Control of Weeds, ed. E.S. Delfosse, Rome, Italy: Ministero dell’Agriculture e delle Foreste, pp. 27Á36. Donnelly, M.J., Green, D.M., and Walters, L.J. (2008), ‘Allelopathic Effects of Fruits of the Brazilian Pepper Schinus terebinthifolius on Growth, Leaf Production and Biomass of 314 O.E. Moeri et al.

Seedlings of the Red Mangrove Rhizophora mangle and the Black Mangrove Avicennia germinans’, Journal of Experimental Marine Biology and Ecology, 357, 149Á156. Dunn, P.H. (1978), ‘Shortcomings in the Classic Tests of Candidate Insects for the Biocontrol of Weeds’,inProceedings of the IV International Symposium on Biological Control of Weeds, ed. T.E. Freeman, Gainesville, FL: University of Florida, pp. 51Á56. Ewel, J., Ojima, D., Karl, D., and DeBusk, W. (1982). ‘Schinus in Successional Ecosystems of Everglades National Park’, http://fulltext10.fcla.edu/cgi/t/text/text459idx?cfeol&idno FI00835676&formatpdf. Ferro, D.N., and Akre, R.D. (1975), ‘Reproductive Morphology and Mechanics of Mating of the Codling Moth, Laspeyresia pomonella’, Annals of the Entomological Society of America, 68, 417Á424. Gogue, G.J., Hurst, C.J., and Bancroft, L. (1974), ‘Growth Inhibition by Schinus terebinthi- folius’, HortScience, 9, 301. Gonzalez, L., and Christoffersen, B. (2006). ‘The Quiet Invasion: A Guide to Invasive Plants of the Galveston Bay Area’, http://www.galvbayinvasives.org/. Greany, P.D., and Carpenter, J.E. (2000), ‘Use of Nuclear Techniques in Biological Control’,in Proceedings: Area-Wide Control of Fruit Flies and Other Insect Pests. International Conference on Area-Wide Control of Insect Pests, and the 5th International Symposium on Fruit Flies of Economic Importance 2000, ed. K.H. Tan, Penang, Malaysia: Penerbit Universiti Sains Malaysia, pp. 221Á227. Habeck, D.H., Bennett, F.D., and Balciunas, J.K. (1994), ‘Biological Control of Terrestrial and Wetland Weeds’,inPest Management in the Subtropics: Biological Control Á A Florida Perspective, eds. D. Rosen, F.D. Bennett and J.L. Capinera, Andover, UK: Intercept, pp. 523Á547. Hight, S.D., Cuda, J.P., and Medal, J.C. (2002), ‘Brazilian Peppertree’,inBiological Control of Invasive Plants in the Eastern United States, eds. R.G. Van Driesche, S. Lyon, B. Blossey, M. Hoddle and R. Reardon, Morgantown, WV: USDA Forest Service, pp. 311Á321. Hight, S.D., Carpenter, J.E., Bloem, S., and Bloem, K.A. (2005), ‘Developing a Sterile Insect Release Program for Cactoblastis cactorum (Berg) (Lepidoptera: Pyralidae): Effective Overflooding Ratios and Release-Recapture Field Studies’, Environmental Entomology, 34, 850Á856. Insightful (2005), ‘S-Plus Guide to Statistics’, http://www.insightful.com/support/splus70win/ uguide.pdf. Krauss, N. (1963), ‘Biological Control Investigations on Christmas Berry (Schinus terebinthi- folius) and Emex (Emex spp.)’, Proceedings of the Hawaiian Entomological Society, 18, 281Á285. LaChance, L.E. (1985), Genetic Methods for the Control of Lepidopteran Species: Status and Potential, Washington, DC: USDA/ARS. Makee, H., and Saour, G. (1997), ‘Inherited Effects in F1 Progeny of Partially Sterile Male Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009 Phthorimaea operculella (Lepidoptera: Gelechiidae)’, Journal of Economic Entomology, 90, 1097Á1101. Makee, H., and Saour, G. (2003), ‘Noninherited Sterility in Irradiated Phthorimaea operculella Females’, Journal of Applied Entomology, 127, 489Á493. Marohasy, J. (1998), ‘The Design and Interpretation of Host-specificity Tests for Weed Biological Control with Particular Reference to Insect Behaviour’, Biocontrol News and Information, 19, 13Á20. Martin, C., Cuda, J., Awadzi, K., Medal, J., Habeck, D., and Pedrosa-Macedo, J. (2004), ‘Biology and Laboratory Rearing of Episimus utilis (Lepidoptera: Tortricidae), a Candidate for Classical Biological Control of Brazilian Peppertree (Anacardiaceae) in Florida’, Environmental Entomology, 33, 1351Á1361. McEvoy, P.B. (1996), ‘Host Specificity and Biological Pest Control’, BioScience, 46, 401Á405. McKay, F., Oleiro, M., Walsh, G.C., Gandolfo, D., Wheeler, G.S., and Cuda, J.P. (2009), ‘Natural Enemies of Brazilian Peppertree (Schinus terebinthifolius: Anacardiaceae) from Argentina: Their Possible Use for Biological Control in the USA’, Florida Entomologist (accepted). Morgan, E.C., and Overholt, W.A. (2005), ‘Potential Allelopathic Effects of Brazilian Pepper (Schinus terebinthifolius Raddi, Anacardiaceae) Aqueous Extract on Germination and Biocontrol Science and Technology 315

Growth of Selected Florida Native Plants’, Journal of the Torrey Botanical Society, 132, 11Á15. Morton, J.F. (1978), ‘Brazilian Peppertree Á Its Impact on People, Animals and the Environment’, Economic Botany, 32, 353Á359. North, D.T. (1975), ‘Inherited Sterility in Lepidoptera’, Annual Review of Entomology, 20, 167Á182. Proverbs, M.D. (1962), ‘Progress on the Use of Induced Sexual Sterility for the Control of the Codling Moth Carpocapsa pomonella’, Proceedings of the Entomological Society of Ontario, 92, 5Á11. Randall, J.J. (2000), ‘Schinus terebinthifolius Raddi’,inInvasive Plants of California’s Wildlands, eds. C.C. Bossard, J.M. Randall and M.C. Hoshovsky, Berkeley and Los Angeles, CA: University of California Press, pp. 282Á286. Razowski, J., and Brown, J.W. (2008), ‘New Species, New Combinations, and New Synonymies in Neotropical Episimus Walsingham, 1892 (Lepidoptera: Tortricidae: Ole- threutinae)’, Acta Zoologica Cracoviensia, 51B, 83Á144. Schaffner, U. (2001), ‘Host Range Testing of Insects for Biological Weed Control: How Can It Be Better Interpreted?’, BioScience, 51, 951Á959. Schmitz, D.C. (1994), ‘The Ecological Impact of Non-indigenous Plants in Florida’,inAn Assessment of Non-indigenous Species in Florida’s Public Lands, Tallahassee, FL: Florida Department of Environmental Protection, pp. 10Á17. Sheppard, A.W., van Klinken, R.D., and Heard, T.A. (2005), ‘Scientific Advances in the Analysis of Direct Risks of Weed Biological Control Agents to Nontarget Plants’, Biological Control, 35, 215Á226. Tate, C.D., Carpenter, J.E., and Bloem, S. (2007), ‘Influence of Radiation Dose on the Level of F1 Sterility in the Cactus Moth, Cactoblastis cactorum (Lepidoptera: Pyralidae)’, Florida Entomologist, 90, 537Á544. Treadwell, L.W., and Cuda, J.P. (2007), ‘Effects of Defoliation on Growth and Reproduction of Brazilian Peppertree (Schinus terebinthifolius)’, Weed Science, 55, 137Á142. Wapshere, A.J. (1974), ‘A Strategy for Evaluating the Safety of Organisms for Biological Weed Control’, Annals of Applied Biology, 77, 201Á211. Withers, T.M., Barton-Browne, L., and Stanley, J. (1999), Host Specificity Testing in Australasia: Towards Improved Assays for Biological Control, Brisbane, Australia: Scientific Publishing. Yoshioka, E., and Markin, G. (1991), ‘Efforts of Biological Control of Christmas Berry (Schinus terebinthifolius) in Hawaii’,inProceedings of the Symposium of Exotic Pest Plants, ed. T. Center, Miami, FL, pp. 377Á385. Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009 Biocontrol Science and Technology, Vol. 19, S1, 2009, 317Á333

RESEARCH ARTICLE Oviposition preference of Cactoblastis cactorum (Lepidoptera: Pyralidae) in caged choice experiments and the influence on risk assessment of F1 sterility C.D. Tatea, S.D. Hightb, and J.E. Carpenterc*

aUSDA-APHIS-PPQ-CPHST, CPHST Laboratory, Phoenix, AZ, USA; bUSDA-ARS- CMAVE at Center for Biological Control, FAMU, Tallahassee, FL, USA; cUSDA-ARS Crop Protection & Management Research Unit, Tifton, GA, USA

Releases of lepidopteran biological control agents have successfully controlled invasive weed species. However, issues with non-target effects of released exotic agents have resulted in stringent pre-release host specificity testing. Use of inherited (F1) sterility, a radiation induced genetic condition that can cause sterility in the F1 generation, could further assess the risk of non-target effects and negative ecological effects under field conditions. This technique may aid in approving potentially effective and safe biological control agents for release. The unintentional arrival of the cactus moth, Cactoblastis cactorum, into the United States provides a unique opportunity to evaluate the potential of F1 sterility. This study was conducted to assess host oviposition preferences of C. cactorum females mated with irradiated and non-irradiated males for cactus species from seven groups based on location, cactus growth characteristics (plant structure), spine densities, genera, and economic importance. No significant differences in female host preference were observed between females mated with normal or irradiated males. Lack of significant differences in oviposition preference suggests that inherited (F1) sterility has potential as a risk assessment tool for potential exotic biological control agents for invasive weed species. Evaluation of the overall analysis of female C. cactorum host preference revealed that significantly different numbers of eggsticks were oviposited on cactus species. In whole plant cages, significantly more eggsticks were oviposited on Opuntia corallicola than any other species, and in cladode cages, significantly more eggsticks were oviposited on Opuntia humifusa than all other species except Opuntia pusilla. Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009 Keywords: cactus moth; inherited sterility; invasive species; risk assessment; weed biological control

Introduction Lepidoptera species have been successful biological control agents of invasive weeds. For example, the cactus moth, Cactoblastis cactorum (Berg) (Lepidoptera: Pyralidae), has been cited as one of the world’s most successful biological control programs against weeds (Dodd 1940; Petty 1948; Moran and Zimmerman 1984). Release and establishment of C. cactorum in Australia, South Africa, the Caribbean, and many other countries resulted in significant reductions in several invasive and native pest Opuntia spp. (Dodd 1940; Simmonds and Bennett 1966; Julien and

*Corresponding author. Email: [email protected]

First Published Online 1 May 2009 ISSN 0958-3157 print/ISSN 1360-0478 online This work was authored as part of the Contributor’s official duties as employees of the United States Government and is therefore a work of the United States Government. In accordance with 17 U.S.C. 105 no copyright protection is available for such works under U.S. law. DOI: 10.1080/09583150902814507 http://www.informaworld.com 318 C.D. Tate et al.

Griffith 1998; Pemberton and Cordo 2001). However, the unintentional arrival of C. cactorum into the United States (specifically the Florida Keys) through either natural dispersal (Zimmerman, Moran, and Hoffmann 2001; Stiling 2002) or human facilitated introductions, e.g., Opuntia cactus commerce (Pemberton 1995) or human-aided dispersal (Stiling and Moon 2001), has become an example of potential risks inherent in biological control agents. Cactoblastis cactorum continues to migrate northward and westward since arriving in the Florida Keys in 1989 (Hight, Bloem, Bloem, and Carpenter 2003). Populations have established as far north as Bull Island, near Charleston, South Carolina and currently, as far west as the Mississippi barrier Islands of Petit Bois and Horn (Hight and Carpenter 2009). The rate of migration and establishment of C. cactorum populations poses a potential threat to Opuntia diversity throughout North America and the Caribbean basin, and to wild and cultivated Opuntia spp. in the southwestern US and Mexico (Strong and Pemberton 2000; Sobero´n, Golubov, and Sarukhan 2001; Perez-Sandi 2001). Presence of C. cactorum in the southeastern US and its dual status, both as a beneficial and pest species, provides a unique model system to conduct inherited sterility proof-of-concept studies using this non-native insect species as a test subject. Inherited sterility (F1 sterility) is a radiation-induced genetic condition that can cause sterility in the F1 generation (LaChance 1985). Insects carrying this condition are not able to reproduce in the field. This unique genetic feature may be used to evaluate, under field conditions, the risks of releasing agents that may cause non- target and negative ecological effects, and to increase the chance of approving valuable and safe biological control agents for release (Greany and Carpenter 2000; Tate, Carpenter, and Bloem 2007; Moeri, Cuda, Overholt, Bloem, and Carpenter 2009). In a classical weed biological control program, an agent’s host specificity must first be defined before it can be considered for release into its new homeland. Over the years, practitioners of biological control, government regulators, and the public have required increasingly narrower host specificity ranges and insurances that non- target native species will not be harmed before a non-native agent is released into the wild. Weed biological control has a rigorous protocol of evaluating host range of biological control agents (Wapshere 1974; Marohasy 1998; Briese and Walker 2002; Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009 Briese 2003). Experiments are designed to evaluate oviposition preference of the females and performance of the larvae (McEvoy 1996). Testing regimes involve quarantine studies on caged native plants in the region of introduction to identify the host range of potential biological control agents. An increasing number of host specificity evaluations include open field tests in the agents’ homeland (Briese, Zapater, Andorno, and Perez-Camargo 2002). However, both of these types of tests have limitations. Quarantine studies are conducted inside cages in a laboratory and have been shown to interfere with the normal behavior and biology of the biological control agent (Marohasy 1998). Open field studies in the agents’ native range are conducted in more realistic settings, but these studies are fraught with problems relating to the availability and seasonality of test plants from outside the native range of the biological control agent. A potential new approach for evaluating host specificity is to conduct open field tests in the region where the target weed has become invasive with promising biological control agents that are rendered safe through F1 sterility (Greany and Carpenter 2000; Moeri et al. 2009). Biocontrol Science and Technology 319

Before a non-native insect is tested in an open field setting, the sterility of the released individuals must be proven. Carpenter, Bloem, and Bloem (2001a) evaluated gamma radiation dose effects on the fecundity and fertility of C. cactorum.From these evaluations, a minimum dose at which irradiated females and males were 100% sterile when mated to fertile males and females was established and the occurrence of F1 sterility in this species was verified. Additionally, Carpenter et al. (2001a) suggested doses between 100 and 200 Gray (Gy) would allow for maximum production of F1 adults while inducing full sterility in the F1 generation. When determining the usefulness of an F1 sterile open field-testing protocol, another important aspect is to ensure that the oviposition preference of females mated to irradiated males is the same as females mated to normal (unirradiated) males. Unaltered oviposition host preference would suggest that F1 sterility is potentially a risk assessment tool for evaluating the host range of non-native biological control agents in a more realistic setting. Open field evaluations of F1 sterile potential biological control agents in their non-native range would be free of risks to non-target species. In this study, we evaluated host preference in greenhouse cages of female C. cactorum mated with either normal or with irradiated male moths. Cactus plants native and introduced to the southeastern US, growth characteristics (plant structure), spine densities, non-Opuntia cactus genera, and economic importance also are considered and discussed in this study.

