Hoy: Myths of Managing Resistance 385

INTRODUCTION TO THE SYMPOSIUM, THE MYTHS OF MANAGING RESISTANCE

MARJORIE A. HOY Entomology & Nematology Department, University of Florida, Gainesville, FL 32611-0620

Resistance to pesticides in arthropod pests is a serious and increasingly perplexing problem in Florida, the USA, and the world. Resistance to pesticides already has created significant economic, ecological, and public health problems in agricultural, household and garden, and medical/veterinary pest management programs. Exten- sive research has been conducted on diverse aspects of pesticide resistance, and we have learned much during the past 40 years. However, to some degree, much of the discussion about ‘resistance management’ has been based on ‘myths’. As an organizer of this symposium, one of my goals was to stress that managing resistance is a formi- dable task that will remain a perpetual pest management dilemma, because resistance is a fundamental survival response to stress by arthropods. Five papers were presented in this symposium at the 1994 annual meeting of the Florida Entomological Society, but one manuscript regarding the response by industry to resistance could not be published in this series. In the first paper, Gary Leibee and John Capinera assess the impact of resistance to pesticides in Florida and cite examples of resistance that limit pest management options. Julie Scott describes what we currently know about the molecular genetics of arthropod resistance to pesticides. The number of genes identified, and the diversity of their effects on the physiology of arthropods, verify that resistance is a normal response to diverse environmental stresses. ‘Pesticide resistance’ is part of a general stress response with a long evolutionary history. Leah Bauer describes what we know about resistance to various toxins of Bacillus thuringiensis (B.t.) strains. B.t. provides microbial control of an increasingly diverse group of arthropod species and is an increasingly important tool for integrated pest management programs. The deployment of transgenic crop plants containing B.t. toxin genes is likely to be an effective method for inducing resistance in agricultural pests. Despite the diversity of B.t. toxin genes isolated and cloned, cross resistances are common. Thus, B.t. toxin genes are limited resources. Finally, I discuss a variety of resistance management methods and point out that we cannot really avoid resistance—we can only delay its onset. I argue that resistance management needs a paradigm shift that can best be accomplished if we recognize that pest management must be changed from a single-tactic strategy to a multi-tactic mode. Delaying resistance, whether to traditional pesticides or to transgenic plants with toxin genes, will require that we develop truly integrated pest management programs, incorporating all appropriate tactics, including host plant resistance, cultural controls, biological controls, genetic controls, and biorational controls. Pesticides should be reserved for situations in which they perform best—as tools to resolve an unexpected pest population outbreak. Effective, fully-integrated IPM programs will delay resistance because the number and rates of pesticide applications can be reduced.

386 Florida Entomologist 78(3) September, 1995

PESTICIDE RESISTANCE IN FLORIDA INSECTS LIMITS MANAGEMENT OPTIONS

GARY L. LEIBEE1 AND JOHN L. CAPINERA2 1Department of Entomology and Nematology, University of Florida, Central Florida Research and Education Center, Sanford, FL 32771

2Department of Entomology and Nematology, University of Florida, Gainesville FL 32611

ABSTRACT

Pesticide resistance in Florida was characterized through a survey and literature review. The survey was conducted in 1994 among public-sector entomologists to determine the current and future status, extent, context, pattern, and instances of pesticide (insecticide and acaricide) resistance in Florida. Results attested to the impact of pesticide resistance on the management of numerous arthropods in Florida. Twenty-ﬁve examples of insecticide and acaricide resistance were cited by survey respondents in agricultural, ornamental and landscape, medical and veterinary, or household and structural pests. It remains possible to manage most arthropods by using chemical pesticides, but the current and anticipated lack of efﬁcacious materials threatens current practices in some areas. Trends in extent, context, or patterns of resistance were noted as follows: high value crops, frequently treated arthropods, smaller arthropods, and pyrethroids were all considered factors associated with resistance. Insecticide resistance and its management were reviewed in depth for the leafminer Liriomyza trifolii and the diamondback moth, Plutella xylostella, two major insect pests in Florida for which management options have become severely limited because of insecticide resistance. Both cultural practices (continuous cropping, isolation, transport of infested seedlings) and pesticide use patterns (frequent application of broad spectrum pesticides) contributed to L. trifolii and P. xylostella resistance development. The history of pesticide resistance in these two insects is probably typical of pest resistance in Florida and may portend similar future problems unless depen- dency on pesticides for pest suppression is reduced through adoption of IPM philosophy and practices.

Key Words: Insecticide resistance, Liriomyza trifolii, Plutella xylostella.

RESUMEN La resistencia a los pesticidas en la Florida fue caracterizada a través de una encuesta y una revisión de la literatura. La encuesta fue conducida en 1994 entre los en- tomólogos del sector público para determinar el estado presente y futuro, extensión, contexto, patrón e instancias de la resistencia a pesticidas (insecticidas y acaricidas) en la Florida. Veinte y cinco ejemplos de resistencia a insecticidas y acaricidas fueron citados por los que respondieron la encuesta sobre plagas agrícolas, de ornamentales y de jardines, de importancia médica y veterinaria, o domésticas y de otras estructuras. Parece posible manejar la mayoría de los artrópodos usando pesticidas químicos, pero la falta actual y anticipada de materiales amenaza las prácticas presentes en al- gunas áreas. La tendencia en la extensión, contexto, o patrones de resistencia fue como sigue: cultivos de alto valor, artrópodos frecuentemente tratados, pequeños ar- trópodos, y piretroides fueron todos considerados como factores asociados con la resistencia.

Leibee and Capinera: Symposium on Pesticide Resistance 387

La resistenca a los insecticidas y su manejo fueron revisados en profundidad para el minador de las hojas Liriomyza trifolii y para la polilla de la col, Plutella xylostella, los insectos plagas principales en la Florida para los cuales las opciones de manejo se han tornado severamente limitadas debido a la resisitencia a los insecticidas. Tanto las prácticas culturales (cosecha continua, aislamiento, transporte de plántulas infestadas) como los patrones de uso de pesticidas (aplicación frecuente de insecticidas de amplio espectro) contribuyeron al desarrollo de la resistencia de L. trifolii y P. xylostella. La historia de la resistencia a los pesticidas en estos dos insectos es probablemente típica para la resistencia de las plagas en la Florida, y podría signiﬁcar problemas futuros similares a menos que la dependencia de los pesticidas para la su- presión de las plagas sea reducida a través de la adopción de ﬁlosofía y prácticas de MIP.

Insecticide resistance has had an impact on the management of insect pests in Florida since the mid-1940s following the widespread adoption of synthetic insecticides, especially the organochlorines, organophosphates, and pyrethroids. Numerous anecdotal reports exist, wherein consistently effective insecticides have become ineffective and remained so for several seasons. Such reports have been considered ample evidence of resistance development (Hoskins & Gordon 1956). In fact, Genung (1957) provided strong evidence based on anecdotal reports and data from field efficacy trials for resistance development in the cabbage looper, Trichoplusia ni Hubner, imported cabbageworm, Artogeia rapae (L.), a Liriomyza sp., and leafhoppers, Empoasca sp., at a session of the Florida State Horticultural Society Meeting in 1957 entitled “Sympo- sium-Vegetable Insect Resistance to Insecticides in Florida” (Brogdon 1957). Resis- tance episodes in Florida have also been documented in a number of species with laboratory studies in which concentration-mortality response has been used to compare resistant and susceptible strains. Much of this work has been conducted within the last 10 years and involves species such as cabbage looper (Shelton & Soderlund 1983), diamondback moth, Plutella xylostella (L.), (Leibee & Savage 1992a, Shelton et al. 1993, Yu & Nguyen 1992), silverleaf whitefly, Bemisia argentifolii Bellows & Per- ring, (G. L. L. unpublished data), house fly, Musca domestica L., (Bailey et al. 1970, Bloomcamp et al. 1987), German cockroach, Blatella germanica (L.), (Milio et al. 1987, Koehler 1991, Hostetler & Brenner 1994), Liriomyza trifolii Burgess (Keil & Parella 1990, G. L. L. unpublished data), fall armyworm, Spodoptera frugiperda (J. E. Smith), (Pitre 1988, Yu 1992), cat flea, Ctenocephalides felis (Bouche), (El-Gazzar et al. 1986), and citrus rust mite, Phyllocoptruta oleivora (Ashmead), (Omoto et al. 1994). In hopes of providing a better understanding of current pesticide resistance and its consequences in Florida, we report here the results of a recent survey of public-sector entomologists conducted to assess the extent of pesticide resistance in Florida, and its current and potential impacts. In addition, we provide an in-depth account of two important insect pests of vegetables in Florida, the dipterous leafminer L. trifolii and the diamondback moth, for which management options have become extremely limited because of insecticide resistance.

RESISTANCE SURVEY During the spring of 1994, 16 public-sector entomologists were sent survey forms to measure their opinion about the extent of pesticide (deﬁned as insecticide and acaricide) resistance in Florida, and its current and potential impact. We polled Univer-

388 Florida Entomologist 78(3) September, 1995 sity and USDA entomologists from various backgrounds, representing the ﬁelds of agricultural, ornamental and landscape, medical and veterinary, or household and structural pest management. Entomologists with considerable ﬁeld experience, and a close relationship with producers or pest control professionals, were favored. We received responses from 14 of those surveyed, and 12 respondents provided useful information. Additional information was sought from other knowledgeable individuals to round out the survey. The questions and responses were as follows:

The Current and Future Status of Pesticide Resistance Respondents were asked to indicate if resistance was: not a problem, a minor problem, a significant problem, or a critical problem. Only a single response was requested. The time frame for future problems was specified as 10 years in the future. The respondents differed in their assessment of the severity of the resistance problem depending on the crop or environment being considered. Resistance was viewed to be a critical problem in greenhouses (foliage plants, flowering plants, and some woody ornamentals), floriculture (both greenhouse and field-grown flowers), and animal production (penned and free-ranging). Ornamental plants have long been considered to be extremely sensitive to damage, hence they are treated frequently and prone to insecticide resistance problems. Resistance in animal production is a more recent phenomenon, however, apparently resulting from widespread use of insecticide im- pregnated ear tags. Resistance was considered to be a significant, but not critical, problem in vegetable crops, some field crops, and households. This might be viewed as surprising, because many vegetable crops, some field crops, and households in Florida receive insecticide treatments at frequencies similar to the aforementioned situations where pesticide resistance was judged to be critical. It is likely that the severity of the problem is due as much to corporate marketing strategies as to pesticide use patterns. Specifically, the pesticide market is smaller for greenhouse, floriculture and animal uses, so pesticide companies support fewer registrations. Therefore, when pesticide failures occur, there are few options, or in some cases none. This, of course, results in a critical situation. For medical pests, which in Florida is principally mosquitoes, the significance of the resistance problem apparently is related to location. Resistance was reported to be a significant problem in coastal locations, but only a minor problem in other areas. Coastal regions not only are extremely favorable for mosquito breeding, but a high proportion of the state’s population (79%) dwells along the coast, so there is frequent need for chemical suppression. Landscape plants seem to be relatively free of resistance problems. Woody ornamentals are not usually planted in large single-species stands, which may help them to avoid development of high pest populations. Such landscape plants often tolerate considerable defoliation or pest density without obvious symptoms, so chemical treatment is not a regular feature of landscape maintenance. Also, in recent years there has been a concerted effort to introduce native, hardy, pest-resistant plants into the landscape, reducing the need for insecticide treatment. Among landscape plants, perhaps only turfgrass is treated regularly, and the southern chinch bug, Blissus insu- laris Barber, exhibits some degree of resistance, particularly in southern Florida. Nursery production of landscape ornamentals is also an exception, and mites can present resistance problems in this environment. Although the number of pests displaying resistance to pesticides has increased markedly in the last two decades, respondents generally did not see the resistance problem worsening greatly in the next 10 years. The only exception was the area of

Leibee and Capinera: Symposium on Pesticide Resistance 389 household pest management, where the situation is anticipated to become critical. This generally optimistic attitude likely reﬂects faith in the agrichemical industry, which has continued to introduce novel pesticide chemistry or biorational materials that allow producers to continue with traditional agriculture and pest control practices despite increasing numbers of pests that have become somewhat resistant to one or more pesticides. The scientiﬁc community has also responded quickly and effectively to the onset of resistance by identifying alternative pest control chemicals and by helping to integrate other types of pest suppression into traditional production systems.

The Extent of Resistance Respondents were asked to indicate whether resistance applied to: a few compounds, numerous compounds, a few pests, or numerous pests. Up to two responses were possible. Respondents were also asked to designate how many pests or compounds were affected and to indicate either a speciﬁc number or range. The extent of the resistance problem was reported to be variable, depending on whether the focus was the number of pests or pesticides. Resistance was generally reported to be limited to few pests in each commodity or environment. The number of resistant species was generally reported to be 3-5 per respondent, with a range given as 1-10 per respondent. Although the number of species was small, the number of chemical compounds to which the pests were reported resistant was considerably larger. Respondents generally indicated that pests exhibiting resistance were resistant to 5-10 compounds. The range in the number of compounds was given as from 1 product to all those on the market.

The Context of Pesticide Resistance Respondents were asked to indicate if particular pests, crops, or environments existed in which resistance occurred more frequently. Respondents most frequently indicated that high value, damage-sensitive crops were prone to have pesticide resistance problems. They cited greenhouse, ﬂoricul- tural, and vegetable crops as examples. The environments next most frequently cited as having resistance problems were those in which frequent or routine pesticide applications were made. Of course this corresponds to the aforementioned high value crops, but there are also situations in which value and damage sensitivity are not a major issue; examples are households, livestock, and certain ﬁeld crops. Only infrequently were the biological characteristics of the pests cited as favoring frequent occurrence of resistance. Pests with short generation times and high intrinsic rates of increase were suggested to be more prone to display resistance.

The Pattern of Pesticide Resistance

Respondents were asked if there were any patterns evident wherein entire classes of pesticide compounds or groups of arthropods displayed a tendency toward increased frequency of resistance, or whether resistance applied only to specific materials or pests. Patterns of pesticide resistance related to chemical or biological taxon were not especially evident to our respondents. Many said that pesticide resistance was species-specific, that biological taxon was not a very good predictor of resistance problems. A few, however, suggested that whiteflies, thrips, and especially mites were re-

390 Florida Entomologist 78(3) September, 1995 sistance prone. Similarly, although cross resistance within chemical classes was acknowledged, respondents indicated that they generally considered each pesticide to have unique chemical properties, so that development of resistance was difﬁcult to predict based on chemical taxon. The exception to this generalization seems to be the pyrethroids, where there is general acknowledgment that resistance is likely to develop.

Instances of Pesticide Resistance

Respondents were asked to name specific instances of pesticide resistance, including the pesticide, pest, and approximate date, and also to indicate whether the information on resistance was “documented” or anecdotal. Instances of pesticide resistance in Florida provided by respondents are shown in Table 1. Surely this is not a complete list, either of pests or problem pesticides, but serves to demonstrate adequately the diversity of arthropod taxa affected. Also, arthropods found in numerous environments or crop systems are affected, and some his- torical trends are evident. Respondents acknowledged that only about one-half of the purported cases of resistance are “documented,” with the remainder based on anecdotal information. However, we carefully selected experienced entomologists and asked them to respond only in their area of expertise. Thus, we are confident that instances of misapplication and other potential sources of erroneous reports of resistance are not included. Because some of the “documented” resistance is from industry sources and not accessible to us, we have not included this specific information. Note also that this table does not include information on the leafminer L. trifolii and the diamondback moth, two insects with well-documented histories of insecticide resistance in Florida. A review of insecticide resistance in these two troublesome insects follows.

INSECTICIDE RESISTANCE IN L. TRIFOLII

Past and Present Situation

Prior to 1945, leafminer problems on celery and other vegetables in Florida were apparently almost nonexistent. Control consisted mainly of clean-up measures and application of nicotine sulphate (Wolfenbarger 1947). Wolfenbarger (1947) recommended chlordane for control of leafminer on potatoes in south Florida. Harris (1962) reported that dimethoate, which was not labeled for celery, and diazinon and naled which were labeled, could control leafminer on celery in 1962. Genung et al. (1979) reported that with the use of diazinon, naled, and azinphos-methyl, the mortality of vegetable seedlings and yield reductions declined and leafminer populations remained low until 1974, when they began to heavily infest celery and tomato. Genung et al. (1979) reported that in 1974 growers could not control leafminers on celery with diazinon, naled, or azinphos-methyl and that dimethoate, which was approved for use on celery the same year, also did not give the desired level of control. Poe & Strandberg (1979) reported that oxamyl, which was approved for use on celery in 1975, was effective for about two years. They also reported that in 1976 and 1977 leafminer on celery was uncontrollable in Florida by any insecticide labeled for use. Florida growers acquired the use of methamidophos in 1977 and permethrin in 1978 for the control of leafminer on celery. Permethrin became ineffective for leafminer control on celery in less than two years. Methamidophos was then considered the only insecticide that gave any amount of control in celery in Florida, and it was considered marginally effective. The possibility of effective chemical control did not come until the spring of

Leibee and Capinera: Symposium on Pesticide Resistance 391

TABLE 1. EXAMPLES OF INSECTICIDE AND ACARICIDE RESISTANCE IN FLORIDA CITED BY RESPONDENTS IN A 1994 SURVEY OF PUBLIC SECTOR ENTOMOLOGISTS.

Date Arthropod (decade) Pesticide

House fly 1940 DDT Musca domestica (L.) 1950 chlordane, dieldrin, lindane, malathion 1970 dimethoate, ronnel, tetrachlorvinphos 1980 cyromazine, methomyl, various pyrethroids German cockroach 1950 chlordane, dieldrin, lindane Blatella germanica (L.) 1960 allethrin, diazinon, malathion 1970 carbaryl, propoxur 1980 cyfluthrin, cypermethrin Cat flea 1950-70 diazinon, malathion Ctenocephalides felis 1970-80 bendiocarb, carbaryl, propoxur (Bouché) 1980-90 cyfluthrin, fenvalerate, permethrin Horn fly 1980 fenthion, fenvalerate, flucythrinate, Haematobia irritans permethrin, stirophos (L.) Salt marsh mosquito 1950 DDT Culex nigripalpus 1960 malathion Theobald 1990 methoprene Soybean looper 1970 acephate, methomyl, various pyrethroids Pseudoplusia includens (Walker) Fall armyworm 1970-80 malathion, carbaryl, methyl parathion, Spodoptera frugiperda diazinon, trichlorfon, fluvalinate, bifenthrin, (J. E. Smith) tralomethrin Southern green stinkbug 1970 carbaryl, methomyl Nezara viridula (L.) 1980 endosulfan Tobacco budworm 1970 ethyl parathion Heliothis virescens (Fabricius) Corn earworm 1950-60 malathion, diazinon, Helicoverpa zea (L.) 1960-70 ethyl parathion, carbaryl Pepper weevil 1990 fenvalerate, permethrin, oxamyl Anthonomus eugenii Cano Beet armyworm 1980 chlorpyrifos, methomyl Spodoptera exigua (Hübner) Tomato pinworm 1970 carbaryl Keiferia 1980 methomyl, fenvalerate lycopersicella 1990 oxamyl (Walsingham) Western flower thrips 1980 pyrethroids Frankliniella occidentalis (Pergande) Mole crickets 1970 chlordane Scapteriscus spp.