Materials and methods Laboratory reared C. cactorum originating from wild populations collected on the causeway connecting the Georgia mainland with Jekyll Island, GA and along the Florida Gulf Coast were used in this study. Larvae were reared on Opuntia ficus- indica (L.) P. Miller cladodes (flattened green stems of Opuntia spp.) in rectangular plastic containers (342413 cm) at 268C, 70% relative humidity (RH), and a photoperiod of 12 h L:12 h D (Carpenter et al. 2001a). Cocoons were collected 2Á3 days after initiation of pupation. Cocoon silk was removed from pupae with a 5% sodium hypochlorite (NaOCL) solution. Pupae were sorted by gender, with male and female pupae held separately in 475 mL plastic cups at the above conditions. Just Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009 before emergence, male and female pupae were placed inside separate screen cages (30.530.530.5 cm) and allowed to emerge at room temperature (23918C) (Carpenter, Bloem, and Bloem 2001b). Virgin male and female C. cactorum were individually placed in separate 30 mL plastic containers (BioServ, Frenchtown, NJ, USA) and held at 68C until quiescent (30 min). When the moth flight risk appeared low, cohorts of 10 or 20 male moths were placed in separate fabric containers (six containers of each density). A piece of paper towel was inside each container to provide a place for moths to rest, and an additional piece was placed on top of the container to enclose the moths. Half of the containers at each density (three containers or 90 moths) were irradiated at 200 Gy in a Cobalt60 gammacell 220 irradiator. Equal numbers of female moths were placed in all containers with irradiated and control males. Containers were transported to the greenhouse for release into oviposition cages. Whole potted cactus plants were evaluated in large cages (1 m3) and excised cactus cladodes were evaluated in small cages (30 cm3). Whole plants were used to 320 C.D. Tate et al.

evaluate plant structure and growth characteristics as possible factors that influence moth oviposition preference. Excised cladodes were used to examine the presence of other factors that may influence moth oviposition preference, such as plant surface properties and plant volatiles. Choice tests were used to evaluate oviposition preference of C. cactorum. Potted plants and cladodes were randomly positioned in each cage. A single container with 20 pairs of C. cactorum moths was released in each large cage and a single container of 10 pairs of C. cactorum moths was released in each small cage. Treatments were randomly assigned to large and small cages, resulting in a completely randomized design with three replications.

Cactoblastis cactorum oviposition preference for five native and one introduced Opuntia species found in Florida Cactoblastis cactorum oviposition was evaluated in greenhouse cages on six Opuntia species commonly found in Florida: O. corallicola (Small) Werdermann, O. ficus- indica, O. humifusa (Rafinesque) Rafinesque, O. pusilla (Haworth) Haworth, O. stricta (Haworth) Haworth, and O. triacantha (Wildenow) Sweet. Irradiated (200 Gy) and normal male moths paired with normal females at densities previously described were placed in large and small cages for 5 days. Female C. cactorum stack their eggs on top of one another so that the group of eggs resembles a cactus spine, and the spine mimic is called an eggstick. Eggsticks were collected daily from each Opuntia spp. in each test cage, and the number of eggsticks and eggs within each eggstick recorded.

Cactoblastis cactorum oviposition preference for cactus species in six select groupings Female C. cactorum moth oviposition preference was evaluated in greenhouse cages on cactus species in the Genera Opuntia, Cylindropuntia, and Harrisia. Six groups of four cacti were selected to evaluate several different characteristics (Table 1). The groups were distinguished as follows: Group 1 Á comparison of Opuntia spp. with similar growth characteristics, i.e., tall upright species; Group 2 Á comparison of Opuntia spp. with differing spine and glochid complements Á virtually lacking spines and glochids to dense glochids or dense spines; Group 3 Á comparison of different Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009 species in the Genera Opuntia, Cylindropuntia,andHarrisia; Group 4 Á comparison of documented C. cactorum hosts and non-hosts; Group 5 Á comparison of cactus species present in Florida and species native to the southwestern US and Mexico; Group 6 Á comparison of economically important Opuntia species. Irradiated (200 Gy) and normal male moths paired with normal females in densities previously described were placed in large and small cages for 3 days. The number of days for oviposition was reduced from the previous experiment because 70% of eggsticks were oviposited in the first 3 days. The number of eggsticks collected after 3 days and the number of eggs per eggstick were recorded by host plant species.

Analyses Cactoblastis cactorum oviposition preference for six species of Opuntia were ranked based on numbers of eggsticks oviposited on each plant (Proc Rank procedure of Biocontrol Science and Technology 321

Table 1. Groupings of cactus species used in experiments evaluating Cactoblastis cactorum oviposition preference for six select groups of four cactus species.

Group no. and characteristic Test species Categorical trait evaluated

1 Opuntia spp. with similar O. cochenillifera Tall plant, flat stems, lack spines growth habit O. corallicola Tall plant, flat stems, dense spines O. falcata Tall plant, flat stems, lack spines O. stricta Tall plant, flat stems, intermediate spine 2 Opuntia spp. with different O. falcata Lack spines/glochids spine/glochid densities O. microdasys Dense glochids O. polyacantha Dense spines O. stricta Intermediate spine/glochid density 3 Species in the Genera Opuntia, C. acanthocarpa Round stems Cylindropuntia,&Harrisia C. spinosior Round stems H. fragrans Angular stems O. stricta Flat stems 4 Documented host and non-host C. spinosior Non-host species H. fragrans Non-host O. streptacantha Host O. stricta Host 5 Species native or not native to FL H. fragrans Native to Florida O. dellinii Native to Florida C. spinosior Native to southwestern US O. streptacantha Native to southwestern US 6 Economically important O. engelmannii Livestock food & ornamental Opuntia spp. O. ficus-indica Human vegetable & fruit; ornamental O. streptacantha Human fruit

Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009 O. stricta Wildlife food

The categorical trait evaluated for each cactus species under each grouping is identified.

SAS†, SAS Institute Inc. 1999). Rankings were based on a scale from 1 to 24, with the lowest number representing the highest rank Á the species that received the most eggsticks. Rankings included all possible treatment combinations among species, radiation, and cage size. Resulting rankings, numbers of eggsticks, and numbers of eggs per eggstick were analyzed with ANOVA using Proc Mixed procedure of SAS† (SAS Institute Inc. 1999). Cage type, cactus species, and treatment (irradiated or normal males) were treated as fixed effects, while replication was treated as a random effect. Oviposition preferences for six select groups of four cactus species separated according to growth habit, spine density, family or genus classification, and economic importance also were ranked (1Á18) based on numbers of eggsticks 322 C.D. Tate et al.

oviposited on each plant using Proc Rank procedure of SAS† (SAS Institute Inc. 1999). Resulting rankings, proportions of eggsticks, and numbers of eggs per eggstick were analyzed with ANOVA using Proc Mixed procedure of SAS† (SAS Institute Inc. 1999). Cage type, group, cactus species, and treatment (irradiated or normal male) were treated as fixed effects, while replication was treated as a random effect. Degrees of freedom were adjusted using Satterthwaite approximation method in all experiments. Means were separated using TukeyÁKramer mean separation procedures. Data were log transformed when standard deviations were proportional to the mean (heteroscedasticity) and distributions were skewed.

Results No significant differences in oviposition preference were found between the two main treatments, C. cactorum females mated to normal males versus C. cactorum females mated to irradiated males. Differences were not found for females that mated with irradiated or unirradiated males in any of the three measured ovipositional parameters (ranking of cactus host species that received eggsticks, proportion of eggsticks oviposited per host species, or mean number of eggs per eggstick oviposited on cactus species) (Table 2). In addition, no significant interactions were found between the main treatment effects and any other fixed effect (cage type or cactus species). Because the measurements between the two treatments were not different, data were pooled and presented as an analysis of oviposition host specificity for C. cactorum.

Cactoblastis cactorum oviposition preference for five native and one introduced Opuntia species found in Florida A significant cactus species by cage type interaction effect was observed on species ranking (Table 3; F12.09, df5, 56.7, PB0.0001). Opuntia corallicola and O. stricta mean rankings were significantly higher ranked in the large cage, while O. humifusa, O. pusilla, and O. tricantha were significantly higher ranked in the small Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009 cage. Opuntia corallicola and O. humifusa had significantly higher rankings than all other cactus species tested, except O. pusilla. Cactus by cage type interaction significantly affected the proportions of eggsticks oviposited on cactus species (Table 3; F29.3, df5, 60, PB0.0001). Mean proportions of eggsticks oviposited on O. corallicola were significantly higher in large cages than in small cages; conversely, O. humifusa, O. pusilla, and O. tricantha were significantly higher in small cages. Significantly higher proportions of eggsticks were oviposited on O. corallicola than any other cactus species (Table 3; F18.2, df5, 60, PB0.0001). A significant cactus and cage type interaction effect was observed on mean numbers of eggs per eggstick (Table 3; F2.9, df5, 49.4, P0.02). No cactus species effect (F1.74, df5, 49.2, P0.14) was observed on mean numbers of eggs per eggstick: however cage type (F3.8, df1, 51.9, P0.05) had a significant effect on mean numbers of eggs per eggstick. Biocontrol Science and Technology 323

Table 2. Mean (9SE) values for oviposition parameters of the two main treatments; female Cactoblastis cactorum mated to normal males, and females mated to irradiated males.

Ranking Proportion of Number of eggsticks eggs/eggstick

Ovipostional preference Normal Irradiated Normal Irradiated Normal Irradiated comparison

Six common Opuntia 10.991.9 11.991.9 0.5190.13 0.4990.13 25.292.1 24.892.1 spp. in FL Opuntia spp. with 8.790.4 9.090.4 0.5090.19 0.5090.19 29.391.7 27.691.7 similar growth characteristics Opuntia spp. with 9.990.8 10.390.8 0.5090.04 0.5090.04 15.893.2 17.093.1 different spine & glochid compliments Species of Opuntia vs. 9.590.8 10.390.8 0.5090.05 0.5090.5 23.192.5 20.292.4 Cylindropuntia vs. Harrisia Documented hosts vs. 9.990.7 9.490.7 0.4990.04 0.5190.04 22.392.2 20.392.4 undocumented hosts Cactus species in FL 8.691.8 8.591.8 0.5090.04 0.5090.04 19.894.2 16.594.7 vs. cactus species in Southwest Economically 9.890.9 11.190.9 0.5490.04 0.4690.04 20.892.2 15.894.1 important Opuntia spp.

No significant differences were found between the two treatments for any of the three parameters at a P0.05.

Table 3. Mean (9SE) Cactoblastis cactorum oviposition preference ranking, proportion of eggsticks per plant, and numbers of eggs per eggstick on five native and one introduced Opuntia species found in Florida.

Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009 Cactus species

O. corallicola O. ficus-indica O. humifusa O. pusilla O. stricta O. tricantha Cage (Native) (Introduced) (Native) (Native) (Native) (Native)

Rankings Small 10.892.6B 10.892.6B 2.392.6A 5.792.6AB 21.392.6C 8.292.6B Large 2.692.4A* 13.392.4B 15.592.4B* 15.192.4B* 15.992.4B* 15.192.4B* Proportion of eggsticks Small 0.1190.04C 0.1390.04B 0.2990.04A 0.269.04AB 0.0190.04D 0.2090.04A Large 0.5990.03A* 0.1290.03B 0.0790.03B* 0.079.03B* 0.0790.03B 0.0890.03B* Eggs per eggstick Small 18.494.2B 26.394.6AB 27.794.2A 24.6 94.2AB 24.099.8AB 17.994.6B Large 34.093.6A* 29.093.6AB 26.494.1B 25.093.9B 27.793.6B 16.893.6C

Uppercase letters denote significant differences between means within a row and an asterisk (*) denotes significant differences between cage means within a column for each variable. 324 C.D. Tate et al.

Cactoblastis cactorum oviposition preference for cactus species in six select groupings Group 1: comparison of Opuntia cacti with similar growth characteristics Cactus species ranking was significantly affected by interaction between cactus species and cage type (Table 4; F18.1, df3, 48, PB0.0001). Mean rank of O. corallicola and O. falcata (Ekman & Werdermann) were significantly higher in large cages than in small cages; while O. stricta ranked higher in small cages than in large cages. Opuntia corallicola (the endangered south Florida species) in large cages was ranked significantly higher than all other tall cactus species evaluated. Cactus by cage type interactions (Table 4; F19.6, df3, 48, PB0.0001) significantly affected proportions of eggsticks oviposited on tall cactus species. Significantly higher proportions of eggsticks were oviposited on O. corallicola in large cages than in small cages; while significantly more eggsticks were oviposited on O. stricta in small cages than large cages. Mean proportions of eggsticks oviposited on the endangered species (O. corallicola) were significantly higher than all tall cactus species evaluated (Table 4; F55.5, df5, 60, PB0.0001). A significant interaction between cage type and cactus species was observed on numbers of eggs per eggstick on cactus species with similar growth characteristics (Table 4; F33.4, df3, 47, PB0.0001). Mean numbers of eggs per eggstick on O. cochenillifera and O. stricta were significantly higher in small cages than large cages. Significantly fewer eggs per eggstick were oviposited on O. cochenillifera and O. stricta plants in large cages than all other cactus species in any size cage; however, significantly more eggs per eggstick were oviposited on O. cochenillifera than on O. stricta plants in large cages.

Group 2: comparison of Opuntia cacti with differing spine and glochid complements Cactus species ranking was significantly affected by interaction between cactus species and cage type for this group of Opuntia spp. (Table 5; F4.1, df3, 48,

Table 4. Group 1: comparison of four Opuntia species with similar growth characteristics for mean (9SE) Cactoblastis cactorum oviposition preference ranking, proportion of eggsticks per plant, and numbers of eggs per eggstick. Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009 Cactus species

Cage O. cochenillifera O. corallicola O. falcata O. stricta

Rankings Small 12.590.8C 6.990.8A 7.090.8A 9.390.8B Large 13.390.8C 1.590.8A* 4.890.8B* 15.490.8D* Proportion of eggsticks Small 0.1090.04B 0.3590.04A 0.3090.04A 0.2590.04A Large 0.0490.04C 0.5190.04A* 0.2790.04B 0.1390.04C* Eggs per eggstick Small 39.293.2A 29.093.0B 38.393.0A 32.193.0B Large 13.393.0B* 32.693.0A 36.693.0A 6.493.0C*

Uppercase letters denote significant differences between means within a row and an asterisk (*) denotes significant differences between cage means within a column for each variable. Biocontrol Science and Technology 325

P0.01). Mean rank of O. polyacantha and O. stricta rankings were significantly higher in small cages than in large cages. Opuntia falcata (lacking spines and glochids) in large cages was ranked significantly higher than all other test cactus species. Cactus by cage type interaction (Table 5; F7.0, df3, 48, P0.0006) significantly affected proportions of eggsticks oviposited on cactus species in the second test group. Mean proportions of eggsticks oviposited on cactus species evaluated differed among cage types, with the exception of O. microdasys. Significantly higher proportions of eggsticks were oviposited on O. falcata in large cages than in small cages. However, significantly higher proportions of eggsticks were oviposited on O. polyacantha and O. stricta in small cages than large cages. Mean proportions of eggsticks oviposited on the non-spiny O. falcata was significantly higher than all cactus species evaluated (Table 5; F19.7, df3, 48, PB0.0001). A significant interaction effect between cage type and cactus species (Table 5; F 4.3, df3, 37.6, P0.01) was observed on numbers of eggs per eggstick. Cactus by cage type interaction (Table 5; F10.5, df3, 37.3, PB0.0001) significantly affected numbers of eggs per eggstick. Mean numbers of eggs per eggstick on O. polyacantha and O. stricta in small cages was significantly higher than in large cages. Significantly higher numbers of eggs per eggstick were observed on non-spiny O. falcata in large cages. The two species with the most glochids/spines (O. microdasys and O. polyacantha) had significantly lower numbers of eggs per eggstick in small cages. Significantly more eggs per eggstick were collected from O. microdasys in large cages than from O. polyacantha, and O. stricta in large cages.