392 Florida Entomologist 78(3) September, 1995

TABLE 1. (CONT.) EXAMPLES OF INSECTICIDE AND ACARICIDE RESISTANCE IN FLORIDA CITED BY RESPONDENTS IN A 1994 SURVEY OF PUBLIC SECTOR ENTOMOLOGISTS.

Date Arthropod (decade) Pesticide

Green peach aphid 1970-80 malathion, diazinon, oxydemeton-methyl, Myzus persicae (Sulzer) dimethoate Cabbage looper 1950-60 DDT, toxaphene, parathion Trichoplusia ni 1960-70 endrin, mevinphos, naled (Hübner) 1970-80 methomyl Cowpea curculio 1980 endosulfan Chalcodermus aeneus Boheman Citrus rust mite 1990 dicofol Phyllocoptruta oleivora (Ashmead) Yellow pecan aphids 1980 various pyrethroids Monellia caryella (Fitch), Monelliopis pecanis Bissell Silverleaf whiteﬂy 1990 bifenthrin, fenvalerate, permethrin, Bemisia argentifolii endosulfan Bellows & Perring Two-spotted spider mite 1980 fenbutatin-oxide Tetranychus urticae 1990 avermectin Koch Melon aphid 1990 acephate Aphis gossypii Glover

1982 when the celery industry secured the use of cyromazine. With the use of cyromazine, leafminer problems were considered under control until late 1989 when an unusual lack of efficacy occurred in the Everglades area. Laboratory studies confirmed the presence of a high level of cyromazine resistance in a suspect strain of L. trifolii (G. L. L., unpublished data). Larval mortality in the resistant strain at 300 ppm of cyromazine, the highest label concentration used in the field, was low enough to explain the loss of efficacy. The cyromazine resistance was expressed as an incompletely recessive trait and not sex-linked. Backcrossing suggested that the resistance was conferred by a major gene. The resistance was considered un- stable since sensitivity returned in the resistant strain (from an LC50 of about 440 ppm to an LC50 of about 85 ppm) within 5 generations of laboratory rearing without selection. This was consistent with a survey of leafminer populations that indicated susceptibility to cyromazine had returned during the summer of 1990 (J. S. Ferguson, unpublished data). This reversion was probably due to the immigration of susceptible individuals during what is traditionally a period of little or no celery production and very little use of cyromazine. The cyromazine-resistant strain was not resistant to abamectin (G. L. L., unpublished data), the only logical alternative insecticide available for control of leafminer in celery. This information contributed to the granting of a crisis exemption (Section 18, FIFRA) in early 1990 for the use of abamectin in celery to control leafminer. Fur- ther efforts of the Florida Fruit and Vegetable Association, celery growers, CIBA, Merck Research Laboratories, and the University of Florida resulted in the subsequent granting by the EPA (Section 18, FIFRA) in October 1990 of a specific exemp- Leibee and Capinera: Symposium on Pesticide Resistance 393 tion for the use of abamectin in celery. Since then, abamectin has been used in celery under specific exemptions. These specific exemptions are unique in that, in order to discourage the onset of resistance to abamectin, only two consecutive applications are allowed, forcing rotation with another insecticide. However, no other effective insecticide was available for rotation except for cyromazine which, due to reversion, had become efficacious again. Since cyromazine and abamectin have different modes of action and no cross resistance was indicated, cyromazine was included in the leafminer control program under well-defined resistance management guidelines.

Management of Cyromazine Resistance in L. trifolii

A program for managing cyromazine resistance in L. trifolii was presented to celery growers. The goal of this program was to control L. trifolii while increasing and preserving susceptibility to cyromazine and minimizing the possibility of selecting for resistance to abamectin. Cyromazine use patterns and celery culture were suggested that would reduce selection of resistant phenotypes and encourage the immigration of feral, hopefully susceptible, leafminers into resistant populations. Recommendations included: using noninfested transplants; initiating the spray program based on a threshold to reduce the number of insecticide applications; starting with abamectin to maximize early control; rotating two sprays of abamectin with two sprays of cyromazine to avoid excessive use of one insecticide; finishing a planting with two applications of abamectin to reduce the number of adults emerging from the soil and the trash after harvest; disking in trash as soon as possible to remove this source of leafminers; and not using pyrethroids, such as permethrin and esfenvaler- ate, to minimize adverse effects on parasites and predators. In addition, since acreage is very low in the production fields as harvesting ends (June) and the transplanting begins (September), it was recommended to not use cyromazine during the summer (June through September) to prevent the continued selection of isolated populations and to encourage the immigration of susceptible individuals when celery acreage is at its lowest. Except for seedling production, July and August are otherwise free of celery. Not using cyromazine at all in seedling beds was recommended, since transplanting from infested seedling beds is considered an important mechanism for transferring resistant leafminers to the production fields. Lastly, seedlings were recommended to be grown distant from the field production areas to reduce the chances of infestation by resistant leafminers.

INSECTICIDE RESISTANCE IN THE DIAMONDBACK MOTH

Past and Present Situation

Historically, the diamondback moth was considered a minor pest, usually included in a complex of cabbage caterpillars along with the cabbage looper and the imported cabbageworm, Artogeia rapae (L.), but of much less importance. Control recommendations for the diamondback moth generally have been the same as for the other cabbage caterpillars (Sanderson 1921, Metcalf & Flint 1939, Watson & Tissot 1942, Metcalf et al. 1951, 1962). Prior to the mid-1940s, insecticides used for cabbage caterpillar control included nicotine, arsenicals, pyrethrum, rotenone, kerosene, and hot water (150°F), and from the mid-1940s through the 1970s included DDT, toxaphene, parathion, methoxychlor, mevinphos, endosulfan, naled, methomyl, and methamidophos. Bacillus thuringiensis was also available, but was not used extensively due to expense and the perception of less than desirable control. In the early 1980s growers 394 Florida Entomologist 78(3) September, 1995 switched to the newly available and extremely effective pyrethroids, permethrin and fenvalerate for control of the cabbage looper and diamondback moth, both of which had become difficult to control with the other insecticides. Insecticide resistance had long been suspected as the cause of the poor cabbage looper control (Genung 1957, Workman & Greene 1970), and Shelton & Soderlund (1983) showed that a population from Florida was one of the most resistant to methomyl in the eastern U.S. The poor control of diamondback moth has been attributed to the destruction of parasites by excessive use of insecticides, such as methomyl, which were applied for cabbage looper suppression, but which were relatively ineffective on diamondback moth. However, the poor control of diamondback moth may have actually been the earliest indications of resistance problems. Permethrin and fenvalerate proved to be very effective for control of all cabbage insects until the mid-1980s when growers observed that these insecticides were no longer providing effective control of the diamondback moth. University trials reflected the same lack of control with fenvalerate (Leibee 1986) and from the winter of 1986- 87 to the present, pyrethroid insecticides provided poor control of diamondback moth at Sanford, FL (G. L. L., unpublished data). Magaro & Edelson (1990) noted that failures to control diamondback moth in south Texas were first reported by cabbage producers in the spring of 1987. Leibee & Savage (1992b) reported a high level of resistance to fenvalerate in a laboratory strain of diamondback moth collected in central Florida in 1987. Loss of efficacy with pyrethroids for control of diamondback moth caused growers to switch to intensive use of several organophosphates, endosulfan, and B. thuringiensis subspecies kurstaki (Btk), all of which did not provide the level of control provided by pyrethroids prior to resistance. At present, many diamondback moth populations have become very difficult to control with any of the registered synthetic insecticides and Btk. The presence of Btk resistance in Florida was immediately suspected because Btk resistance in diamondback moth had been reported in Hawaii (Tabashnik et al. 1990), Japan (Tanaka & Kimura 1991), and Malaysia (Syed 1992); it was eventually confirmed for Florida (Leibee & Savage 1992a, Shelton et al. 1993). With the presence of Btk-resistance, there were essentially no effective insecticides available for control of many diamondback moth populations in Florida until the recent introduction of B. thuringiensis subspecies aizawai (Bta)-based insecticides. Bta-based insecticides (those possessing the Cry1C toxin) are being successfully used in areas where Btk- based insecticides have failed. Lack of resistance to Bta in diamondback moth resistant to Btk has been documented in Japan (Hama et al. 1992), Malaysia (Syed 1992), and Florida (Leibee & Savage 1992a, Shelton et al. 1993). Diamondback moth abundance has been considered low for several seasons in central Florida (G. L. L., personal observation). This is due in part to the return of substantial amounts of natural control from parasites, which in turn is attributed to reduced pyrethroid use. Growers are not spraying as frequently for diamondback moth and are able to use Btk-based insecticides, suggesting a return of susceptibility to Btk.

Management of Resistance in the Diamondback Moth Crop culture and control recommendations were made that would reduce the selection of resistant phenotypes of diamondback moth and encourage the immigration of feral, hopefully susceptible, individuals into resistant populations. These recommendations were based on the following knowledge. Susceptibility to Btk had been greatly reduced in some populations. Bacillus thuringiensis resistance in diamond- Leibee and Capinera: Symposium on Pesticide Resistance 395 back moth in Hawaii was shown to be inherited as a recessive trait (Tabashnik et al. 1992) and observations from field and laboratory studies in Florida suggested the same (G. L. L., unpublished data). Bacillus thuringiensis subspecies aizawai-based products (those possessing the Cry1C toxin) appeared to be effective in populations where Btk susceptibility was reduced. Tank-mixing Bt with mevinphos was shown to be quite effective at reducing infestations in early season (G. L. L. unpublished data); however, the use of mevinphos was to be discontinued in 1995, eliminating the most effective insecticide other than Bt for diamondback moth control on cabbage in Flor- ida. Use of pyrethroids and carbamates can select for resistance that might further reduce the efficacy of organophosphate insecticides and endosulfan, and also destroy the parasites and predators providing natural control of the diamondback moth. Crop culture recommendations included: not growing cabbage in the warmest months (May through September in central Florida) when insect pressure is the highest and Bt-based insecticides are the least efficacious; immediately disposing of crop residues to prevent migration from heavily selected populations into new plantings and seedling production areas; and using noninfested transplants, which not only contributes to control but also reduces spread of diamondback moths to new locations. Diamondback moths that infest purchased transplants may be highly resistant due to heavy usage of insecticides on the transplants or in fields near transplant production areas. Producers growing their own transplants are at an advantage because they have more control over infestation levels and have specific knowledge about the resistance problems in their production areas. Control recommendations included: inspecting crops frequently (about twice per wk) to determine the presence of the pest of concern or unexpected pests; beginning inspections at the seedling stage because reducing infestations in early season appears to be critical to managing diamondback moth; minimizing insecticide applications whenever possible by using action thresholds developed through research or by intuition; and using pheromone traps to monitor the presence or absence of diamondback moth before and during the growing season, and also for monitoring peaks of adult activity (Baker et al. 1982) for timing insecticide applications. Specific insecticide recommendations included: using Btk and Bta as the principle insecticides for control of diamondback moth; if the population was known to be susceptible to Btk, alternating a Bta-based product with a Btk-based product to avoid re- petitive applications of the same insecticide to reduce the selection of resistance to any one product; using only Bta to insure maximum control if Btk resistance was known to be present or the status of Btk susceptibility was unknown; applying Bt twice weekly and tank-mixing with mevinphos weekly, starting with the tank-mix to maximize the control that is critical early in crop; including endosulfan, chlorpyrifos, and methamidophos as alternatives or substitutes for mevinphos in the tank-mixes with Bt, especially endosulfan, since it belongs to a different chemical class than mevinphos; avoiding the use of carbamates; and, not using pyrethroids.

OBSERVATIONS ON INSECTICIDE RESISTANCE IN L. TRIFOLII AND DIAMONDBACK MOTH Probably the greatest factor contributing to the development of insecticide resistance in L. trifolii and diamondback moth was long term and frequent use of single insecticides or classes of insecticides. However, celery and cabbage cultural practices in Florida probably contribute to the rapidity and degree with which insecticide resistance develops in these two insects. Both crops are grown in relatively small and isolated areas, or pockets, of agricultural activity. In addition, both crops are grown in some form year round. This isolation and the lack of a substantial crop-free period result in the containment and “cycling” of resistant populations. This results in the 396 Florida Entomologist 78(3) September, 1995 same population being exposed to insecticides continually. In addition, the constant use of insecticides removes susceptible individuals that may be immigrating into the population, thus preventing the opportunity for reversion. These factors are believed to be especially evident with the development of high levels of resistance in the diamondback moth in the 1980s. Several changes in the production of cabbage in the 1980s contributed greatly to the “cycling” of resistant populations in the production areas and the movement of resistant populations between production areas within and outside Florida. Among these changes was the lengthen- ing of the crucifer production period by harvesting later in the spring and transplanting earlier in the summer. Prior to this situation, much of the summer (June-September) in central Florida was basically a crucifer production-free period and diamondback moth populations were very low. Crucifer production was thrust into the warmer, drier parts of the year when diamondback moth became more of a problem, resulting in increased use of insecticides and subsequent exposure of additional generations to more selection pressure. With development of the container grown transplant industry and the ability to grow transplants in the summer, a situation was eventually created in which crucifers were continually produced throughout the year in the same localities, either as transplants or field crops, or both. Insecticide resistant diamondback moth could move from the fields to transplants in the summer, and be redistributed back to the fields in late summer and early fall. Therefore, heavy populations of resistant diamondback moth were being perpetuated locally throughout the year and continually exposed to insecticides. Opportunities for the return of susceptibility by the immigration of individuals with susceptible phenotypes was essentially eliminated. Transplants probably were a major factor in the development of resistance problems in diamondback moth on a national level during the 1980s; container-grown transplants were popular and a healthy containerized transplant industry developed in the south. Much of the transplant production in the south in the winter and spring supplied more northern growers with transplants to establish stands in the spring when environmental conditions were not conducive to direct seeding. Transplants were also produced during summer months in the north to facilitate establishment of stands in the south when soil temperatures are too high for germination. Transplant growers likely were oblivious to the fact that they were shipping resistant diamondback moth. They were probably controlling the later larval stages that cause obvious damage and shipping what appeared to be uninfested plants. Adults flying into the open-sided greenhouses from the field could maintain a supply of eggs and early mining instars before shipment. In addition, larvae have been observed to be deep inside the bud of the transplant, out of reach of the insecticide and the eye of the grower. It is possible that avoidance of the insecticide deposits drives diamondback moth larvae emerging from the leafmine into the bud. Another aspect of the transplant industry that could have contributed to the development of resistance was that only a few large growers in areas of high levels of resistance produced most of the transplants used in Florida and the rest of the U.S. This greatly increased the probability of many growers receiving transplants from areas with resistant diamondback moth populations.

CONCLUSION The results of the survey attest to the impact of insecticide resistance on past and present pest management in Florida. The review of resistance problems with L. trifolii and diamondback moth illustrates how insecticide resistance can complicate the management of pestiferous arthropods. Insecticide resistance seems to be pervasive Leibee and Capinera: Symposium on Pesticide Resistance 397 in Florida, suggesting that we are not adequately considering the consequences of the way we use pesticides. Resistance management strategies should be integrated with nonchemical control whenever possible. In the case of the aforementioned L. trifolii and diamondback moth problems, crop management (isolation and continuous cropping) were key factors in development of resistance. In most cases, pesticides are the most efficient, easiest, and cheapest methods of controlling pestiferous insects and mites. The number of pesticides available is dwin- dling rapidly, however, due to cancellation of registrations and lack of re-registration. As a result, the conservation of arthropod susceptibility to the remaining pesticides, and to newly developed pesticides, is becoming extremely important. Unfortunately, we have not made adequate effort to conserve susceptibility of arthropods to pesticides. The minimum effort should include development of baseline data that would allow the investigation of potential resistance episodes in a timely manner. Conserving pesticide susceptibility takes the development of knowledge and a commitment from those responsible for producing, using, and conducting research on pesticides. Thus, educational efforts should be enhanced. Something everyone can do to help manage resistance to any pesticide is to refine its use. Application of the correct amount and type of insecticide in a timely and efficient manner would help forestall the onset of resistance, particularly if nonchemical techniques could be used to reduce the numbers and frequency of application. This is best accomplished following research on, and implementation of, IPM strategies. In- creasingly, pesticide resistance management must be considered an important component of IPM.

ACKNOWLEDGMENTS

We gratefully acknowledge the assistance of the following individuals who gave generously of their time to make the survey component of this paper possible: Joe Funderburk, Jerry Hogsette, Freddie Johnson, Joe Knapp, Phil Koehler, Russ Mizell, Charlie Morris, Lance Osborne, Jim Price, Dakshina Seal, Dave Schuster, Phil Stansly, and Simon Yu. We are also grateful to Brett Highland for his critical review of the manuscript. Florida Agricultural Experiment Stations Journal Series No. R- 04551.

REFERENCES CITED

BAILEY, D. L., G. C. LABRECQUE, AND T. A. WHITFIELD. 1970. Resistance of house flies (Diptera: Muscidae) to dimethoate and Ronnell in Florida. Florida Entomol. 53:1-5. BAKER, P. B., A. M. SHELTON, AND J. T. ANDALORO. 1982. Monitoring of diamondback moth (Lepidoptera: Yponomeutidae) in cabbage with pheromones. J. Econ. En- tomol. 75:1025-1028. BLOOMCAMP, C. L., R. S. PATTERSON, AND P. G. KOEHLER. 1987. Cyromazine resistance in the house fly (Diptera: Muscidae). J. Econ. Entomol. 80:352-357. BROGDON, J. E. 1957. Symposium—vegetable insect resistance to insecticides in Flor- ida. Proc. Florida State Hort. Soc. 70:143-144. EL-GAZZAR, L. M., J. MILIO, P. G. KOEHLER, AND R. S. PATTERSON. 1986. Insecticide resistance in the cat flea (Siphonaptera: Pulicidae). J. Econ. Entomol. 79:132- 134. GENUNG, W. G. 1957. Some possible cases of insect resistance to insecticides in Flor- ida. Proc. Florida State Hort. Soc. 70:148-152. GENUNG, W. G., S. L. POE, AND C. A. MUSGRAVE. 1979. Insect and mite pests of celery, pp. 29-52 in S. L. Poe and J. O. Strandberg [eds.], Opportunities for integrated 398 Florida Entomologist 78(3) September, 1995