Group 3: comparison of Opuntia, Cylindropuntia, and Harrisia cacti Cactus species ranking was significantly affected by cage type (Table 6; F13.4, df1, 48, P0.0006) and cactus species (Table 6; F3.39, df3, 48, P0.03).

Table 5. Group 2: comparison of four Opuntia species with differing spine and glochid complements for mean (9SE) Cactoblastis cactorum oviposition preference ranking, proportion of eggsticks per plant, and numbers of eggs per eggstick. Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009 Cactus species

Cage O. falcata O. microdasys O. polyacantha O. stricta

Rankings Small 6.391.5A 11.991.5C 9.691.5BC 8.491.5AB Large 3.991.5A 11.091.5B 15.191.5C* 14.891.5BC* Proportion of eggsticks Small 0.4090.07A 0.1690.07B 0.1890.07B 0.2690.07AB Large 0.7890.07A* 0.1690.07B 0.0390.07B* 0.0390.07B* Eggs per eggstick Small 24.394.2A 12.796.4B 21.394.6AB 26.694.4A Large 30.494.2A 11.494.2B 1.594.2C* 2.594.2C*

Uppercase letters denote significant differences between means within a row and an asterisk (*) denotes significant differences between cage means within a column for each variable. 326 C.D. Tate et al.

Cactus species rankings, based upon numbers of eggs oviposited, were significantly higher in small cages compared to large cages, except for C. spinosior whose ranking did not differ between the small and large cage. In addition, H. fragans ranked significantly lower overall than all other test species. Proportions of eggsticks oviposited was not significantly affected by cage type (Table 6; F9.9, df1, 45, P0.19). Harrisia fragans tended to receive the lowest proportion of eggsticks in both small and large cages, although not always significantly fewer than all other test species. Similar to mean proportions of eggsticks, no significant interaction and no treatment or cactus species main effects were observed on mean numbers of eggs per eggstick oviposited on these cactus species. However, numbers of eggs per eggstick oviposited was significantly affected by cage type (Table 6; F82.8, df1, 43, PB0.0001). Significantly more eggs per eggstick were collected in small cages than large cages with all four species.

Group 4: comparison of C. cactorum hosts and non-hosts Cactus species ranking was significantly affected by interaction between cactus species and cage type (Table 7; F4.6, df3, 48, P0.006). Ranking was significantly higher in small cages than in large cages for C. spinosior and O. stricta. Mean ranking for H. fragrans and O. steptacantha were not significantly different in large and small cages. The non-documented host plant species C. spinosior in small cages ranked significantly higher than all other cactus species evaluated, while the other non-documented host plant H. fragrans in large cages ranked significantly lower than other species. Cactus by cage type interaction (Table 7; F8.4, df3, 45, P0.0006) significantly affected the proportions of eggsticks oviposited on cactus species. Significantly higher proportions of eggsticks were oviposited on O. spinosior in

Table 6. Group 3: comparison of Opuntia, Cylindropuntia, and Harrisia cacti for mean (9 SE) Cactoblastis cactorum oviposition preference ranking, proportion of eggsticks per plant, and numbers of eggs per eggstick. Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009 Cactus species

Cage C. acanthocarpa H. fragrans C. spinosior O. stricta

Rankings Small 7.491.7AB 10.591.7B 8.191.7AB 4.991.7A Large 12.191.7B* 15.591.7B* 8.091.7A 12.691.7B* Proportion of eggsticks Small 0.3390.09A 0.1790.09A 0.2490.09A 0.2790.09A Large 0.3290.09AB 0.1790.09B 0.2490.09A 0.2790.09A Eggs per eggstick Small 28.695.0A 28.795.4A 34.295.4A 36.094.7A Large 6.094.7BC* 1.994.7C* 23.094.7A* 14.694.7AB*

Uppercase letters denote significant differences between means within a row and an asterisk (*) denotes significant differences between cage means within a column for each variable. Biocontrol Science and Technology 327

Table 7. Group 4: comparison of Cactoblastis cactorum hosts and non-hosts for mean (9SE) Cactoblastis cactorum oviposition preference ranking, proportion of eggsticks per plant, and numbers of eggs per eggstick.

Cactus species

Cage H. fragrans C. spinosior O. streptacantha O. stricta

Rankings Small 13.591.4C 1.891.4A 11.091.4C 6.691.4B Large 13.491.4B 10.491.4A* 10.591.4A 10.491.4A* Proportion of eggsticks Small 0.0390.07C 0.6490.07A 0.1490.07BC 0.2090.07B Large 0.1090.07B 0.3490.07A* 0.2890.07A 0.2890.07A Eggs per eggstick Small 45.696.9A 25.594.0B 21.495.1B 30.094.0B Large 7.294.0B* 15.494.0A* 14.194.0AB 11.194.0AB*

Uppercase letters denote significant differences between means within a row and an asterisk (*) denotes significant differences between cage means within a column for each variable.

small cages than in large cages. No significant difference in mean proportions of eggsticks on O. stricta, O. streptacantha and H. fragrans in large and small cages was observed. Mean proportions of eggsticks oviposited on a non-documented host (C. spinosior) in small cages was significantly higher than all cactus species evaluated; while mean proportions of eggsticks on the non-documented host H. fragrans was significantly lower than other species in both small and large cages, except for O. stricta in small cages. No significant interaction was found between cactus and cage type for mean number of eggs per eggstick. However, mean number of eggs per eggstick was significantly affected by cage type (Table 7; F60.0, df1, 40, PB0.0001). Significantly more eggs per eggstick were collected from cladodes in small cages than from plants in large cages. The highest mean number of eggs per eggstick was

Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009 found on H. fragrans in small cages.

Group 5: comparison of cactus species, H. fragrans and O. dellinii, represented in Florida and species, C. spinosior and O. streptacantha, represented in the Southwestern US and Mexico Cactus species ranking was significantly affected by cage type (Table 8; F7.0, df1, 38.3, P0.01) and cactus species (Table 8; F9.9, df3, 36.6, PB0.0001). Cactus species rankings were significantly higher in small cages compared to large cages. Mean ranking of O. streptacantha was significantly higher than other test species, with the exception of C. spinosior. Harrisia fragrans ranked significantly lower than all cactus species evaluated. Proportions of eggsticks oviposited were significantly affected by cactus species (Table 8; F9.9, df3, 36.2, PB0.0001). Opuntia streptacantha had significantly 328 C.D. Tate et al.

Table 8. Group 5: comparison of species, Harrisia fragrans and Opuntia dellinii, represented in Florida and species, Cylindropuntia spinosior and Opuntia streptacantha, represented in the southwestern US and Mexico for mean (9SE) Cactoblastis cactorum oviposition preference ranking, proportion of eggsticks per plant, and numbers of eggs per eggstick.

Cactus species

Cage H. fragrans C. spinosior O. streptacantha O. dellinii

Rankings Small 12.692.4C 6.292.4AB 1.992.4A 7.192.4B Large 13.992.2B 9.092.2A 7.192.2A* 10.692.2AB Proportion of eggsticks Small 0.0390.09C 0.3490.09AB 0.4190.09A 0.2290.09B Large 0.0290.09C 0.3390.09AB 0.4990.09A 0.1690.09BC Eggs per eggstick Small 18.597.6B 28.895.2AB 28.494.8AB 33.495.5A Large 2.894.3B* 11.194.3AB* 19.294.3A 6.094.3B*

Uppercase letters denote significant differences between means within a row and an asterisk (*) denotes significant differences between cage means within a column for each variable.

higher proportions of eggsticks than all test species. Significantly lower proportions of eggsticks were oviposited on H. fragrans than remaining test species. Numbers of eggs per eggstick was significantly affected by cage type (Table 8; F86.1, df1, 31.8, PB0.0001) and cactus species (Table 8; F5.4, df3, 30.9, P0.004). Significantly more eggs per eggstick were collected from small cages than large cages. Significantly higher numbers of eggs per eggstick were collected from O. streptacantha in small and large cages than from H. fragrans. Numbers of eggs per eggstick collected from O. dellinii in large cages also was significantly higher than from H. fragrans.

Group 6: comparison of economically important Opuntia species No significant cactus species main effect was observed on mean ranking (Table 9; Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009 F1.6, df3, 48, P0.19). Cactus species ranking tended to be higher for O. engelmannii in small cages but lower in large cages than the other test species. Cactus species ranking was significantly affected by cage type (Table 9; F6.6, df 1, 48, P0.01); oviposition ranking for O. engelmannii was higher in small cages than large cages. No significant cactus species main effects were observed on mean proportions of eggsticks oviposited on these cactus species (Table 9; F2.4, df3, 48, P0.08). A significant interaction between species and cage type was observed (Table 9; F5.9, df3, 48, P0.01). The proportion of eggsticks oviposited in small cages was significantly higher for O. engelmannii, but significantly lower for O. stricta In addition, although O. engelmannii received the highest mean proportion of eggsticks in small cages, this same species received significantly fewer eggsticks than O. streptacantha or O. stricta in large cages. Numbers of eggs per eggstick collected was significantly affected by cage type (Table 9; F185.4, df1, 36, PB0.0001) and cactus species (Table 9; F3.5, df3, Biocontrol Science and Technology 329

Table 9. Group 6: comparison of economically important Opuntia species for mean (9SE) Cactoblastis cactorum oviposition preference ranking, proportion of eggsticks per plant, and numbers of eggs per eggstick.

Cactus species

Cage O. engelmannii O. ficus-indica O. streptacantha O. stricta

Rankings Small 5.591.8C 11.891.8A 7.591.8BC 10.591.8AB Large 14.891.8A* 13.491.8AB 9.891.8B 10.491.8B Proportion of eggsticks Small 0.5390.08A 0.1090.08B 0.2390.08B 0.1490.08B Large 0.1490.08B* 0.1990.08AB 0.3490.08A 0.3390.08A* Eggs per eggstick Small 18.493.7B 28.894.3A 33.394.1A 28.894.7A Large 2.193.4B* 5.793.4AB* 10.493.4A* 9.093.4A*

Uppercase letters denote significant differences between means within a row and an asterisk (*) denotes significant differences between cage means within a column for each variable.

35.3, P0.03). Significantly more eggs per eggstick were collected from cladodes of all four species in small cages than from plants in large cages. Mean number of eggs per eggstick collected from O. ficus-indica, O. streptacantha, and O. stricta was significantly higher than from O. engelmannii.

Discussion Open field tests have been used to clarify host specificity test results obtained from more traditional cage tests conducted in quarantine facilities. Many weed biological control researchers consider open field tests as the most realistic method to evaluate host specificity for potential biological control agents, especially for ovipositing females that are no longer behaviorally constrained by caging (Sheppard 1999; Briese et al. 2002; Heard, Zonneveld, Segura, and Martinez 2004). However, to date, open field tests are conducted in the native range of the potential biological control agent Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009 and not in the native range of the non-target host plants of concern. Potential host plants may be shipped to the native range of the control agent but often these non- native plants cannot be planted outside a containment facility because of concerns and prohibitions in introducing non-native species. Even if the non-native test plant can be planted in an open field setting, additional problems may arise such as the time required producing a healthy, mature test plant, presence of specific micro- habitat growth conditions, or attack/destruction by generalist natural enemies not present in the test plants’ native range. The most realistic test for non-target impacts from a potential biological control agent would be conducted in the native area of the non-target host plants. A safe alternative for assessing the host range of potential Lepidoptera biological control agents is open field-testing in the requested area of introduction with the F1 sterility technique (Moeri et al. 2009). Carpenter et al. (2001b) discussed using F1 sterility as a technique in elucidating potential host range of C. cactorum for key Opuntia species across the southern US. As an oligophagous feeder, C. cactorum feeds on a wide range of Opuntia spp. within 330 C.D. Tate et al.

the subgenus Platyopuntia in the insects’ native region of South America (Mann 1969; Zimmermann, McFadyen, and Erb 1979) and the insects’ adventive range in the southeastern US (Hight et al. 2002). Host specificity for C. cactorum females mated to either irradiated or normal males was evaluated in this study as a proof of concept for using the F1 sterility technique. Several arrangements of host plant choice tests were conducted to insure a rigorous comparison of the two oviposition preference treatments. In all tests, female C. cactorum oviposition preference was not significantly different for females mated to irradiated males versus females mated to normal males. This study supports the use of F1 sterility as a tool for assessing the safety of exotic Lepidoptera for biological control of invasive weeds. F1 sterile moths could be released in the proposed area of introduction as part of a program to evaluate the risk of non-target impacts without the risk of the proposed biological control agent becoming established. Although female C. cactorum oviposition preference was not significantly different when mated with irradiated or normal males, there were significantly different oviposition preferences in the choice tests conducted in this study. Cactus species significantly affected female C. cactorum oviposition preference. Of the six Opuntia spp. from Florida, significantly more eggsticks were oviposited on O. corallicola. This species was also found to be the preferred oviposition host by Johnson and Stiling (1996) in caged tests with three other Florida native Opuntia spp., even though the larval performance was significantly poorer. Opuntia corallicola is spinier than other test cacti of similar height (Group 1, Table 2) and, for most comparisons, was the preferred host. However, in the evaluation of species with and without spines (Group 2, Table 3), the plant without spines (O. falcata) received significantly more eggsticks than the species with numerous spines, although the test did not include the species O. corallicola. This inconsistency is further born out by Mafokoane, Zimmermann, and Hill (2007) who found that C. cactorum avoided ovipositing on plants with dense spines, but preferred O. ficus-indica,an upright, large stemmed plant, with few spines. Evaluating data only from our study, a potential pattern of oviposition preference may be found in that tall species (O. corallicola, O. falcata and O. ficus-indica) are preferred over short species, spiny (O. polyacantha) or less spiny (O. pusilla), but tall spiny species (O. corallicola) are preferred over tall not spiny species (O. cochenillifera). Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009 Our oviposition test of the three different genera (Group 3) identified variations in oviposition preference. Certain taxonomists regard Cylindropuntia as sub-genera of Opuntia (Benson 1982) and others regard them as independent genera (Anderson 2001). Either way, the species are closely related and, in general, numbers of eggsticks oviposited on species of Opuntia and Cylindropuntia cacti were similar. Oviposition on closely related non-Opuntia cacti in a field situation will likely occur. In fact, in our studies evaluating non-hosts and known hosts of C. cactorum (Group 4, Table 6), females oviposited significantly more eggsticks on C. spinosior cladodes in small cages than on the other four cactus species including O. stricta, a known host plant. Oviposition on a species in the cactus genera Harrisia received significantly fewer eggsticks. The genus Harrisia is in the subfamily Cactoideae while the genera Opuntia and Cylindropuntia are in the subfamily Opuntioideae. Not only cactus species significantly affected C. cactorum oviposition preference, but so did cage type. However, the trend was not consistent. Some species (such as O. corallicola) showed a relatively consistent increase in preference measures from Biocontrol Science and Technology 331

small to large cage. Other species (O. spinosior) showed a decrease in preference measures from small to large cage. We did not evaluate the physiological changes in cladodes excised from plants and placed in the small cages. Cladodes from some cactus species may have changed more than others when they were excised. Cage size may also have impacted preference by interfering with female visual cues in concert with plant volatiles to identify potential hosts. Removal of the cladode from the plant also could have caused biochemical changes in excised cladodes. However, in addressing the primary objective of this study, we identified no difference in oviposition preference for females mated to irradiated or normal males. These data support the use of F1 inherited sterility as a tool for assessing the safety of exotic lepidopterans for biological control of invasive weeds. Irradiated males and non-irradiated females could be released in areas without the risk of the biological control agent becoming established. This technique will allow open field studies that can be used to evaluate host specificity under more stringent and realistic testing procedures to obtain permission for importation and release of exotic biological control agents.