pest management in celery production. University of Florida, Institute of Food & Agricultural Sciences. UF-IFAS-IPM 2. 104 pp. HAMA, H., K. SUZUKI, AND H. TANAKA. 1992. Inheritance and stability of resistance to Bacillus thuringiensis formulations of the diamondback moth, Plutella xylostella (Linnaeus) (Lepidoptera: Yponomeutidae). Appl. Entomol. Zool. 27:355-362. HARRIS, E. D., JR. 1962. Insecticides and intervals between applications for leafminer control on celery. Proc. Florida State Hort. Soc. 75:184-189. HOSKINS, W. M., AND H. T. GORDON. 1956. Arthropod resistance to chemicals. Annu. Rev. Entomol. 1:89-122. HOSTETLER, M. E., AND R. J. BRENNER. 1994. Behavioral and physiological resistance to insecticides in the German cockroach (Dictyoptera: Blatteridae): an experimental reevaluation. J. Econ. Entomol. 87:885-893. KEIL, C. B., AND M. P. PARELLA. 1990. Characterization of insecticide resistance in two colonies of Liriomyza trifolii (Diptera: Agromyzidae). J. Econ. Entomol. 83:18-26. KOEHLER, P. G. 1991. Toxicity of hydramethylnon to laboratory and ﬁeld strains of German cockroach (Orthoptera: Blattellidae). Florida Entomol. 74:345-349. LEIBEE, G. L. 1986. Caterpillar control on cabbage, spring 1985. CFREC-Sanford Re- search Report SAN 86-10. Univ. Florida IFAS. 2 p. LEIBEE, G. L., AND K. E. SAVAGE. 1992a. Observations on insecticide resistance in diamondback moth, pp. 41-46 in Seminar Proceedings: Global Management of Insecticide Resistance In The 90s. Abbott Laboratories. Sept. 15-17, 1992. Lake Bluff, IL. LEIBEE, G. L., AND K. E. SAVAGE. 1992b. Toxicity of selected insecticides to two laboratory strains of insecticide-resistant diamondback moth (Lepidoptera: Plutel- lidae) from central Florida. J. Econ. Entomol. 85:2073-2076. MAGARO, J. J., AND J. V. EDELSON. 1990. Diamondback moth (Lepidoptera: Plutel- lidae) in south Texas: a technique for resistance monitoring in the ﬁeld. J. Econ. Entomol. 83:1201-1206. METCALF, C. L., AND W. P. FLINT. 1939. Destructive and useful insects. McGraw-Hill Book Co., New York. METCALF, C. L., W. P. FLINT, AND R. L. METCALF. 1951. Destructive and useful insects. McGraw-Hill Book Co., New York. METCALF, C. L., W. P. FLINT, AND R. L. METCALF. 1962. Destructive and useful insects. McGraw-Hill Book Co., New York. MILIO, J. F., P. G. KOEHLER, AND R. S. PATTERSON. 1987. Evaluation of three methods for detecting chlorpyrifos resistance in german cockroach (Orthoptera: Blattel- lidae) populations. J. Econ. Entomol. 80:44-46. OMOTO, C., T. J. DENNEHY, C. W. MCCOY, S. E. CRANE, AND J. W. LONG. 1994. Detec- tion and characterization of the interpopulation variation of citrus rust mite (Acari: Eriophyidae) resistance to dicofol in Florida Citrus. J. Econ. Entomol. 87:566-572. PITRE, H. N. 1988. Relationship of fall armyworm (Lepidoptera: Noctuidae) from Flor- ida, Honduras, Jamaica, and Mississippi: susceptibility to insecticides with reference to migration. Florida Entomol. 71:56-61. POE, S. L., AND J. O. STRANDBERG. 1979. Crop protection through prevention and management, pp 1-4 in S. L. Poe and J. O. Strandberg [eds.], Plant protection through integrated pest management. Opportunities for integrated pest management in celery production. University of Florida, Institute of Food & Agri- cultural Sciences. UF-IFAS-IPM 2. 104 pp. SANDERSON, E. D. 1921. Insect pests of farm, garden, and orchard. John Wiley and Sons, Inc., London. SHELTON, A. M., AND D. M. SODERLUND. 1983. Varying susceptibility to methomyl and permethrin in widely separated cabbage looper (Lepidoptera: Noctuidae) populations within eastern North America. J. Econ. Entomol. 76:987-989. Leibee and Capinera: Symposium on Pesticide Resistance 399

SHELTON, A. M., J. L. ROBERTSON, J. D. TANG, C. PEREZ, S. D. EIGENBRODE, H. K. PREISLER, W. T. WILSEY, AND R. J. COOLEY. 1993. Resistance of diamondback moth (Lepidoptera: Plutellidae) to Bacillus thuringiensis subspecies in the ﬁeld. J. Econ. Entomol. 86:697-705. SYED, A. R. 1992. Insecticide resistance in diamondback moth in Malaysia, pp 437-442 in N. S. Talekar [ed.], Proceedings of the second international work- shop on the management of diamondback moth and other crucifer pests. Asian Vegetable Research and Development Center, Tainan, Taiwan. TABASHNIK, B. E., N. L. CUSHING, N. FINSON, AND M. W. JOHNSON. 1990. Field development of resistance to Bacillus thuringiensis in diamondback moth (Lepi- doptera: Plutellidae). J. Econ. Entomol. 83:1671-1676. TABASHNIK, B. E., J. W. SCHWARTZ, N. FINSON, AND M. W. JOHNSON. 1992. Inherit- ance of resistance to Bacillus thuringiensis in diamondback moth (Lepidoptera: Plutellidae). J. Econ. Entomol. 85:1046-1055. TANAKA, H., AND Y. KIMURA. 1991. Resistance to Bt formulation in diamondback moth, Plutella xylostella L., on watercress. Japan J. Appl. Entomol. Zool. 35:253-255. WATSON, J. R., AND A. N. TISSOT. 1942. Insects and other pests of Florida vegetables. Univ. of Florida Agric. Exp. Stn. Bull. 370. WOLFENBARGER, D. O. 1947. The serpentine leaf miner and its control. Univ. of Flor- ida Agric. Exp. Sta. Press Bul. 639. WORKMAN, R. B., AND G. L. GREENE. 1970. Cabbage looper control at Hastings and Sanford Florida during 1970. Proc. Florida State Hort. Soc. 83:166-167. YU, S. J. 1992. Detection and biochemical characterization of insecticide resistance in fall armyworm (Lepidoptera: Noctuidae). J. Econ. Entomol. 85:675-682. YU, S. J., AND S. N. NGUYEN. 1992. Detection and biochemical characterization of insecticide resistance in the diamondback moth. Pestic. Biochem. Physiol. 44:74- 81.

♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦

Scott: Symposium on Pesticide Resistance 399

THE MOLECULAR GENETICS OF RESISTANCE: RESISTANCE AS A RESPONSE TO STRESS

JULIE A. SCOTT Knipling-Bushland USDA Livestock Insect Research Laboratory Agricultural Research Service Kerrville, TX 78028-9184

ABSTRACT

In this overview of the molecular genetics of resistance, pesticides are regarded as one of the many environmental stresses against which insects must defend themselves to survive. Examined at the genetic level, pesticide resistance appears to be a preadapted response to stress and not due to novel mutations caused by pesticide exposure. The genetic mutations—gene ampliﬁcation, altered gene regulation, structural alteration of a gene—which result in resistance are described and explained and a possible distribution mechanism of resistance genes is considered. Resistance mechanisms, their associated biological processes and the types of genetic mutations associated with each are detailed. Finally, the potential of molecular technology for the development of novel methods to detect and monitor for resistance is examined and compared to more traditional technology.

400 Florida Entomologist 78(3) September, 1995

Key Words: Pesticide resistance, molecular, mutation, genetic, adaptation, stress, insect.

RESUMEN

En esta revisión de la genética molecular de la resistencia, los pesticidas son considerados como uno de los muchos estréses ambientales de los cuales los insectos de- ben defenderse para sobrevivir. Examinada a nivel genético, la resistencia a los pesticidas parece ser una respuesta preadaptada al estrés y no debida a nuevas mutaciones causadas por la exposición a los pesticidas. Son descritas y explicadas las mutaciones genéticas-ampliﬁcación genética, regulación génica alterada, alteración estructural de un gen-que producen la resistencia, y es considerado un posible meca- nismo de distribución de los genes de resistencia. Son detallados los mecanismos de resistencia, los procesos biológicos y los tipos de mutaciones asociados con cada uno de ellos. Finalmente, es examinado y comparado a la tecnología más tradicional el poten- cial de la tecnología molecular para el desarrollo de nuevos métodos de deteccción y monitoreo de la resistencia.

When we discuss pesticide resistance, we generally refer to our ability to control a pest, not the pest’s ability to defend itself. Yet resistance is really a form of self-defense. To an insect, exposure to a pesticide is just one of the myriad of dangers which must be avoided in order to survive. In this sense, pesticide exposure may be described as an environmental stress and resistance as the overt expression of an insect’s natural response to that stress. Insects have been confronted with lethal and nonlethal stresses for as long as they have existed. And, they have evolved effective defense mechanisms to deal with these stresses. Their potential to adapt and develop resistance to stress becomes most apparent when examined at the molecular genetic level. It is the objective of this paper to examine how and why insects adapt to stress, particularly pesticide exposure at the molecular genetic level and to emphasize that resistance is a part of the normal response of insects to stress. In addition, the advantages and limitations of traditional and molecular technologies for monitoring for resistance are brieﬂy compared. This paper is a general overview to introduce readers to concepts of molecular genetics and pesticide resistance. It is not a technical review of the molecular biology of speciﬁc resistance mechanisms.

STRESS, RESISTANCE AND TOLERANCE: DEFINITIONS AND INTERACTIONS Before discussing the molecular genetics of resistance, some definition and discussion of the relationships of stress, resistance, and tolerance is required. Stress has been broadly defined as “any environmental change that acts to reduce the fitness of an organism” (Koehn & Bayne 1989). Stresses may have physical, biotic, and/or toxic components (Fig. 1) which, in turn, may be acute, chronic, and/or seasonal and affect insects at the community, population, and/or individual level. The deleterious effects of excesses in temperature or humidity are self-evident. Exposure to ultraviolet radiation can influence trophic-level interactions of entire communities to the advantage of one population at the expense of another (Bothwell et al. 1994). Predation, parasitism, disease, inter- and intraspecies competition all act to determine the success or failure of a population to occupy a particular ecological niche. Toxic components of the environment are also stresses that can affect populations and can be divided into three groups: pollutants, which may be natural or artificial, pesticides, and plant al-

Scott: Symposium on Pesticide Resistance 401

Figure 1. Examples of environmental stresses which can act on living organisms and force individuals, populations, and entire communities to continuously adapt to new and ever changing conditions in order to survive. lelochemicals. Each of these stresses can affect insects differently, but all must be adequately dealt with for continued survival. Resistance has been defined as the development of the ability in a strain of an organism to tolerate doses of a toxicant which would prove lethal to the majority of individuals in a normal (i.e. susceptible) population of the same species (Anonymous 1957). This definition is somewhat imprecise, because it infers that resistance can develop in an individual either before or after exposure to a toxin, two very different events. Resistance is the phenotypic expression throughout a population of a heritable trait that was already expressed in at least some of the individuals in the population prior to exposure to a toxicant. The development of a measurable shift in a population’s susceptibility to a toxin is due to the specific selection of these pre- adapted individuals in the population, often over several generations, by exposure to amounts of toxicant which are sublethal to the pre-adapted individuals but may or may not be sublethal to others in the population. For many years this adaptive event was difficult to understand; however, we now know that the toxic components of some pesticides are similar to ones present in plants (e.g. pyrethrum) and that the detoxification systems that deal with these plant allelochemicals are the same systems that detoxify pesticides. Therefore insects which can detoxify certain plant allelochemicals well are pre-adapted to detoxify and develop resistance to pesticides which have the same mode of action as the allelochemicals even before the insects are ever exposed to the pesticides. Resistance and tolerance are often used interchangeably in the literature and to define one another; however, tolerance is also used to describe shifts in susceptibility that occur within a single generation after exposure to stress which is not expressed by succeeding generations until after exposure to a similar stress. This phenomenon is different from the one previously described. For example 6-day-old bollworms, He-

402 Florida Entomologist 78(3) September, 1995 liothis zea (Boddie), only detoxify relatively small amounts of methyl parathion, but at 12 days, they can detoxify 30-fold more methyl parathion. If gossypol, a toxic allelochemical present in cotton, is added to the diet, then 12-day-old larvae can detoxify not 30-fold but 75-fold more methyl parathion. The presence of gossypol in the diet induced the larvae to produce more detoxifying enzymes (Muehleisen et al. 1989). Yet, the 6-day-old progeny of these individuals do not retain either the 30- or 75-fold increase in tolerance of their parents but must develop it over time and after exposure to gossypol. The ability to metabolize methyl parathion is inherent; the increase in tolerance to it is not. The difference between resistance and tolerance can become blurred when a population is subjected to a strong selection pressure, such as chronic pesticide exposure, and the mechanism that is induced by the pesticide exposure to yield tolerance is also the mechanism that is speciﬁcally selected to yield resistance. Resistance, as it will be discussed here, refers to a decrease in susceptibility which is heritable and does not need to be induced before it can be expressed; however, exposure to a stress, such as an insecticide, may result in an increase in the expression of the resistance gene(s) which may or may not be heritable. Two other terms which are often used and may be confused with one another are cross-resistance and multi-resistance. Cross-resistance is resistance to two or more classes of pesticides which occurs because the pesticides have the same, or very similar, modes of action. Organophosphate and carbamate pesticides intoxicate by similar modes of action, and resistance to one usually results in resistance to the other. Multi-resistance refers to resistance to two or more classes of pesticides because of the coexistence of two or more different resistance mechanisms. For example, a resistant insect may have both metabolic resistance to organophosphates and target-site resistance to pyrethroids.

THE WAYS AND MEANS OF ALTERING GENETIC MATERIAL It should be apparent that pre-exposure to plant allelochemicals can only help explain those cases of resistance where the pesticide is rendered inactive by the same detoxification mechanism. It does not explain resistance mechanisms which do not appear to be selected for by plant allelochemicals or seem to arise spontaneously or increase in amplitude after exposure to a pesticide. How can exposure of the parental generation to stress result in their progeny being more resistant to that stress? The answer is that exposure to a stress causes the genetic material (i.e. the DNA) to be altered. There appear to be three general types of alterations, i.e. mutations, that can occur and result in resistance (Fig. 2). A gene may be amplified so that instead of only having one copy of the gene, there are now many copies present in the DNA. If an insect has ten copies of a gene, then it can make ten times as much product as an insect with only one copy of the gene. If the amplified gene encodes for a detoxifying enzyme, then that insect can detoxify 10-fold as much toxicant as the insect with only one gene. The expression of a gene may also be altered to yield resistance. In this case, there is only one copy of the gene present in the mutated insect but that gene’s regulation is altered so that it produces more (or less) product compared to a susceptible individual. For example, in a susceptible insect the gene to gene-product ratio may be 1:1 but in a resistant insect that ratio may be changed. The gene may be up-regulated to produce more product, that is, the gene to gene-product ratio is now 1:10 or down-regulated to make less product (the ratio is now 1:0.1). When a pesticide is applied in its toxic form, up-regulation of a detoxifying enzyme will increase resistance. When a pro-insecticide, i.e. the material must be metabolized first in order to become toxic, is applied, down-regulation of the metabolizing enzyme will increase resistance.

Scott: Symposium on Pesticide Resistance 403

Figure 2. Graphic representation of the types of genetic mutations which occur and cause resistance. (a) A gene is ampliﬁed to increase its number of copies in the genome and consequently increase the amount of gene product made (b) the regulatory expression of a gene is altered to increase the amount of gene product made (note that gene expression may also be altered to decrease the amount of product made) (c) the genetic code is rewritten to produce a structurally different product.

The third type of mutation that can result in resistance is a structural change in a gene which yields a corresponding structural change in its product. A single point mutation, i.e. one nucleotide in the gene’s coding region is substituted with a different nucleotide so that a different amino acid is encoded for at a specific position and this change causes the gene product to have a different three-dimensional structure, can result in resistance in several ways. It may decrease the ability of the insecticide to physically bind to its site of action, or increase or decrease the gene product’s ability to metabolize the insecticide. A structural change does not alter the quantity of the product made but alters the quality of the product made. It is important to recognize that these alterations to an insect’s DNA do not create new genes. They only affect pre-existing genes. The idea that exposure to a pesticide causes resistance genes to be “created” has been debated periodically; however, no substantiating data have ever been proffered. It is much more likely that resistance genes already exist in the pesticide-naive population at low frequency prior to selection by a pesticide. How resistance genes are spread throughout a population is not completely certain. It is clear that pesticide exposure plays a key role. However, other factors may also be important in the spread of resistance genes. One hypothesis which has received considerable attention is that transposable or mobile elements play a significant role in some cases of gene amplification. Transposable or mobile elements (TEs) are discrete sections of DNA that can move to new chromosomal locations and prolif- erate at a higher frequency relative to other genomic sequences (i.e. more and more copies of the TE are inserted into the genome) after they have moved. TEs can also move genes that were previously not mobile and whose functions are not related to the transposition. The genes that are moved with the TEs are also replicated at a higher frequency (Berg 1989). Gene amplifications may be initially distributed throughout a population by TEs or a gene may be transposed to a new location where its expression is altered to yield resistance. For example, it has been indirectly demonstrated in the laboratory that the transposition of alleles, alternative forms, of the gene Met results in insecticide resistance in Drosophila (Wilson & Turner 1992). But

404 Florida Entomologist 78(3) September, 1995 so far there is absolutely no evidence that such an event has occurred in a ﬁeld population.

RESISTANCE MECHANISMS AND ASSOCIATED TYPES OF MUTATIONS Interestingly, only specific types of genetic mutations appear to be associated with specific resistance mechanisms (Table 1). The two types of mechanisms that cause high levels of resistance are generally referred to as metabolic resistance and target- site insensitivity, respectively. Each of these consists of several biological mechanisms. Metabolic resistance can be divided into three principle enzyme systems: cytochrome P450 monooxygenases (P450s), nonspecific esterases, and glutathione S- transferases (GSTs). Components of each of these enzyme systems may be mutated to alter the detoxification of a pesticide. Cytochrome P450s catalyze a variety of detoxification reactions in insects, including the hydroxylation of DDT, the epoxidation of cyclodienes, the aromatic hydroxylation of the carbamates carbaryl and propoxur, and oxidation of phosphorothioates (Feyereisen et al. 1990). Given the variety of reactions stimulated by these enzymes, it is likely that several different P450 enzymes are present in any one insect and that several alleles of each gene may exist. Such is the case for the mosquito Anopheles albimanus in which seventeen P450 genes were discovered (Scott et al. 1994) and for the termites Mastotermes darwiniensis and Coptotermes acinaciformis in which multiple isoenzymes of cytochrome P450s were detected biochemically (Haritos et al. 1994). There is no evidence to suggest that P450 genes are amplified or structurally altered to yield insecticide resistance. But there are numerous examples of their expression being altered by various substances (Rose et al. 1991, Jeong et al. 1992, Wax- man & Azaroff 1992, Snyder et al. 1993). By definition, if the expression of a P450 is

TABLE 1. THE GENETIC MUTATIONS ASSOCIATED WITH ENZYMES AND RECEPTORS THAT RESULT IN DIFFERENT TYPES OF RESISTANCE.