Acknowledgements We thank Susan Drawdy and Robert Caldwell (USDA-ARS, Tifton, GA) for technical assistance, and Denny Bruck (USDA-ARS, Corvallis, OR) and David Coyle (University Of Wisconsin, Madison) for comments on earlier drafts of this manuscript. Mention of trade names or commercial products in this publication is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the US Department of Agriculture.

References Anderson, E.F. (2001), The Cactus Family, Portland, OR: Timber Press. Benson, L. (1982), The Cacti of the United States and Canada, Stanford, CA: Stanford University Press. Briese, D.T. (2003), ‘The Centrifugal Phylogenetic Method Used to Select Plants for Host- specificity Testing of Weed Biological Control Agents: Can and Should it be Modernized?’, in Improving the Selection, Testing and Evaluation of Weed Biological Control Agents, eds. H. Spafford Jacob and D.T. Briese, Glen Osmond, Australia: CRC for Australian Weed Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009 Management, pp. 23Á33. Briese, D.T., and Walker, A. (2002), ‘A New Perspective on the Selection of Test Plants for Evaluating the Host-specificity of Weed Biological Control Agents: The Case of Deuterocampta quadrijuga, a Potential Insect Control Agent of Heliotropium amplexicaule’, Biological Control, 25, 273Á287. Briese, D.T., Zapater, M., Andorno, A., and Perez-Camargo, G. (2002), ‘A Two-phase Open- field Test to Evaluate the Host-specificity of Candidate Biological Control Agents for Heliotropium amplexicaule’, Biological Control, 25, 259Á272. Carpenter, J.E., Bloem, S., and Bloem, K.A. (2001a), ‘Inherited Sterility in Cactoblastis cactorum (Lepidoptera: Pyralidae)’, Florida Entomologist, 84, 537Á542. Carpenter, J.E., Bloem, S., and Bloem, K.A. (2001b), ‘Applications of F1 Sterility for Research and Management of Cactoblastis cactorum (Lepidoptera: Pyralidae)’, Florida Entomologist, 84, 531Á536. Dodd, A.P. (1940), The Biological Campaign Against Prickly Pear, Brisbane, Queensland: Commonwealth Prickly Pear Board. Greany, P.D., and Carpenter, J.A. (2000), ‘Use of Nuclear Techniques in Biological Control’, in Proceedings: Area-Wide Control of Fruit Flies and Other Insect Pests. International Conference on Area-Wide Control of Insect Pests, and the 5th International Symposium on 332 C.D. Tate et al.

Fruit Flies of Economic Importance, Penang, Malaysia: Penerbit Universiti Sains Malaysia, pp. 221Á227. Heard, T.A., Zonneveld, R., Segura, R., and Martinez, M. (2004), ‘Limited Success of Open Field Tests to Clarify the Host Range of Three Species of Lepidoptera of Mimosa pigra’,in Proceedings of the XI International Symposium on Biological Control of Weeds, eds. J.M. Cullen, D.T. Briese, D.J. Kriticos, W.M. Lonsdale, L. Morin, and J.K. Scott, Canberra, Australia: CSIRO Entomology, pp. 277Á282. Hight, S.D. and Carpenter, J.E. (2009), ‘Flight Phenology of Male Cactoblastis cactorum (Lepidoptera: Pyralidae) at Different Latitudes in the Southeastern United States, Florida Entomologist (forthcoming). Hight, S.D., Carpenter, J.E., Bloem, K.A., Bloem, S., Pemberton, R.W., and Stiling, P. (2002), ‘Expanding Geographical Range of Cactoblastis cactorum (Lepidoptera: Pyralidae) in North America’, Florida Entomologist, 85, 527Á529. Hight, S.D., Bloem, S., Bloem, K.A., and Carpenter, J.E. (2003), ‘Cactoblastis cactorum (Lepidoptera: Pyralidae): Observations of Courtship and Mating Behaviors at Two Locations on the Gulf Coast of Florida’, Florida Entomologist, 86, 400Á407. Johnson, D.M., and Stiling, P.D. (1996), ‘Host Specificity of Cactoblastis cactorum (Lepidoptera: Pyralidae), an Exotic Opuntia-feeding Moth, in Florida’, Environmental Entomologist, 28, 743Á748. Julien, M. and Griffith, M.W. (eds) (1998), Biological Control of Weeds. A World Catalogue of Agents and their Target Weeds, Wallingford: CABI Publishing. LaChance, L.E. (1985), ‘Genetic Methods for the Control of Lepidopteran Species: Status and Potential’, U.S. Department of Agriculture, Research Series, ARS-28. Mafokoane, L.D., Zimmermann, H.G., and Hill, M.P. (2007), ‘Development of Cactoblastis cactorum (Berg) (Lepidoptera: Pyralidae) on Six North American Opuntia Species’, African Entomology, 15, 295Á299. Mann, J. (1969), ‘Cactus Feeding Insects and Mites’, U.S. National Museum Bulletin, 256, 1Á158. Marohasy, J. (1998), ‘The Design and Interpretation of Host-specificity Tests for Weed Biological Control with Particular Reference to Insect Behaviour’, Biocontrol News and Information, 19, 13NÁ20N. McEvoy, P.B. (1996), ‘Host Specificity and Biological Pest Control’, BioScience, 46, 401Á405. Moeri, O.E. Cuda, J.P. Overholt, W.A. Bloem, S. and Carpenter, J.E. (2009), ‘F1 Sterile Insect Technique: A Novel Approach for Risk Assessment of Episimus unguiculus (Lepidoptera: Tortricidae), a Candidate Biological Control Agent of Schinus terebinthifolius in the Continental USA, Biocontrol Science Technology (in press this volume). Moran, V.C., and Zimmermann, H.G. (1984), ‘The Biological Control of Cactus Weeds: Achievements and Prospects’, Biocontrol News and Information, 5, 297Á320.

Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009 Pemberton, R.W. (1995), ‘Cactoblastis cactorum (Lepidoptera: Pyralidae) in the United States: An Immigrant Biological Control Agent or an Introduction of the Nursery Industry?’, American Entomologist, 41, 230Á232. Pemberton, R.W., and Cordo, H. (2001), ‘Potential and Risk Of Biological Control of Cactoblastis cactorum (Lepidoptera: Pyralidae) in North America’, Florida Entomologist, 84, 513Á526. Perez-Sandi, C.M. (2001), ‘Addressing the Threat of Cactoblastis cactorum (Lepidoptera: Pyralidae) to Opuntia in Mexico’, Florida Entomologist, 84, 499Á502. Petty, F.W. (1948), ‘The Biological Control of Prickly Pears in South Africa’, Science Bulletin, Department of Agriculture of the Union of South Africa, 271, 1Á163. SAS Institute Inc. (1999), SAS systems for Windows, version 8.0, Cary, NC: Author. Simmonds, F.J., and Bennett, F.D. (1966), ‘Biological Control of Opuntia spp. by Cactoblastis cactorum in the Leeward Islands (West Indies)’, Entomophaga, 11, 183Á189. Sheppard, A.W. (1999), ‘Which Test? A Mini Review of Test Usage in Host Specificity Testing’,inHost Specificity Testing in Australasia: Towards Improved Assays for Biological Control, eds. T.M. Withers, L. Barton Browne, J. Stanley Brisbane, Australia: CRC for Tropical Pest Management, pp. 60Á69. Biocontrol Science and Technology 333

Sobero´n, J., Golubov, J., and Sarukhan, J. (2001), ‘The Importance of Opuntia in Mexico and Routes of Invasion and Impact of Cactoblastis cactorum (Lepidoptera: Pyralidae)’, Florida Entomologist, 84, 486Á492. Stiling, P. (2002), ‘Potential Non-target Effects of a Biological Control Agent, Prickly Pear Moth, Cactoblastis cactorum (Berg) (Lepidoptera: Pyralidae), in North America, and Possible Management Options’, Biological Invasions, 4, 273Á281. Stiling, P., and Moon, V.C. (2001), ‘Protecting rare Florida Cacti from Attack by the Exotic Cactus Moth, Cactoblastis cactorum (Lepidoptera: Pyralidae)’, Florida Entomologist, 84, 507Á509. Strong, D.R., and Pemberton, R.W. (2000), ‘Biological Control of Invasive Species Á Risk and Reform’, Science, 288, 1969Á1970. Tate, C.D., Carpenter, J.E., and Bloem, S. (2007), ‘Influence of Irradiation Dose on the Level of F1 Sterility in the Cactus Moth, Cactoblastis cactorum (Lepidoptera: Pyralidae)’, Florida Entomologist, 90, 537Á544. Wapshere, A.J. (1974), ‘A Strategy for Evaluating the Safety of Organisms for Biological Weed Control’, Annals of Applied Biology, 77, 201Á211. Zimmermann, H.G., McFadyen, R.E, and Erb, H.E. (1979), ‘Annotated List of Some Cactus- feeding Insects of South America’, Acta Zoologica Lilloana, 32, 101Á112. Zimmermann, H.G., Moran, V.C., and Hoffmann, J.H. (2001), ‘The Renowned Cactus Moth, Cactoblastis cactorum: Its Natural History and Threat to Native Opuntia in Mexico and the United States of America’, Florida Entomologist, 84, 543Á551. Downloaded By: [Hendrichs, Jorge] At: 16:01 6 November 2009 Biocontrol Science and Technology, Vol. 19, S1, 2009, 335362

Radiation sources supporting the use of natural enemies for biological control of agricultural pests Kishor Mehta

Vienna, Austria

Augmentative biological control as a component of integrated pest management programmes involves the release of natural enemies of the pest, such as parasitoids and predators. Several potential uses for nuclear techniques have been identified which can benefit such programmes; these benefits include facilitating trade, protecting the environment and increasing the overall efficacy of the programmes. This may involve sterilising feed material, hosts or even the control insects. Radiation is currently the most favoured sterilising agent, although availability and cost of radiation sources are considered as limiting the use of radiation in support of biological control. This paper reviews various radiation sources that may be used for this purpose, including a comparison of several key parameters such as cost estimates of these radiation sources that should assist in making a judicious selection of a suitable irradiator. Keywords: natural enemies; pest management; irradiators; radiation; sterilisation

Introduction Ionising radiation Ever since ionising radiation was discovered by Marie Curie more than 100 years ago, it has been beneficially used in nearly every activity that touches human life, from medicine to communication to hydrology to food and agriculture. Electromagnetic radiation spectrum includes radiowaves with very long wave- length to the exceedingly penetrating X-rays and gamma rays with very short 8 1 Downloaded By: [Hendrichs, Jorge] At: 16:02 6 November 2009 wavelength, all of which travel at the velocity of light (Â310 ms in vacuum). The energy associated with radiation is inversely proportional to the wavelength; the longer the wavelength, the less is the energy. The energy associated with X-rays and gamma rays is greater than the binding energy of an atomic electron, thus this type of radiation can ionise an atom or break the molecular bonds. Radiation with such high energy is referred to as ionising radiation. Besides gamma rays and X-rays, ionising radiation includes high-energy electrons (generally80 keV). Ionising radiation breaks down molecules, modifying chemical, physical or biological properties of the irradiated material. Thus, radiation can cause polymerisation of plastics, kill pathogens and microorganisms, and damage DNA molecules, leading to applications in industry and food processing, sterilisation of health care products, and reproductive sterilisation of insects.

*Email: [email protected]

First Published Online 14 October 2008 ISSN 0958-3157 print/ISSN 1360-0478 online # 2009 Taylor & Francis DOI: 10.1080/09583150802417849 http://www.informaworld.com 336 K. Mehta

Applications of ionising radiation in entomology There are a number of applications of ionising radiation in entomology (Bakri, Heather, Hendrichs, and Ferris 2005a), including disinfestation of commodities for quarantine and phytosanitary purposes, and reproductive sterilisation of insects for pest management programmes using the Sterile Insect Technique (SIT) (Dyck, Hendrichs, and Robinson 2005). Radiation can also be applied in various ways to facilitate the use of biological agents for control of arthropod pests and weeds (Carpenter 1997, 2000; Greany and Carpenter 2000). These authors cite a number of potential advantages of nuclear techniques for biological control: avoidance of the emergence of pest insects from non-parasitised hosts, allowing earlier transport and facilitating trans-boundary shipment, improvements in rearing media (either artificial diets or natural hosts/prey), provision of sterilised natural prey to be used as food during predator shipment, to ameliorate concerns relating to the incidental presence of hitch- hiking pests, provision of supplemental food or hosts in the field, to increase the initial survival and build-up of released natural enemies, and reproductive sterilisation of weed-feeding insects that are candidates for biological control allowing their risk-free field assessment of host specificity.

Currently, there are hundreds of producers of natural enemies reared specifically for biological control purposes that produce several types of parasitoids and predators (BCPC 2004). There are no complete data available regarding such information, although sales worldwide are generally estimated at about US$ 100 million (www.anbp.org; www.ibma.ch; www.amrqc.org). In North America alone, there are 2530 major producers, not including small local producers or collectors of garden products such as ladybugs, etc. Very few producers or the pest management programmes using biological control agents are currently using nuclear techniques; either there is no appreciation for such techniques and the subsequent benefit to the pest management programme, or they are not aware of a suitable radiation source that is also economical to use, or some organic growers using biological control agents may object to the use of radiation. This review Downloaded By: [Hendrichs, Jorge] At: 16:02 6 November 2009 provides relevant and useful information that may motivate these producers to apply this beneficial technology. There are three types of ionising radiation used in radiation processing, namely gamma rays, X-rays and electrons. All have similar effects on the irradiated materials (since they have similar relative biological effectiveness), and in particular on the irradiated insects (Bakri, Mehta, and Lance 2005b). In a biological organism composed of differentiated and undifferentiated cells, mitotically active cells such as stem cells and germ cells are the most radiation-sensitive elements. Thus, radiation can make an insect reproductively sterile by damaging the DNA of gonial cells. For certain insect life stages, several studies found no significant difference in the lethal effects between electrons and gamma rays (Adem, Watters, Uribe-Rendo´n, and de la Piedad 1978; Watters 1979; Dohino, Tanabe, and Hayashi 1994). Numerous mutagenic chemicals were tested as alternatives to radiation for sterilisation of insects in the 1950s and 1960s (Knipling 1979). Efficacy of irradiated and chemosterilised insects for population control was generally similar (Guerra, Biocontrol Science and Technology 337

Wolfenbarger, Hendricks, Garcia, and Raulston 1972; Flint, Wright, Sallam, and Horn 1975; Moursy, Eesa, Cutkomp, and Subramanyam 1988). However, chemos- terilants are rarely used today. Most are carcinogenic, mutagenic and/or teratogenic, leading to environmental and human health issues in such areas as the integrity of ecological food chains, waste disposal (e.g. spent insect diet), and worker safety (Hayes 1968; Bracken and Dondale 1972; Bartlett and Staten 1996). Insect resistance to chemosterilants is an additional concern (Klassen and Matsumura 1966). Exposure to ionising radiation is now the principal method for inducing reproductive sterility in mass-reared insects. Irradiation of insects is a relatively straightforward process with reliable quality control procedures (FAO/IAEA/USDA 2003). The key parameter is the absorbed dose of radiation; efficacy of the irradiation process is guaranteed as long as the dose is correctly delivered (Bakri et al. 2005b). Other advantages of using radiation (gamma rays, X-rays and electrons) include (1) insignificant increase in temperature during the process, (2) treated insects can be used immediately after processing, (3) irradiation does not add residues that could be harmful to human health or the environment, and (4) radiation can pass through packaging material, thus allowing the insects to be irradiated after packaging.

Radiation dose The absorbed dose, D, is radiation energy absorbed in unit mass of a material, and is mathematically expressed as the quotient of do by dm, where do is the mean energy imparted to matter of mass dm; thus, Ddo/dm (ICRU 1998). The unit is J/kg. The special name for the unit is gray (Gy); thus, 1 Gy1 J/kg. The unit of absorbed dose used earlier was rad (1 Gy 100 rad). Quite often, ‘absorbed dose’ is simply referred to as ‘dose’.

Radiation technology Energy transfer from radiation to matter Ionising radiation is classified under two categories: directly ionising radiation, and Downloaded By: [Hendrichs, Jorge] At: 16:02 6 November 2009 indirectly ionising radiation (Attix and Roesch 1968). Charged particles such as electrons belong to the first category since they can transfer energy directly and ionise the atoms of the irradiated matter (e.g. tissues, insects). On the other hand, uncharged particles like photons transfer energy indirectly by first transferring their energy to electrons, which in turn ionise the atoms. The nature of this energy transfer from radiation (photons and electrons) to the irradiated matter influences the distribution of dose (McLaughlin, Boyd, Chadwick, McDonald, and Miller 1989; IAEA 2002a). Dose distribution (or dose variation) in the irradiated matter is one of the important parameters for insect irradiation. When matter is irradiated with a photon beam, the intensity of the beam, and thus the dose imparted to the irradiated matter, decreases exponentially with depth as radiation penetrates into the matter. The rate of decrease in the intensity depends on the photon energy, composition and density of the irradiated material, and irradiation geometry. On the other hand, electrons can be roughly characterised by a common pathlength, traced out by most such particles of a given energy in a 338 K. Mehta

specific medium. This is generally referred to as ‘range’. The range depends on the electron energy, composition and density of the irradiated material, and the irradiation geometry. Dose initially increases with distance, but eventually decreases as electron energy is spent. At the end of their range, the electrons come to rest in the material having spent all their kinetic energy. Figure 1 shows the variation of dose with distance in water for 5 MeV electrons and three types of photons (cobalt-60 gamma rays, and 5 MeV and 160 keV X-rays). There is no definite ‘range’ for photons, unlike the case with electrons.

Radiation energy To maintain fitness of the irradiated insects and for the safety of the operating personnel, induction of radioactivity in the irradiated materials, such as canisters (reusable containers) and insects, must be avoided. This is achieved by restricting the energy of the radiation used for treating insects as follows (FAO/IAEA/WHO 1999; IAEA 2002b; Codex Alimentarius 2003): for photons, the energy should be less than 7.5 MeV, and for electrons, the energy should be less than 10 MeV.

Thus, gamma rays from cobalt-60 (photon energies are 1.173 and 1.332 MeV) and caesium-137 (0.662 MeV), electrons generated by accelerators with energy less than 10 MeV, and X-rays generated from electrons with energy below 7.5 MeV are acceptable for irradiation of insects. Gamma rays as well as X-rays are photons. However, by convention the photons created outside the atomic nucleus are referred to as X-rays, and those created inside the nucleus during radioactive decay as gamma rays.

180

160 5 MeV electrons Downloaded By: [Hendrichs, Jorge] At: 16:02 6 November 2009 140

120

100

80 5 MeV X rays Relative dose 60

40 Co-60 gamma rays

20 160 keV X rays

0 0 5 10 15 20 25 cm of water

Figure 1. Variation of dose in water with distance for a narrow beam of radiation, for 5 MeV electrons, cobalt-60 gamma rays, and X-rays of energy 160 keV and 5 MeV. Biocontrol Science and Technology 339

Gamma irradiators To expose insects to gamma radiation, they are treated in a gamma irradiator, which consists of an isotopic radiation source, a mechanism for transporting the insects through the radiation field, and an operating system to control the exposure of insects to radiation. More than 1000 different radioisotopes emit gamma radiation, but only two are used for radiation processing, namely cobalt-60 and caesium-137. Both have well determined half-lives, emit relatively high-energy gamma rays, and decay into stable isotopes. Both radioisotopes are produced in nuclear reactors. Source capsules are doubly sealed in stainless steel or zirconium alloy for maximum security, which ensures that the radioactive material cannot come into contact with the product to be irradiated (Brinston and Norton 1994). Because of the difference in the photon energies emitted per disintegration (0.662 MeV for caesium compared to total energy of 2.505 MeV for cobalt), caesium sources require about four times more activity than cobalt sources to provide the same processing capacity. Since production of caesium- 137 is quite elaborate and it is water-soluble, presently all large gamma irradiators employ cobalt-60 as a radiation source. Cobalt is water-insoluble with a high melting point, and thus well suited for use as an intense source of gamma radiation. Cobalt-60 is deliberately produced in a nuclear power reactor; its production starts with natural cobalt (metal) that is an element with 100% abundance of the stable isotope cobalt-59 (Brinston and Norton 1994). On absorption of a neutron, a cobalt-59 atom converts into a cobalt-60 atom. The radioisotope cobalt-60 decays into a stable nickel isotope by principally emitting one negative beta particle (of maximum energy 0.313 MeV) with a half-life of about 5.271 years (Unterweger, Hoppes, and Schima 1992). Nickel-60 thus produced is in an excited state, and it immediately emits two photons of energy 1.173 and 1.332 MeV in succession to reach its ground state (Lide 1990). These two gamma-ray photons are responsible for radiation processing in the cobalt-60 gamma irradiators. With decay of every cobalt- 60 atom, the strength or the activity level of the cobalt source is decreasing, such that the decrease amounts to 50% in about 5.271 years. The radiation source in a gamma irradiator typically consists of several pencils of the radioactive isotope cobalt-60. The only variation in the source output is the

Downloaded By: [Hendrichs, Jorge] At: 16:02 6 November 2009 known reduction in activity (strength) caused by the radioactive decay mentioned above, which can have a significant impact on the programme (financial as well as scheduling) if not taken into account. Because the activity of a cobalt source decreases by about 12% annually, the irradiator operator compensates for this loss of activity by incrementally increasing irradiation time to maintain the same radiation dose to the insects. Because the irradiation times eventually become impractically long, the source needs to be replenished at regular intervals, depending on the operational requirements. Source pencils are eventually removed from the irradiator at the end of their useful life. Generally they are returned to the supplier for re-use, recycling or disposal. In about 50 years, 99.9% of cobalt-60 will decay into non-radioactive nickel. Two types of gamma irradiators may be used for exposing insects to radiation: self-contained irradiators and panoramic irradiators. Thus, irradiators may be divided into two broad types: small-scale self-contained irradiators, and large-scale panoramic irradiators. 340 K. Mehta

Figure 2. Self-contained dry-storage gamma irradiator suitable for research and small-scale irradiations. This irradiator employs cobalt-60 as the radiation source. In preparation for irradiation, a canister is being placed in the irradiation chamber when it is in the loading (shielded) position. The control panel is visible at bottom right.

Small-scale self-contained irradiators Self-contained irradiators are specially designed for research and for applications that need small doses and relatively small throughputs, such as blood irradiation to help prevent transfusion-induced graft-versus-host disease (GVHD) and exposure of insects for pest management programmes. Most irradiation of insects is currently carried out in such irradiators. Generally, these are dry-storage irradiators; the radiation source is predominantly cobalt-60 (Figure 2). There are a few old caesium irradiators still being used. These irradiators house the source within a protective shield of lead or other material; thus, they can be placed very conveniently in an existing laboratory or a room without needing extra shielding (hence, self-contained). The advantages of such small irradiators are that they provide a high dose rate and

Downloaded By: [Hendrichs, Jorge] At: 16:02 6 November 2009 good dose uniformity within the irradiated sample, which is essential especially for radiation research. These characteristics are achieved by surrounding the sample with several radiation source pencils, such that it receives radiation from all directions. Such an irradiation arrangement places restriction on the sample size, limiting it to typically 15 L. However, this volume is quite adequate for research and small-throughput applications. Considering the amount of material to be treated and the required dose, such small irradiators would be most suitable for biological control companies. They are easy to install and operate, although their installation and use have to follow national licensing procedures. To irradiate, a canister of insects is placed in the irradiation chamber while it is in the loading (shielded) position, and the timer is set to deliver the pre-selected dose (see Figure 2). On the push of a button located on the control panel, the irradiation chamber (along with the insect canister) is auto- matically lowered to the irradiation position, and is returned to the unloading (shielded) position at the end of the pre-set irradiation time. Biocontrol Science and Technology 341

These self-contained irradiators are classified by the IAEA as Category I (dry storage) and Category III (wet storage). Applications and the procedures for use for these two categories of irradiators are described in a Practical Radiation Safety Manual published by the IAEA (1996a).

Large-scale panoramic irradiators For large-volume applications, panoramic irradiators are more suitable, where the radiation source consists of either several cobalt-60 rods (pencils) arranged in a plane, or a single rod. The source is moved into a large irradiation room for treatment of the product (e.g. insects), and when not in use it is returned to a separate storage room, which is shielded by either water (wet storage) or concrete (dry storage). Since isotopic sources emit gamma radiation in all directions, they may be surrounded on all sides by canisters with insects for irradiation to increase the energy utilisation efficiency; thus, several insect canisters are typically irradiated simulta- neously (unlike self-contained irradiators). Many panoramic irradiators are run in a continuous operation mode, wherein canisters are carried on a conveyor around a central source, which also improves dose uniformity in the product. The speed of the conveyor is selected to give the required dose to the product. The source is moved to the storage room only when the irradiator is not in use. An example of such an irradiator is the unit operated in Mexico by the joint USA/Mexico Moscamed programme, which was installed more than 25 years ago. Currently it contains about 30 kCi of cobalt and processes about 15 000 L of fruit fly pupae per week. Generally, this type of gamma irradiator is too large and too expensive for the purpose of insect irradiation only, especially if small volumes are involved (see discussion in Total treatment costs). An alternate method is batch operation, which is more suitable for relatively small throughputs. In this mode of operation, several insect canisters (a batch of canisters) are placed in the irradiation room around the source position while the source is in the storage container (or a separate room below the irradiation room). After the irradiation room is vacated and locked, the source is moved into the room for the length of time required to deliver the desired absorbed dose. To improve dose uniformity, each canister is rotated on its own axis during irradiation. After Downloaded By: [Hendrichs, Jorge] At: 16:02 6 November 2009 irradiation, the source is returned to the storage container, and the irradiated canisters are replaced with a new batch of canisters for the next irradiation. An example of a typical batch irradiator, which is commercially available, is shown in Figure 3. This irradiator features a ring of product turntables surrounding the cobalt-60 source. The radius of the ring is adjustable so that the dose rates, irradiation time and product batch volume can be varied to a large extent. The configuration can facilitate simultaneous irradiation of products of various densities because the products on the turntables do not shield each other at any time. The rotation of the turntables ensures good dose uniformity for all product densities. When the source is raised from the storage container for irradiation, the biological shield of thick concrete and the safety interlock system protect personnel. There are a few laboratories (insect facilities) where such batch irradiators are currently being used. In 1996, such an irradiator with 12 turntables was commissioned by Servicio Nacional de Sanidad Agraria (SENASA), Ministerio de Agricultura, Peru, for the purpose of insect sterilisation (see Figure 4). Each turntable takes a large cylinder 342 K. Mehta

Figure 3. A typical panoramic gamma batch irradiator. This unit, marketed by MDS Nordion, has four turntables (50 cm in diameter) that can each support product up to 320 kg. The number of turntables and their arrangement can be modified as per customer’s requirements. (Figure courtesy of MDS Nordion, Canada.)

that in turn contains 12 smaller cylinders of 10 cm diameter and 70 cm long. The present activity is about 17 kCi of cobalt-60, which yields an average dose rate of about 1.7 Gy/min at the insect location. There is a similar irradiator at the ARC Infruitec-Nietvoorbij Fruit, Vine & Wine Research Institute in Stellenbosch, South Downloaded By: [Hendrichs, Jorge] At: 16:02 6 November 2009

Figure 4. A gamma batch irradiator in use at SENASA, Lima, Peru for insect irradiation. Biocontrol Science and Technology 343

Africa. To improve dose uniformity to the insects, each canister is rotated around its own axis, while the entire table (90 cm diameter) on which the eight canisters are situated is rotated around the source. This irradiator yields a dose rate of about 5 Gy/min with cobalt-60 activity of about 6.5 kCi. The volume of each canister is about 5 L. These panoramic irradiators are classified by the IAEA as Category II (dry storage) and Category IV (wet storage). Applications and the procedures for use for these two categories of irradiators are described in a Practical Radiation Safety Manual published by the IAEA (1996b).