Associated Genetic Mutations

Types of Gene Altered Structural Resistance Ampliﬁcation Expression Change

Metabolic P450 oxidases ND1 +ND Esterases + ND + GSTs ND + ? Target site insensitivity Acetylcholinesterase ND ND + GABA receptor ND ND + Sodium channel ND ? ? JH receptor ND ? ? Other Reduced penetration — — — Behavioral change — — —

1ND = not detected; + = conﬁrmed or strongly indicated; ? = implied but not conﬁrmed; — = no data available.

Scott: Symposium on Pesticide Resistance 405 altered to yield resistance, then susceptible and resistant strains will have quantita- tively different amounts of that P450. Three P450s have been demonstrated to be over-expressed by resistant insect strains: P450Lpr (Wheelock & Scott 1992), CYP6A1 (Cariño et al. 1994), and CYP6A2 (Waters et al. 1992), respectively. P450Lpr has been directly implicated as the major enzyme causing pyrethroid resistance in one strain of house fly (Wheelock & Scott 1992, Hatano & Scott 1993). CYP6A1 appears to be a major cyclodiene-metabolizing enzyme in the house fly (Andersen et al. 1994). The primary function of CYP6A2 has not been reported. A major handicap to determining which P450s are involved in resistance is our inability to distinguish the activities of individual uncharacterized P450s from each other with a high degree of accuracy. The activity of a specific P450 towards an insecticide must be determined because its over-expression does not necessarily prove that it is the enzyme responsible for resistance. The over-expression of CYP6A1 by the house fly is a case in point. CYP6A1 is over-expressed in a house fly strain that is resistant to DDT, organophosphates, and carbamates but does not have significant resistance to cyclodienes. To determine which P450s are actually responsible for resistance, more specific substrates are needed. Esterases are a large group of enzymes which metabolize a wide variety of substrates. All esterases are able to hydrolyze ester bonds in the presence of water. Since many insecticides, especially organophosphates and carbamates, contain ester bonds, it is not surprising that the mechanism of resistance in many cases is caused by elevated levels of esterases (Fournier et al. 1987, Field et al. 1988, Carlini et al. 1991, Kettermen et al. 1992, Chen & Sun 1994). Esterase levels can be elevated by either gene amplification or altered gene expression. So far, no data have indicated that the expression of esterase genes is altered to yield resistance, but the molecular characterization of esterases is limited to a very few insect species and this type of mutation cannot be discounted. Considerable data show that certain non-specific esterase genes are amplified to yield resistance. The esterases which cause resistance in Myzus persicae Sulz. and Culex mosquitoes have been particularly well-studied. In these insects the resistant esterase genes are highly amplified and up to 250 copies of the same gene may be found in a single individual (Mouchès et al. 1986, Poirié et al. 1992). The more the esterase genes are amplified, the greater the level of resistance that they provide (Field et al. 1988, Poirié et al.1992). This increase in resistance appears to be because the esterases interact with the insecticides more readily than the insecticides’ own target. When the esterases are present in approximately an equal molar ratio to the insecticides, they are able to effectively sequester the insecticides and then slowly hydrolyze the insecticides (Devonshire & Moore 1982, Ketterman et al. 1992, Karunaratne et al. 1993). How recently these esterases have been amplified in Culex populations has been a matter of some debate. One group has suggested that two esterases associated with resistance, A2 and B2, were amplified in a single event within the past forty years, i.e. since the use of organophosphate insecticides became widespread, and that A2 and B2 have been distributed across three continents by migration since that event (Ray- mond et al. 1991). If this hypothesis is correct then all A2 and B2 genes should be identical. But data show that they are not. Kinetic studies of the insecticidal interaction of purified A2 and B2 esterases found that different forms of each enzyme were present in a number of resistant strains (Ketterman at al. 1993), and three amino acid differences have been found between the two B2 genes that have been sequenced (Vaughan et al. 1995). Therefore, not all of the A2 and B2 genes are identical and either these genes were amplified in at least two separate events or they were amplified

406 Florida Entomologist 78(3) September, 1995 once long ago and have since diverged. Examination of other amplified esterase genes from different Culex strains clearly reveals numerous differences at the molecular level and strongly suggests that multiple amplification events have occurred with them (Vaughan et al. 1995). In addition to amplification, esterases may be mutated to produce structurally different enzymes which are able to metabolize insecticides more efficiently. In the Aus- tralian sheep blow fly Lucilia cuprina and the mosquito C. tarsalis, a carboxylesterase appears to be structurally altered in resistant populations to produce high levels of resistance to malathion (Whyard et al. 1994a, 1994b). In neither species is more of the enzyme produced. The difference between susceptible and resistant populations is strictly a qualitative difference in the enzyme produced. Whether or not the carboxylesterases from the blow fly and mosquito are homologous to each other will only be known through further molecular and biochemical analysis. The final group of enzymes which may provide metabolic resistance are the GSTs. Both organophosphate and cyclodiene pesticides can be detoxified by GST pathways. These enzymes have been somewhat less studied at the biochemical and molecular level in insects than the P450s and esterases. Elevated GST levels are found in many resistant insect strains (Motoyama & Dauterman 1975, Ottea & Plapp 1984, Aham- mad-Sahib et al. 1994, Hoffman & Fisher 1994) and increased GST activity is clearly the underlying resistance mechanism in some cases (Kao & Sun 1991, Wang et al. 1991, Prapanthadara et al. 1993). But in other resistant populations the increased GST activity does not cause resistance (Bush et al. 1993, Hemingway et al. 1993, Ar- gentine et al. 1994). Both insecticides and plant allelochemicals induce increased GST production (Yu 1992a, Lagadic et al. 1993, Leszczynski et al. 1994), and generalist plant feeders seem to rely on GST pathways more heavily than do specialist feeders to metabolize plant allelochemicals (Yu 1992b). Because GSTs can metabolize a wide variety of substances, increased GST activity may be part of a generalized compensa- tory change due to exposure to an environmental stress. How GST activity is increased has only been examined in a few Diptera. In dipterans, increased GST activity does not appear to be the result of gene amplification. In both the house fly and Drosophila, several GSTs contribute to resistance and their expression appears to have been increased through a regulatory change (Wang et al. 1991, Cochrane et al. 1992, Fournier et al. 1992). In addition, at least one resistance gene is structurally altered in Drosophila (Cochrane et al. 1992). Whether or not this structural change affects the resistance level in the flies remains to be determined. As numerous GSTs have been found in several insects, each of which appears to be encoded by a different gene (Cochrane et al. 1992, Fournier et al. 1992, Baker et al. 1994), it is likely that resistance caused by GSTs is due to altered gene expression and/or structural changes and is not due to gene amplification in most, if not all, cases. The second major resistance mechanism is target-site insensitivity, which refers to an alteration of the molecule(s) that directly interacts with the pesticide to reduce toxicity. Both acetylcholinesterase (AChE) and the gamma-aminobutyric acid (GABA) receptor are known targets of insecticides, and resistant alleles of each have been found. Voltage-gated sodium channels and the juvenile hormone (JH) receptor are putative targets of insecticides. Their direct interaction with insecticides has not been confirmed, but it is clearly evident that they play a key role in the intoxication process. Acetylcholinesterase is the target site of both organophosphates and carbamates. These pesticides bind to AChE and prevent the enzyme from stopping the action of the neurotransmitter acetylcholine. Multiple forms of AChE that confer varying degrees of resistance have been found in a variety of arthropods (Nolan et al. 1972, Devonshire & Moore 1984, Pralavorio & Fournier 1992). In each case examined so far, the affinity