Electron-beam irradiators To expose insects to electrons, they are treated in an electron-beam irradiator, which consists of a source of electrons (an electron accelerator that emits a narrow beam of electrons), a device to broaden the beam to cover the insect canister, a mechanism for transporting the canisters through the electron beam, and an operating system to control the exposure. An electron accelerator does not involve any radioactive material. It yields a narrow and intense electron beam, and thus the dose rate can be up to 1000 times greater than that for an isotopic gamma ray source. In an accelerator, electrons are introduced into an accelerating structure from an injector, where they are accelerated to the designed energy; the energy for the acceleration is derived from a variety of sources depending on the type of accelerator (ASTM 2007a). Medium-energy (between 300 keV and 5 MeV) electrons are commonly produced by potential-drop accelerators, whereas high-energy (more than 5 MeV) electrons are commonly produced using microwave-powered accelerators. Electrostatic accelerators can be used to accelerate electrons, but such systems have seldom been used for radiation processing applications. For all types of accelerators, the beam tube through which the electrons travel is under vacuum. Steering and focusing of the electron beam is accomplished with electromagnets and/or electrostatic fields. After leaving the accelerating structure and prior to reaching the insects, the electron beam is usually

Downloaded By: [Hendrichs, Jorge] At: 16:02 6 November 2009 dispersed to accommodate the canister size. This is typically done by sweeping the beam back and forth across the canister using an electromagnet with a varying magnetic field (referred to as ‘scanning’), although defocusing elements and scattering foils are also used. The electron beam reaching the insect canister may be characterised by the following parameters: electron beam energy (in MeV), which determines the penetration distance in the insect canister and thus dictates the useful canister size; average beam current (in mA), which determines the dose rate; beam power (product of beam energy and beam current, in kW), which determines throughput for a selected dose; beam (scan) width, which is generally selected to cover the canister size; and scan uniformity (uniformity of dose on the canister surface along the scanned direction). 344 K. Mehta

Since the penetration by electrons is relatively shallow, electron energy of less than a few MeV will not be suitable for insect irradiation. As seen in Figure 1, 5 MeV electrons can penetrate about 3.5 cm of water (density of 1 g cm3), and thus, since the density of packaged insects is about half of that of water (Parker, personal communication, 2006), the penetration will be about 7 cm (including the canister wall facing the electron beam). This defines the size of the canister (the dimension along the electron beam) suitable for acceptable dose variation. A larger canister may be used if a two-sided irradiation is carried out; however, one-sided irradiation is preferable for electron irradiation. Because of the slow decrease in dose with distance, this is easily done with photon irradiation; however, because of the sharp fall-off of dose in the case of electrons (Figure 1), a two-sided irradiation is much more complicated and can be problematic. Basically, the electron beam energy dictates the useful size of the canister for irradiation, and the average beam current (for a fixed beam energy) determines the throughput, e.g. number of canisters irradiated per hour. The dose rate at the location of the insect canister is generally much higher than that for a gamma irradiator, and hence the canisters move through the electron beam at a relatively high speed. However, unlike gamma irradiation, only one canister (in reality, only part of a canister) is irradiated at any instant. While in operation, there is a high radiation field in the irradiation room just as in a panoramic gamma irradiator; the irradiation room is shielded. However, when the accelerator is not operating, there is no radiation in the room. For irradiation, the canisters are placed on the conveyor outside the irradiation room, and they enter the room through a shielded labyrinth (similar to a panoramic gamma irradiator). The conveyor speed is adjusted according to the beam current and the dose requirement. Considering the dose required for the treatment of insects (taken to be an average of about 100 Gy) and the number of insects irradiated at a facility, the power requirement will be less than 1 kW, which is equivalent to about 70 kCi of cobalt-60. Currently, there is no commercial treatment of insects using electron-beam irradiators, since accelerators with 5 MeV and such low power levels are not commercially available. The power levels of commercial accelerators currently available are generally 10200 kW. However, use of electrons for insect irradiation will likely increase in the near future. In principle, similar to gamma irradiators, there could be two types of electron Downloaded By: [Hendrichs, Jorge] At: 16:02 6 November 2009 irradiators: one with a shielded room to house the accelerator where the products enter the room from outside for irradiation, and another being a self-contained irradiator where the shielding is part of the irradiator. Some manufacturers are considering producing small self-contained 5 MeV accelerators that would be suitable for insect treatment (Brown, personal communication, 2005). One possibility is a self-contained irradiator that houses a 5-MeV, 750-W accelerator along with a beam-scanning device that spreads the beam over 40 cm. This irradiator could deliver a dose of 100 Gy to insects at a throughput of about 6 kg per min. The cost of the entire irradiator would be about US$ 1 million.

X-ray irradiators When a beam of electrons strikes a high-atomic number material (called a converter), such as tungsten, a fraction of the electron energy is converted into X-rays. Biocontrol Science and Technology 345

Radiation generated in this manner (by rapid deceleration of electrons) is also known as bremsstrahlung (literally ‘braking radiation’ in German). Although their effects on irradiated materials (insects in this case) are generally similar to those of gamma rays, these types of radiation differ in their energy spectra, angular distributions, and dose rates. While gamma radiation from a cobalt-60 source has discrete energies, X-rays have a broad energy spectrum with a maximum equal to the energy of the incident electrons. Conventionally, the incident electron energy is referred to as the energy of the X-ray beam. For example, 5 MeV X-rays are generated by 5 MeV electrons, but the average photon energy in this X-ray beam is even less than that of cobalt-60 gamma rays. X-rays have the benefit of high penetration like that of gamma rays, and also the benefit of electrons in that X-rays do not need radioactive material (unlike gamma rays) for its generation.

High-energy X-ray irradiators A high-energy X-ray irradiator consists of a source of high-energy electrons, a converter to generate X-rays from these electrons, and a mechanism to transport the insect canisters through the X-ray beam. Since high-energy X-rays are produced by high-energy electrons, an electron accelerator is essential for generating them. Various types of accelerators referred to above may be used for this purpose. The X-ray conversion efficiency (X-ray power emitted in the forward direction divided by the electron power incident on the converter) increases with the electron energy and the atomic number of the converter material. The heavy metals, such as tantalum, tungsten or gold, are suitable materials because of their high atomic number and high melting point. In contrast to radiographic and therapeutic X-ray generators, which use small- diameter electron beams to generate well-collimated X-ray beams, radiation processing applications require electron beams with large cross-sections and X-ray converters with large areas to cover the insect canisters. As mentioned above, electron beams may be dispersed by scanning magnets, defocusing magnetic lenses, or scattering foils. The dose rate associated with high-energy X-rays is considerably smaller than Downloaded By: [Hendrichs, Jorge] At: 16:02 6 November 2009 that for the incident electron beam. Since high-energy X-ray irradiators need electron accelerators, they have the same shielding requirements, and the associated high cost. This type of irradiator requires a mechanism for transporting the insect canisters through the X-ray beam, such as a conveyor system. Because of the high cost, there are currently no high-energy X-ray irradiators in use for insect irradiation.

Low-energy X-ray irradiators A low-energy X-ray irradiator consists of an X-ray tube, and a device to transport the insect canister through the X-ray beam. The X-ray tube consists of an electron source (generally a heated wire, a filament which emits electrons), and a converter to generate X-rays. These electrons are electrostatically accelerated through a small potential difference (thus no large and costly accelerators are needed), and thus the energy will be in the range of a few hundred keV. These X-ray irradiators require 346 K. Mehta

much less shielding, and thus self-contained irradiators, just like self-contained gamma irradiators (Figure 2), are possible. Such self-contained X-ray irradiators have been marketed for the last several years for the specific purpose of blood irradiation (which requires a dose of about 25 Gy), and between 50 and 100 units are operating successfully at hospitals and medical institutes in North America. Canister volume is about 1.5 L, and the dose rate for this irradiator is about 5 Gy/min, which may be relatively low for insect irradiation (requiring dose of about 100 Gy) on a commercial basis. However, the original developer has recently introduced a new patented tube design that yields much higher dose rates. These can be configured to address the requirements of the programme/customer (dose and throughput). The tube configuration can also further be optimised to be more efficient, thus reducing the cost. A prototype has been used in the USA for seafood research for about 2 years. It used two tubes with a rating of 5 kW each, and yielded a dose rate of about 25 Gy/min (Kirk, personal communication, 2005). The IAEA has recently purchased a unit with one tube rated at 10 kW and surrounded by 5 canisters for their SIT programmes (Figure 5). Another one at the design stage uses two tubes with a power rating of 10 kW each, and would yield a dose rate of 100 Gy/min. A semi-automatic unit has been recently installed in Panama with the dose rate calibrated to 12 Gy/min for insect irradiation, and uses four tubes with a total power of 25 kW. This unit is about 1.52.8 m and 2 m in height (Figure 6). It utilises a conveyor and the insects are continuously irradiated in flat trays. These low-energy X-ray irradiators use tubes that generate X-rays with a maximum energy of 160 keV, with an average photon energy in the range of 7075 keV; penetration is thus not very deep (see Figure 1). Hence, the canisters are smaller compared to those used for gamma irradiators; generally, flat trays are more suitable. Downloaded By: [Hendrichs, Jorge] At: 16:02 6 November 2009

Figure 5. This self-contained X-ray irradiator (RS 2400) comes in two units. The unit in the foreground houses the X-ray tube and the five canisters which can be loaded/unloaded through an opening at the top next to the control panel. The other is the water-cooling unit for the X-ray tube. Biocontrol Science and Technology 347

Figure 6. A self-contained X-ray irradiator. (Courtesy of Rad Source Technologies, Inc., USA.)

Technical parameters of four types of irradiators For the sterilisation process, the technical parameters of primary interest are canister size, dose rate and throughput. These are generally governed by the radiation energy, the current/flux (i.e. number of photons or electrons emitted), and the power (product of energy and current), respectively. These characteristics for the four types of irradiators discussed above are listed in Table 1.

Selection of radiation source General To make a meaningful selection of the type of irradiator that is suitable for a specific application, such as irradiation of insects, it is essential to review and compare the

Downloaded By: [Hendrichs, Jorge] At: 16:02 6 November 2009 Table 1. Relevant technical parameters of four types of irradiator.

Irradiator 0 Gamma rays X-rays X-rays Parameter ¡ (Co-60) Electrons (high-energy) (low-energy)

Energy (MeV) Fixed Choice at Choice Presently (determines canister purchase at purchase fixed size) Current/flux Activity (Ci) Current (mA) Current (mA) Current (determines dose Choice at Variable Variable (mA) rate) purchase Variable Decays Power (determines Activity Currentenergy Current Current throughput) energy energyconversion energy Decays efficiency conversion efficiency 348 K. Mehta

key characteristics of the various available irradiators. The needs of the technology and the specific programme can then be compared against these various character- istics; this can assist the relevant decision-makers in making an optimum selection. Such key characteristics relevant to insect irradiation include: dose rate, canister size, dosimetry, cost of equipment and accessories, total treatment cost: unit cost, technology, utility requirements, and safety and security.

Dose rate There is some limited data that show that low dose rates reduce the adverse effect of radiation on the insects (Bakri et al. 2005b). As regards to the irradiation process, the dose-rate value at the location of the insect canister determines how long it will take to deliver a required dose, and thus the throughput. If the dose rate is too small, the long treatment time will result in a low and, depending on the size of the insect facility, uneconomical throughput. On the other hand, if the dose rate is too high, the canisters will have to be in the irradiation position for a very short time, which will make it difficult to consistently give the correct dose. Thus, there is in essence a window (albeit a wide one) of dose rate that is suitable for the process. Dose rate depends principally on two factors: current/flux of the source (that is, radioactivity for an isotopic source, and current for an electron accelerator, see Table 1) and the irradiation geometry (including the distance from the source). For example, in a self-contained gamma irradiator, the canister is surrounded by the source, and thus the dose rate is relatively high as compared to the situation for a panoramic irradiator where the gamma rays reach the canister from one direction only. In the later case, the disadvantage of low dose rate and thus longer irradiation time is mitigated by the fact that several large canisters can be treated simultaneously Downloaded By: [Hendrichs, Jorge] At: 16:02 6 November 2009 (see Figure 4). There are basic differences, however, among gamma rays from a radioisotope (like cobalt-60), electrons from an accelerator, and X-rays. These differences are partly based on how the radiation is emitted from the source. The fact that the gamma radiation is emitted in all directions, and that it penetrates much deeper, makes the dose rate for a gamma irradiator much smaller than in the case of an electron accelerator (see Table 2). However, this also allows a gamma irradiator to process several large boxes simultaneously. Also, the dose rate can be increased for a gamma irradiator by adding more activity to the source, which may however require more shielding. For an electron accelerator, the dose rate can be reduced to a manageable value by selecting a low-powered accelerator. However, as mentioned earlier, commercially there are no accelerators available for low power (less than 1 kW) and high electron energy (more than 5 MeV). It is of course possible to operate a high-power accelerator at very low power (by reducing the current), but that would be completely uneconomical since such units would cost US$ 1 to 2 million. Biocontrol Science and Technology 349

Table 2. Radiation field distribution for gamma rays, X-rays and electron sources.

Gamma rays X-rays (isotopic source, Electron beam X-rays (low-energy, panoramic irradiator) (accelerator) (high-energy) RS-2400)

Emitted in all Emitted mainly in one Emitted mainly in Emitted in all directions, direction, radiation is one direction, directions, radiation is radiation is focused radiation is focused distributed distributed High penetration Low penetration High penetration Low penetration Low dose rate High dose rate Moderate dose rate Moderate dose rate Long processing time Short processing time Moderate processing Moderate processing time time Several large canisters Single flat canister Single large canister Several canisters processed together processed at a time processed at a time processed together

X-rays interact with matter in a similar manner to gamma rays, so they have high penetration. High-energy X-rays generated from high-energy (more than 5 MeV) and high-power accelerators will have a reasonable dose rate since the X-ray conversion efficiency is only a few percentage points. Low-energy X-ray irradiators with enough power will also have an adequate dose rate.

Canister size The absorbed dose that is used to induce the required effect in insects is of prime importance to the pest management programmes that use these insects. As dose increases, the desirable effect increases; however, insect fitness will decrease. Thus, optimisation is necessary in selecting treatment dose to balance the desired effect and insect fitness, taking into consideration programme requirements (Parker and Mehta 2006). In reality, because of the unavoidable dose variation within a canister (Figure 1),

Downloaded By: [Hendrichs, Jorge] At: 16:02 6 November 2009 an acceptable range of dose to be given to the insects has to be defined according to the specific programme requirements. Most often, programmes or regulatory officials specify a minimum dose that all insects must receive to ensure sufficient radiation. Due to dose variability, most insects actually receive a dose that is somewhat higher than that minimum. Thus, it is essential that dose is fairly uniform within a canister; the goal is to expose all insects sufficiently without treating large proportions with doses that are high enough to substantially reduce their fitness. Irradiator operators can often adjust process parameters to achieve the desirable dose uniformity in the canister. Canister size is one of those parameters; dose variation within the canister depends significantly on the canister size, besides the radiation energy and the type of radiation. As seen in Figure 1, photons (gamma rays and X-rays) can travel much deeper in the irradiated matter than electrons of similar energy. However, dose is decreasing in both the cases with distance, and thus there is a limit on the size of the canister that can yield acceptable dose variation. If the size of the canister is larger (in 350 K. Mehta

the direction of the radiation) compared to the rate of decrease of dose, the dose uniformity in the canister would be significantly reduced making the canister unsuitable for the irradiation treatment. Thus, the limits on the size of the canister depend on the type of irradiator. It is obvious from Figure 1 that large canisters can be used with cobalt-60 gamma rays and high-energy X-rays. And of course the acceptable size can be substantially increased (generally doubled) by irradiating from two or more sides (for example, by rotating the canister). As mentioned earlier, this is easily done with photon irradiation; however, two-sided irradiation is much more complicated and can be problematic in the case of electron irradiation. First, irradiating a flat canister of insects from two sides involves either two accelerators or passing the canister under the beam a second time after turning it over. Also, because of the sharp fall off of dose, slight variation in the bulk density of the insects from canister to canister or variation in the electron beam energy during irradiation can cause either under-dosing or over-doing the insects in the centre part of the canister.

Dosimetry As mentioned earlier, dose and distribution (variation) of dose in an insect canister are very important parameters, and thus should be determined and controlled for the efficacy of the irradiation process. A dosimetry system is generally used to determine dose. There are several types of dosimetry systems available commercially that are suitable for photons and electrons (ASTM 2007b,c). Some of the systems can be used for both types of radiation, while some may be suitable only for photons. Also, a majority of the dosimeters that are suitable for gamma rays are also suitable for X- rays (Mehta, Kojima, and Sunaga 2003). Thus, careful selection of the dosimetry system is necessary depending on the irradiator type. Considering various factors, the Gafchromic† dosimetry system has been selected by the IAEA based on the specific requirements of insect irradiation (IAEA 2004).

Cost of equipment and accessories Since small self-contained gamma irradiators have been available for a long time, it is Downloaded By: [Hendrichs, Jorge] At: 16:02 6 November 2009 not surprising that currently a large majority of irradiators used for insect sterilisation in pest control programmes with an SIT component are of this type; with either cobalt or caesium as a radiation source (cobalt is more common). The wide usage of such irradiators also opens up the market, resulting in a reasonable cost. The main cost components for such irradiators are the radiation source (initial source plus periodic replenishment because of radioactive decay), lead shielding and transportation. The construction itself is quite simple. Another possibility is small- size panoramic irradiators, preferably used in batch mode. A few are currently in use for insect sterilisation as mentioned earlier. For a relatively large biological control programme, these may be affordable. Nearly all the commercially available electron accelerators with energy greater than a few MeV have high power. High energy also requires heavy shielding. Thus, accelerator and shielding both tend to increase cost for electron irradiators. In addition, an accelerator needs a conveyor system. For some applications, such as polymerisation of cables and sterilisation of medical products, these high-power Biocontrol Science and Technology 351

accelerators are necessary since the relevant doses are in the region of 20100 kGy, and the quantity of the product irradiated is very large. A high-energy X-ray irradiator costs as much as or slightly more than a high- energy electron accelerator, due to the added cost of the X-ray converter. On the other hand, low-energy X-ray irradiators, which can be self-contained, are reason- ably priced. As mentioned earlier, there are many units already operating for several years for blood irradiation, and more recently, other equipment with higher dose rates have been introduced. If these prove to be suitable, the cost of these low-energy X-ray irradiators would be reasonable. The first two rows of Table 3 list the capital cost and operational cost for four types of irradiator described in the next Section.

Total treatment costs Four types of irradiator have been selected for calculating the treatment cost for a unit volume (1 L) of insects since they are currently most suitable for insect irradiation. In some cases, specific models have been selected to illustrate some realistic specifications; this does not, however, imply any endorsement for them. These four irradiators are:

Table 3. Economic and technical characteristics of four types of irradiators.

RS 2400 - low-energy Semi-automatic Panoramic X-rays - low-energy X-rays Gammacell 220 gamma -10kW -25kW - 20 kCi cobalt irradiator - self-contained - self-contained - self-contained - 60 kCi cobalt - 1 tube - 4 tubes

Capital cost 178,000 1,200,000 260,000 400,000 (equipment 20,000 (incl. transport) 5,000 10,000 transport) (US$) 5,000* Operational cost 18,300 (cobalt)$ 42,000 (cobalt) X-ray tubes X-ray tubes (US$/year) (183,000/10) (210,000/5) electricity electricity Operators 1212 per shift Downloaded By: [Hendrichs, Jorge] At: 16:02 6 November 2009 Average 139 4.5# 12.5 12 dose rate (average over (average over (Gy/min) 10 years) 5 years) Volume of 2 (1 canister) 700# 20 16 (4 trays) irradiation (L) (several (5 canisters) canisters) Exchange time 1§ 10 3 Continuous with Tex (min) 10% dead time Throughput’ 70 1310 110 105 (litres/hour)

*This irradiator is supplied with an internal conveyor. However, it would need a short conveyor outside the unit at an estimated cost of US$ 5000. $This replenishing estimate is by the supplier of the Gammacell 220. However, there are other companies that offer this service at a lesser cost. For example, the estimate of the Institute of Isotopes, Hungary, is about 65% of this value; however, this procedure involves transfer of cobalt at the insect facility, and would require a water pool, at least 1.5 m deep. #This value is based on the irradiation configuration of the Lima, Peru irradiator. ’The treatment dose is assumed to be 100 Gy. §A smaller exchange time will probably entail an additional operator per shift. 352 K. Mehta

Gammacell 220 manufactured by MDS Nordion, Inc1. This is a self- contained gamma irradiator and thus needs no external shielding. The assumed loading is 20 kCi of cobalt-60. Only one canister can be irradiated at a time (Figure 2). Panoramic irradiator operated in a batch mode, similar to the one in use in Peru (Figure 4). The assumed loading is 60 kCi of cobalt-60. It requires a shielded irradiation room as well as a shielded storage container (or a room) to house the source when not in use. Several canisters can be irradiated at a time. RS 2400 low-energy X-ray irradiator manufactured by Rad Source Technol- ogies, Inc. This is a self-contained unit and thus no external shielding is needed. It uses one tube rated at 10 kW. The maximum energy of these X-rays is 150 keV, with the average energy between 70 and 75 keV. Five canisters can be irradiated at a time (Figure 5). Semi-automatic low-energy X-ray irradiator manufactured by Rad Source Technologies, Inc. This is a self-contained unit and thus no external shielding is needed. It uses four tubes with a total power of 25 kW. The X-ray energy is similar to that for RS 2400. The insects can be continuously irradiated in four trays with a conveyor system (Figure 6).

Table 3 compares relevant economic data for operating these four irradiators. It also gives values of some technical parameters used for these analyses. The rationale for selecting these values as well as the assumptions made for the analyses are explained below. The capital cost has two components as shown in Table 3; the first is an estimated cost of the equipment. The supplier should be contacted for a more current and accurate price. The second figure is the cost of transportation from the supplier to the buyer, which will vary depending on the destination. For the calculation of the unit cost, it is assumed that the total capital cost is amortised over 10 years; thus, 10% of the capital cost may be viewed as an annual debt-servicing cost. The operational cost for the gamma irradiators includes the cost of replenish- ment of the cobalt source. Replenishment is a major procedure and is carried out either by the original supplier of the irradiator or some other organisation such as the national nuclear regulator. The irradiator will be out of service for a few days during this procedure. The radiation source is replenished when the throughput Downloaded By: [Hendrichs, Jorge] At: 16:02 6 November 2009 decreases to an uneconomical or impractical level for the pest control programme (e.g. 50% of the initial value). The initial dose rate is quite high in the case of Gammacell 220 (see Table 3), and thus the handling time between two irradiations is longer than the irradiation time. This results in a relatively slow decrease in the throughput with time (see Equation 1). For example, in the case of a Gammacell 220 with an initial activity of 20 kCi, even though the activity decreases to 50 and 25% in about 5.3 and 10.6 years, respectively, the corresponding throughput decreases only to 80 and 56%. Thus, for such a unit it is assumed for the present analysis that cobalt would be replenished after 10 years2. However, for a panoramic irradiator, the hourly

1 During the final preparation of this article, Nordion has discontinued this product line. 2 At that point, however, the insect facility may decide to purchase a new irradiator instead of replenishing cobalt in the existing one. It may also decide to keep the old one with a lower dose rate. This decision will depend on the insect production rate and other prevailing conditions at that time. Biocontrol Science and Technology 353

throughput decreases to 61% after 5.3 years and to 35% after 10.6 years, and thus it is assumed that the source will be replenished after 5 years. The annual operational cost is then arrived at by dividing the cost of replenishing cobalt by 10 or 5, as shown in Table 3. The operational cost for the X-ray irradiators includes the cost for tube replacement and the use of electricity for these tubes. The manufacturer guarantees 2000 h of operation for these tubes. Since the tube is dismountable, should the filament fail, it can be replaced without the need for a completely new tube. The complete tube is expected to be serviceable for 20,000 h. The replacement cost for one X-ray tube for the total 20,000-h cycle is about US$ 45,000. Thus, the average cost for a 2000-h period is US$ 4500, which is the value averaged over the entire cycle of changes. The cost of electricity can be between US$ 0.1 and 0.3 per kWh. The cost of labour is based on an annual salary of US$ 15 000 per person per shift (can be adjusted based on the local labour costs), a shift being 8 h a day for 7 days a week (that is, 56 h/week). It is expected that only one person would be necessary for the operation of the Gammacell 220 and for the RS 2400, while two persons would be needed for routine operation of the panoramic gamma irradiator and also for the semi-automatic X-ray irradiator. The values in Table 3 for the dose rate and the irradiation volume are estimates and could vary by 1020%. The value for the dose rate is an average value within the canister and is based on the supplier information, except for the panoramic irradiator, which is based on the measured values in the Peru irradiator (Figure 4). For the gamma irradiators, the dose rate shown in Table 3 is the value at the midpoint between source replenishments. The exact value of the irradiation volume will depend on the acceptable dose variation in the canister. For the calculations of the throughput in litres of insect treated in 1 h, it is assumed that the goal is to deliver an average dose of 100 Gy. The throughput also depends not only on the dose rate and the volume of irradiation, but also on the handling time between the irradiation of two canisters or two batches. The estimated value of this handling or exchange time (Tex) is also listed in Table 3. The value of the hourly throughput for the four irradiators listed in Table 3 is based on the following expression: Downloaded By: [Hendrichs, Jorge] At: 16:02 6 November 2009 60 (min=h) irradiation volume (L) Hourly throughput (L=h) : (1) [dose (Gy)=dose rate (Gy=min)] T ex (min)

To calculate the unit cost for the treatment, it is assumed that the irradiator is operated continuously either for one shift or two shifts. Assuming that the irradiator is operating for 50 weeks per year, one shift would be 2800 h/year (56 h/week50 weeks/year) and two shifts would mean 5600 h/year (112 h/week50 weeks/year) of operation. The unit cost (total cost for irradiation of 1 L of insects) is divided into three components, one arising from the capital cost (equipmenttransport), the second from the operational cost, and the last from the labour cost. The unit cost is calculated using the following expressions:

Unit cost (US$=L)capital costoperational costlabour cost (2) 354 K. Mehta

10%of capital cost (US$=year) Capital cost (3) throughput (L=h) UR (h=year) replacement cobalt (US$=year) Operational (gamma) (4) throughput (L=h) UR (h=year) Operational (X-ray) [(UR=2000 h) 4500 (US$) tubes] [power (kW) UR ER] throughput (L=h) UR (h=year) (2:25 No of tubes) (power ER) (5) throughput (L=h)

where, UR is the usage rate (number of hours the irradiator is used per year, either 2800 or 5600), and ER is the cost of electricity (assumed here to be US$ 0.2 per kWh of electricity). ‘Power’ is the total power (kW) consumed by the X-ray irradiator, which is given in Table 3. US$15;000 number of operators number of shifts Labour cost : (6) throughput (L=h) UR (h=year) It can be seen from these expressions that the capital and the labour components of the unit cost for all irradiators, and the operational component for the gamma irradiators are inversely proportional to the total quantity of insects treated during the year (throughputUR). However, the operational component for the X-ray irradiators is constant, i.e. it is independent of the quantity treated. Table 4 shows the results of the unit cost analyses for these four irradiators for the case when the irradiator is operating for either 2800 h (full 1 shift) or 5600 h (full 2 shifts) during the year. It would be rare, however, that the weekly production of the insect rearing facility would exactly match the maximum weekly irradiation capacity of the irradiator

Table 4. Weekly throughput and unit cost for four types of irradiator when used at maximum capacity for one- and two-shift operations. Downloaded By: [Hendrichs, Jorge] At: 16:02 6 November 2009 Weekly throughput (L/week) Unit cost (US cents/L)

1-shift 2-shift 1-shift 2-shift

Gammacell 220 3920 7840 27.1* 17.5 (Gamma rays) (10.1 9.3 7.7) (5.1 4.7 7.7) Panoramic 73 360 146 720 5.2 3.0 (Gamma rays) (3.3 1.1 0.8) (1.6 0.6 0.8) RS 2400 6160 12 320 17.4 13.1 (X-rays) (8.6 3.9 4.9) (4.3 3.9 4.9) Semi-automatic 5880 11 760 37.6 30.6 (X-rays) (14.113.310.2) (7.113.3 10.2)

*The three components are, respectively, capital cost, operational cost (replenishment for gamma irradiators or tube replacement cost plus electricity use for X-ray irradiators), and the labour cost (either one- or two-shift). Biocontrol Science and Technology 355

when operated for 56 or 112 h per week (as shown in Table 4). Thus, the irradiators may not be used for all the 56 or 112 h per week. The unit costs would then be different than those shown in Table 4. To calculate the unit cost for this non-optimal use of the irradiator, it is convenient to use ‘weekly throughput’ of the rearing facility instead of ‘UR’ in the above expressions. These two quantities are related through the following expression:

UR (h=year) hourly throughput (L=h) Weekly throughput (L=week) : (7) 50 (week=year)

The dependence of the total unit cost on the weekly throughput is shown in Figure 7; where weekly throughput is the weekly production of the insect rearing facility. For the Gammacell 220, at the lowest value of the weekly throughput shown (1000 L/week), the total unit cost would be slightly more than US$ 1. This value then decreases rapidly (open circles) as the weekly throughput increases and reaches a minimum of US$ 0.271 per L for a 1-shift operation at about 3920 L/week (see Table 4). At this point, the irradiation capacity of the Gammacell is reached for a one-shift operation. If the insect rearing facility produces more insects than this, it will need a two-shift operation; and the unit cost first increases slightly (filled circles in Figure 7) and eventually decreases again till it reaches a minimum value of US$ 0.175 for a two-shift operation at 7840 L/week (Table 4). This is the best performance that can be achieved with this type of irradiator. If the production rate of the insectary is more than 7840 L of insects per week, then either a three-shift operation or two irradiators are needed. In the present analysis it is assumed that two irradiators with the same characteristics are used for a facility with a higher production rate. A one- shift operation for this second unit is shown by open circles (Figure 7), and a two- shift operation for the second unit by filled circles. However, the first unit will be operating at two shifts all the time. As mentioned above, it is assumed that the cost of electricity (for the X-ray irradiators) is US$ 0.2/kWh. If the cost is either US$ 0.1 or 0.3 per kWh, the unit cost curve for these irradiators will shift down or up by US$ 0.009 or US$ 0.024 for RS 2400 and the semi-automatic X-ray irradiator, respectively. Figure 7 will assist the insect rearing facility manager to select a suitable type of Downloaded By: [Hendrichs, Jorge] At: 16:02 6 November 2009 irradiator for the specific needs. Based on the weekly production of the facility, the value selected on the x-axis determines the unit cost for the various types of irradiator. It is recommended that the value of the production rate should also reflect the future requirements, at least over the following few years. The outcome of such analyses would to an extent depend on each specific case and the location of the irradiation facility. Thus, this analysis should be considered only as a guideline to assist with further detailed analysis for a specific case. In selecting the most suitable irradiator, unit cost is only one of the parameters. The others include available capital for investment, concerns with radioactive material, availability of technology, etc.

Technology The technology necessary for gamma irradiators is simple, especially for the self- contained types, and for the panoramic irradiators with batch operation and 356 K. Mehta Downloaded By: [Hendrichs, Jorge] At: 16:02 6 November 2009

Figure 7. Dependence of unit cost for insect irradiation on weekly throughput (production) of the insect rearing facility for two types of cobalt-60 gamma irradiator and two types of low- energy X-ray irradiator. Open circles (k) refer to a one-shift operation of the irradiator, while close circles () refer to a two-shift operation. The first set of open and close circles refers to the first irradiator, while the second set refers to the second irradiator of the same type. Biocontrol Science and Technology 357

dry-storage. In both cases, no conveyor system is necessary, and the electronics is not complex. In most cases, the only moving parts are either the source or a device (for example, drawer or shuttle) that moves the canister into and out of the radiation field. In some cases, rotating turntables are also used. Such simple technology makes these gamma irradiators more reliable, and in the case of malfunction they are easy to fix. It also means very little downtime during which the irradiator is not available for irradiation. Regular periodic maintenance includes replenishment of the radiation source to compensate for the radioactive decay; this procedure is elaborate and will require external assistance, generally provided by the supplier of the source. Electron accelerators (necessary for electron-beam irradiators as well as for high- energy X-ray irradiators) are intrinsically more elaborate, and failure in the system requires expertise in the electronics and related technology to repair it. However, suppliers are making these systems much more robust and reliable, with a resultant decrease in downtime. The conveyor system will require periodic maintenance; however, this should not pose a problem. A low-energy X-ray irradiator is more robust and much less can malfunction. Like the isotopic source in a gamma irradiator, the X-ray tube has a finite life and thus needs to be replaced periodically, a task that is relatively easy. Being a self- contained X-ray irradiator, there is no conveyor system to be maintained.

Utility requirements The main utility requirements are an electrical power supply and, in some cases, cooling water: Gamma irradiators: self-shielded units: power for the canister movement (normal line voltage); panoramic unit (batch mode): power for the source movement, and ventilation of the irradiation room and if applicable, the source storage room. Electron irradiators: electrical power to operate the accelerator (any interruption would shut down the electron beam) and for the canister conveyor system. Downloaded By: [Hendrichs, Jorge] At: 16:02 6 November 2009 X-rays irradiators: high-energy: same as for electron irradiators, as well as cooling for the X-ray converter; and low-energy: power for the electron injector and for the X-ray tubes, as well as cooling for the X-ray converter.

Safety and security All irradiators are designed to keep the radiation exposure and dose to workers ‘as low as reasonably achievable’ (ALARA), and within predetermined levels. These dose limits are based on the recommendations of several agencies of the United Nations (UN), including the International Atomic Energy Agency (IAEA), Food and Agriculture Organization of the United Nations (FAO), and World Health Organiza- tion (WHO) (IAEA 1996c). Appropriate safety methods and procedures have been 358 K. Mehta

developed for each type of irradiator, andwhen operated correctly with the appropriate safeguards, they are safe and easy to use. Gamma irradiators are usually licensed by national atomic energy authorities, which set requirements such as restricting access to certain areas and authorised persons, a periodic survey of the radiation field in the vicinity where workers could be present, the use of personal radiation dosimeters, and the availability of radiation survey meters. On the other hand, electron accelerators or X-ray units are generally regulated by occupational safety and health agencies, which also require operator training, etc., and may require use of personal dosimeters. All these requirements are specifically aimed at protecting the workers from radiation. In addition, irradiator design incorporates interlocks that prevent unintentional access to areas with high-radiation fields. When the useful life of a gamma source is over, the irradiator or the source pencils are usually returned to the supplier for storage, reuse, recycling, or disposal. This is now becoming an elaborate and costly procedure. Security is an issue because of the use of radioactive material, either cobalt-60 or caesium-137. The source material is doubly sealed according to ISO standards and there is minimal risk of the radioactive material leaking from the capsule, especially in dry storage facilities. The main concern is the physical security of the radioactive material against unauthorised use of the material. Thus, in recent times, the transportation of radioactive material is getting much more elaborate and restrictive. The public is also more concerned nowadays with the eventual disposal of the radioactive material. However, in reality, that is a much smaller problem compared to the disposal of the used uranium fuel from the nuclear power reactors. Additional security is always prudent where a radioactive material is used. These issues are coming increasingly to the fore and are introducing difficulties with the transporta- tion and operation of isotopic radiation sources. These particular issues do not arise with electron accelerators or X-ray irradiators; when the power switch is turned off, the radiation disappears along with the related concerns.

The selection process Photons and electrons have similar radiation effects, so the choice of the source for treating insects is based on other considerations, such as those mentioned above. The

Downloaded By: [Hendrichs, Jorge] At: 16:02 6 November 2009 power emitted by a cobalt-60 source of 50 kCi is roughly equivalent to that of a 750- W electron accelerator. Since caesium-137 emits about one-quarter of the energy per disintegration compared to cobalt-60, about 200 kCi of caesium would be needed to generate the same power. As discussed above, there are advantages and disadvantages of each type of irradiator. Thus, it is the responsibility of the facility manager to review the requirements of the programme concerned and select the most appropriate irradiator. The selection process will require optimisation and some level of flexibility. To summarise: Gamma-ray irradiators (self-contained and batch mode panoramic): relatively easy to operate, dose rate can be controlled by selecting appropriate activity of the source, especially for panoramic irradiator (to match the required throughput), high penetration allows the use of larger canisters, Biocontrol Science and Technology 359

do not need conveyor systems, and issues related to associated radioactivity.

Accelerator based irradiators (electron-beam irradiators and high-energy X-ray irradiators): maintenance-intensive and relatively less easy to operate, need conveyor systems to move the canisters, electrons have shorter range, and thus smaller canisters are required, high throughput 5MeV accelerators with low power are not currently commercially available (thus, there is not yet much field experience), and no issues related to radioactivity.

Low-energy X-ray irradiators (self-contained): relatively easy to operate, acceptable dose rate, low penetration, and thus smaller canisters are required, do not need conveyor systems, currently only one supplier who is just coming into the market (thus, there is not yet much field experience), no issues related to radioactivity, less regulatory requirements, and low security risk.

There is an important advantage of having a self-contained unit (either gamma ray or X-rays). It can be very easily re-located to another location within the insect rearing facility or even moved entirely to a new location. On the other hand, it would be very complex and costly to relocate a panoramic facility.

Conclusions This review discusses several relevant characteristics of various radiation sources that Downloaded By: [Hendrichs, Jorge] At: 16:02 6 November 2009 are commercially available and suitable for pest control programmes that use natural enemies. The initial investment can vary from US$ 200,000 to about US$ 1 million. The selection would depend on several factors discussed extensively in this review. From the price of these irradiators, it is obvious that they may not be attractive to producers of garden products. However, they may be attractive to major producers of sterile insects or natural enemies. Also, with time it is envisaged that small producers of biological control agents will merge into a few major producers, especially with the rising international market. With usage, the price of these irradiators would also become more reasonable and affordable for some smaller producers. This is apparent from the fact that some of the irradiators are just coming onto the market. It is hoped that this review will make the managers of the pest management programmes aware of the available radiation sources and their advantages and disadvantages, and help them to select the most appropriate irradiator so as to improve their programme at an affordable cost. 360 K. Mehta

Acknowledgements I would like to thank the staff of the Insect Pest Control Section of the IAEA for many stimulating discussions, especially W. Enkerlin, J. Hendrichs and M. Vreysen, who also provided support for some of the figures in this paper.

References Adem, E., Watters, F.L., Uribe-Rendo´n, R., and de la Piedad, A. (1978), ‘Comparison of Co- 60 Gamma Radiation and Accelerated Electrons for Suppressing Emergence of Sitophilus spp. in Stored Maize’, Journal Stored Products Research, 14, 135142. American Society for Testing and Materials (ASTM), International (2007a3), ‘Standard Practice for Dosimetry in an Electron Beam Facility for Radiation Processing at Energies between 300 keV and 25 MeV, ISO/ASTM 51649’,inAnnual Book of ASTM Standards (Vol. 12.02), Conshohocken, PA, USA: Author, pp. 10801108. American Society for Testing and Materials (ASTM), International (2007b3), ‘Standard Guide for Selection and Calibration of Dosimetry Systems for Radiation Processing, ISO/ASTM 51261’,inAnnual Book of ASTM Standards (Vol. 12.02), Conshohocken, PA, USA: Author, pp. 968986. American Society for Testing and Materials (ASTM), International (2007c3), ‘Standard Guide for Dosimetry for Sterile Insect Release Programs, ISO/ASTM 51940’,inAnnual Book of ASTM Standards (Vol. 12.02), Conshohocken, PA, USA: Author, pp. 11821193. Attix, F.H., and Roesch, W.C. (eds.) (1968), Radiation Dosimetry (Vol. 1), New York, London: Academic Press. Bakri, A., Heather, N., Hendrichs, J., and Ferris, I. (2005a), ‘Fifty Years of Radiation Biology in Entomology: Lessons Learned from IDIDAS’, Annals of the Entomological Society of America, 98, 112. Bakri, A., Mehta, K., and Lance, D.R. (2005b), ‘Sterilizing Insects with Ionizing Radiation’, in Sterile Insect Technique Principles and Practice in Area-Wide Integrated Pest Management, eds. V.A. Dyck, J. Hendrichs and A.S. Robinson, Dordrecht, The Nether- lands: Springer, pp. 233268. Bartlett, A.C., and Staten, R.T. (1996), The Sterile Insect Release Method and Other Genetic Control Strategies, http://ipmworld.umn.edu/chapters/bartlett.htm British Crop Protection Council (BCPC) (2004), The Manual of Biocontrol Agents, Hampshire, UK: BCPC Publications. Bracken, G.K., and Dondale, C.D. (1972), ‘Fertility and Survival of Achaearanea tepidariorum (Araneida: Theridiidae) on a Diet of Chemosterilized Mosquitoes’, Canadian Entomologist, 104, 17091712. Brinston, R.M. and Norton, J.L. (1994), ‘Cobalt-60: The Heart of Gamma-Radiation Sterilization’, Medical Device Technology, May, 1416. Downloaded By: [Hendrichs, Jorge] At: 16:02 6 November 2009 Carpenter, J.E. (1997), ‘Development and Integration of Alternative Management Strategies Using Inherited Sterility and Natural Enemies to Control Lepidopteran Pests’, Report for IAEA Consultants Group on Use of Nuclear Techniques in Biological Control, 1997 April 1418; Vienna, Austria: International Atomic Energy Agency. Carpenter, J.E. (2000), ‘Area-wide Integration of Lepidopteran F1 Sterility and Augmentative Biological Control’,inArea-wide Control of Fruit Flies and other Insect Pests, ed. K.H. Tan, Penerbit Universiti Sains Malaysia, Penang, pp. 193200. Codex Alimentarius (2003), Codex General Standard for Irradiated Foods, CODEX STAN 106-1983, Rev.-2003, and Recommended International Code of Practice for the Radiation Processing of Food, CAC/RCP 19-1979, Rev. 2003, Rome: Food and Agriculture Organization and World Health Organization, www.codexalimentarius.net/. Dohino, T., Tanabe, K., and Hayashi, T. (1994), ‘Comparison of Lethal Effects of Electron Beams and Gamma Rays on Eggs of Two Spotted Spider Mite, Tetranychus urticae Koch (Acari: Tetranychidae)’, Research Bulletin of the Plant Protection Service, Japan, 30, 6973.

3 Each Standard is updated about every 5 years. The latest Annual book of ASTM standards should always be consulted. Biocontrol Science and Technology 361

Dyck, V.A., Hendrichs, J., and Robinson, A.S. (eds.) , (2005), Sterile Insect Technique Principles and Practice in Area-Wide Integrated Pest Management, Dordrecht, The Netherlands: Springer, 787 p. Food & Agriculture Organization of the United Nations, International Atomic Energy Agency, World Health Organization (FAO/IAEA/WHO) (1999), High-Dose Irradiation: Wholesomeness of Food Irradiated with Doses above 10 kGy, Joint FAO/IAEA/WHO Study Group, Technical Report Series 890, Geneva, Switzerland: World Health Organiza- tion. Food & Agriculture Organization of the United Nations, International Atomic Energy Agency, U.S. Department of Agriculture (FAO/IAEA/USDA) (2003), Manual for Product Quality Control and Shipping Procedures for Sterile Mass-Reared Tephritid Fruit Flies. Version 5.0, Vienna, Austria: International Atomic Energy Agency, www.iaea.org/pro grammes/nafa/d4/index-tephritid.html Flint, H.M., Wright, B., Sallam, H., and Horn, B. (1975), ‘A Comparison of Irradiated or Chemosterilized Pink Bollworm Moths for Suppressing Native Populations in Field Cages’, Canadian Entomologist, 107, 10691072. Greany, P.D., and Carpenter, J.E. (2000), ‘Use of Nuclear Techniques in Biological Control’,in Area-wide Control of Fruit Flies and other Insect Pests, ed. K.H. Tan, Penerbit Universiti Sains Malaysia, Penang, pp. 221227. Guerra, A.A., Wolfenbarger, D.A., Hendricks, D.E., Garcia, R.D., and Raulston, J.R. (1972), ‘Competitiveness and Behaviour of Tobacco Budworms Sterilized with Reserpine and Gamma Irradiation’, Journal of Economic Entomology, 65, 966969. Hayes Jr, W.J. (1968), ‘Toxicological Aspects of Chemosterilants’,inPrinciples of Insect Chemosterilization, eds. G.C. Labecque and C.N. Smith, New York: Appleton-Century- Crofts, pp. 315347. International Atomic Energy Agency (IAEA) (1996a), Manual on Self-Contained Gamma Irradiators (Categories I and III), IAEA-PRSM-7, Vienna, Austria: Author. International Atomic Energy Agency (IAEA) (1996b), Manual on Panoramic Gamma Irradiators (Categories II and IV), IAEA-PRSM-8, Vienna, Austria: Author. International Atomic Energy Agency (IAEA) (1996c), International Basic Safety Standards for Protection against Ionizing Radiation and for the Safety of Radiation Sources, Safety Series No. 115, Vienna Austria: Jointly published by FAO, IAEA, ILO, OECD/NEA, PAHA and WHO. International Atomic Energy Agency (IAEA) (2002a), Dosimetry for Food Irradiation, TRS No. 409. Vienna, Austria: Author. International Atomic Energy Agency (IAEA) (2002b), Natural and Induced Radioactivity in Food, IAEA-TECDOC-1287, Vienna, Austria: Author. International Atomic Energy Agency (IAEA) (2004), Dosimetry System for SIT: Standard Operating Procedure for Gafchromic Film. Vienna, Austria: Author. (http://www.iaea.org/ Downloaded By: [Hendrichs, Jorge] At: 16:02 6 November 2009 programmes/nafa/d4/public/d4_pbl_5_2.html). International Commission on Radiation Units and Measurements (ICRU) (1998), Funda- mental Quantities and Units for Ionizing Radiation, ICRU Report 60, Bethesda, MD, USA: Author, p. 24. Klassen, W., and Matsumura, F. (1966), ‘Resistance to a Chemosterillant, Metepa, in Aedes aegypti, Mosquitoes’, Nature, 209, 11551156. Knipling, E.F. (1979), The Basic Principles of Insect Population Suppression and Management, Agriculture Handbook No. 512. SEA. Washington, DC, USA: USDA, p. 659. Lide, D.R. (ed.), (1990), CRC Handbook of Physics and Chemistry, Boca Raton, FL: CRC Press. McLaughlin, W.L., Boyd, A.W., Chadwick, K.H., McDonald, J.C., and Miller, A. (1989), Dosimetry for Radiation Processing, New York: Taylor & Francis, p. 251. Mehta, K., Kojima, T., and Sunaga, H. (2003), ‘Applicability Study on Existing Dosimetry Systems to High-Power Bremsstrahlung Irradiation’, Radiation Physics Chemistry, 68, 959 962. Moursy, L.E., Eesa, N.M., Cutkomp, L.K., and Subramanyam, B.H.E. (1988), ‘Effects of Gamma Radiation and a Chemosterilant Applied to Late Nymphal Instars of the Large Milkweed Bug (Hemiptera: Lygaeidae)’, Journal of Economic Entomology, 81, 459462. 362 K. Mehta

Parker, A., and Mehta, K. (2006), ‘Sterile Insect Technique: A Model for Dose Optimization for Improved Sterile Insect Quality’, Florida Entomologist, 89, 8895. Unterweger, M.P., Hoppes, D.D., and Schima, F.J. (1992), ‘New and Revised Half-Life Measurement Results’, Nuclear Instrument Measurements, A312, 349352. Watters, F.L. (1979), ‘Potential of Accelerated Electrons for Insect Control in Stored Grain’,in Proceedings of the Second International Working Conference on Stored-Product Entomology, pp. 278286. Downloaded By: [Hendrichs, Jorge] At: 16:02 6 November 2009