Scott: Symposium on Pesticide Resistance 407 of AChE for the pesticide has been reduced. Neither gene amplification nor altered expression of the gene encoding AChE has been detected. Instead, point mutations have occurred to structurally change the enzyme. Recently, Mutero et al. (1994) identified five point mutations in D. melanogaster which are associated with reduced sensitivity. Several strains of resistant flies were found to have a combination of mutations. Indi- vidually, the mutations gave only low levels of resistance, but when several of them were combined, high levels of resistance resulted. Mutero et al. (1994) hypothesize that decreased sensitivity by AChE is the result of a combination of several mutations, each of which provides a little resistance instead of the appearance of a single mutation which yields strong resistance. GABA receptors are the primary target of cyclodiene insecticides. In vertebrates, these receptors group together to form a complete chloride ion channel. It is inferred that invertebrates have similar ion channels. Most cases of cyclodiene resistance appear to be due to decreased sensitivity of the GABA subtype A receptor (ffrench-Con- stant et al. 1991), an integral part of the chloride ion channel. Like AChE, decreased sensitivity by GABA receptors is due to a structural change of the protein. Neither amplification nor altered expression of the GABA receptor gene has been detected. Unlike AChE, only a single point mutation which causes one specific amino acid to be substituted with another results in high levels of resistance to cyclodienes. Other point mutations have been detected, but no others appear to cause resistance or are consistently associated with resistance (ffrench-Constant et al. 1993, Thompson et al 1993). Voltage-gated sodium channels play an integral role in the transmission of neural impulses. Pyrethroids disrupt neural transmissions by interrupting the normal functioning of voltage-gated sodium channels. Target-site insensitivity to pyrethroids, a phenotypic response commonly referred to as knockdown resistance (kdr), results in these sodium channels becoming less sensitive to intoxication. Although it is clear that sodium channels are adversely affected by pyrethroids, there is disagreement as to whether or not kdr is the result of a mutation to the sodium channels or to some other molecule which is integral to the functioning of the sodium channels. A point mutation to a sodium channel, which could structurally alter it, was detected in a resistant insect strain (Amichot et al. 1992) but so far this mutation has not been shown to cause resistance. Other reports indicate that the mutation which causes kdr is closely linked to (physically close to or a part of) the gene encoding one type of sodium channel (Knipple et al. 1994, Dong & Scott 1994) and this linkage has been inter- preted as strong evidence that a mutation(s) to the sodium channel gene is associated with kdr. On the other hand, a mutation to a regulatory protein or receptor could also result in kdr. There is some indirect evidence to support this alternative hypothesis (Rossignol 1991, Osborne & Pepper 1992). A third hypothesis that has been proposed and for which there is limited electrophysiological evidence is that kdr is caused by changes to two closely linked genes. One involves an altered sodium channel and the other may be associated with calcium-activated phosphorylation of a protein(s) involved with neurotransmitter release (Pepper & Osborne 1993). At this time, it can only be stated that there is no indication that a gene amplification event is associated with kdr. It is most likely that either the expression of a gene associated with sodium channel function has been altered and/or a structural mutation to sodium channels results in kdr. Juvenile hormone analogs such as methoprene compete with the natural hormone for the JH receptor. Most resistance to JH analogs is either metabolic and/or a reduction in penetration of the insecticide through the cuticle. The only reports of target site insensitivity to JH analogs are in mutants isolated from laboratory colonies of

408 Florida Entomologist 78(3) September, 1995

Drosophila (Shemshedini & Wilson 1990, Wilson & Turner 1992). The resistance gene Met that has been isolated from these colonies is associated with a less sensitive cys- tolic JH binding protein. Transposition of alleles of Met by a TE can induce resistance (Turner 1993); however, transposition is not required for the resistance gene to be expressed. Therefore, it seems most likely that the insensitive JH binding protein is either the result of a mutation that structurally changes the protein or the expression of the insect growth cycle has been altered. Gene ampliﬁcation of JH receptors has not been implicated. Two other types of resistance that have been described are the reduced penetration of a pesticide and altered behavior to avoid a pesticide. It is presumed that the cu- ticular structure is somehow altered to reduce the rate of penetration of a pesticide. Avoidance behavior appears to be stimulated by brief contact, either through tactile or olfactory receptors, with a pesticide. Alone, neither of these mechanisms cause high levels of resistance, but they are often found in combination with other types of resistance and can make a signiﬁcant contribution to the overall resistance displayed by an insect. For example, Raymond et al. (1989) calculated that reduced penetration interacts with any other resistance mechanism multiplicatively. Experimental data support this conclusion (Hoyer & Plapp 1968, Plapp & Hoyer 1968). The underlying physiological, genetic and molecular mechanisms that cause these types of resistance can only be speculated and the genetic mutations which cause them cannot be inferred at this time.

DETECTING AND MONITORING FOR RESISTANCE

Resistance is a widespread phenomenon and resistant populations of nearly all economically important pests can now be found (Georghiou 1994, Leibee & Capinera this issue). Where control failures have occurred, the history of pesticide application usually indicates that chronic pesticide exposure resulted in high levels of resistance which caused the failures. Does the development of a resistant population then mean that a control failure is inevitable? Certainly the continuous selection of the same resistance mechanism(s) over and over will result in resistance levels that are high enough to cause a control failure. But if the selection pressure (i.e. the pesticide) for each resistance mechanism is removed prior to signiﬁcant resistance developing, then a control failure may be avoided (for an alternative view on this subject, see Hoy this issue). Keys to the successful avoidance of a control failure are the detection of resistance at low level(s), the routine monitoring for changes in the level(s) and type(s) of resistance present in the pest population and the implementation of, and strict adher- ence to, a multi-tactic pest management program. In many cases, the two former elements have not been fully utilized to allow for the design of an effective pest management program. This lack of implementation is due in part to the unavailabil- ity of easy to use resistance detection methods. The ideal detection method is fast, inexpensive, easy to use, diagnostic for all types of resistance and able to detect resistance at frequencies as low as 1%. Numerous methods to detect resistance are currently available but all fall short of being the ideal technique. In fact, it is very unlikely that such an ideal technique will ever be developed. Instead, we must rely upon traditional methods, principally bioassays and a limited number of biochemical assays, and novel molecular techniques to detect and monitor for resistance. Certain general features are shared by the detection assays within each of these two technological groupings, i.e. traditional and molecular; and consequently, they have similar advantages and disadvantages which are summarized in Table 2.

Scott: Symposium on Pesticide Resistance 409

In general, detection methods which utilize molecular technology are better able to distinguish between the different resistance genotypes, i.e. heterozygotes (SR), homozygous susceptible (SS) and homozygous resistant (RR), than traditional detection assays. Because they detect only genetic differences, molecular assays can eliminate the environmental components which often increase the variability in bioassay and biochemical results. Direct comparison of a molecular assay and a bioassay (Aron- stein et al. 1994) indicates that molecular assays better approximate the perfectly diagnostic assay described by Roush & Miller (1986) than traditional bioassays and, therefore, can require up to 5-fold fewer insects to yield the same information. And unlike traditional assays, molecular techniques can use material from a single insect to perform several different assays so that the resistance levels to a wide variety of pesticides can be determined from the same individuals. However, molecular assays are limited to the detection of known resistance genes and a separate assay must be done for each gene. Traditional bioassays are better able to detect the overall level of resistance present in a population in a single test. In fact, molecular assays cannot detect many types of resistance at this time since with few exceptions we do not know which specific genes cause resistance. In time, molecular assays will be developed to detect more resistance genes; however, it is unlikely that molecular assays will completely replace the traditional ones. In addition, molecular assays are more costly in both material and equipment and require greater technical training than simple bioassays. They cannot be used in the field and usually take significantly longer to complete. Although many resistance genes have not been isolated, it is clear from the molecular and genetic data that it is highly unlikely that resistance genes are the result of novel mutations which create new genes. Instead resistance, a phenotypic response, is the result of the selection by pesticides of alleles of pre-existing genes, i.e. specific genotypes, that regulate or enhance particular defensive mechanisms. How many

TABLE 2. COMPARISON OF THE ADVANTAGES AND DISADVANTAGES OF DETECTION METHODS THAT ARE BASED ON EITHER MOLECULAR OR TRADITIONAL TECH- NOLOGY.

Molecular Detection Assay vs. Traditional Detection Assay

can distinguish resistant genotypes less sensitive; only detects resistant even at low levels phenotypes, not genotypes distinguishes SS, SR, and PR cannot distinguish SR and RR from genotypes reliably even when each other unless the phenotypes are phenotypes are hard to distinguish distinct from one another generally greater accuracy and less less accurate and more variable variability because environmental because environmental components components are eliminated cannot be separated from genetic ones fewer insects can yield more requires more insects to get the same information because material from one data because the same individuals insect can be used for several assays cannot be used in several bioassays currently not adapted for field use many are easily used in the field generally less rapid, it may take days very rapid, often only requiring a to complete a test couple of hours to do material and equipment often costly; generally inexpensive and simple to more technical expertise often needed prepare and execute limited to detection of known can detect any type of resistance, even resistance genes if the resistance gene is not known 410 Florida Entomologist 78(3) September, 1995 genes can cause resistance is not known. Data on various esterases and cytochrome P450s suggests that multiple genes may yield the same type of resistance. Only further analyses will determine if this is true. How resistant genes are distributed within and between populations is also not clear and requires further study. Under- standing the genetic flow of resistance genes and the biological costs of resistance are of particular importance because without this information it will be difficult, if not im- possible, to design successful control programs for many pests. Hopefully, the elucida- tion of the molecular genetics of resistance genes will help in the design of effective control programs and suggest novel control methods.

ACKNOWLEDGMENTS

The author gratefully thanks M. A. Hoy and D. C. Herzog for reviewing this manuscript and J. Hemingway for providing useful insights and a preprint of a relevant manuscript. This paper was presented as part of the symposium “The Myths of Man- aging Resistance” at the Florida Entomological Society Annual Meeting, August 8-11, 1994.

REFERENCES CITED

AHAMMAD-SAHIB, K. I., R. M. HOLLINGWORTH, M. E. WHALON, P. M. IOANNIDIS AND E. J. GRAFIUS. 1994. Polysubstrate monooxygenases and other xenobiotic-metabolizing enzymes in susceptible and resistant Colorado potato beetle. Pestic. Biochem. Physiol. 49:1-12. AMICHOT, M., C. CASTELLA, A. CUANNY, J. B. BERGÉ AND D. PAURON. 1992. Target modification as a molecular mechanism of pyrethroid resistance in Drosophila melanogaster. Pestic. Biochem. Physiol. 44:183-190. ANDERSEN, J. F., J. G. UTERMOHLEN AND R. FEYEREISEN. 1994. Expression of house fly CYP6A1 and NADPH-cytochrome P450 reductase in Escherichia coli and reconstitution of an insecticide-metabolizing P450 system. Biochemistry 33:2171-2177. ANONYMOUS. 1957. World Health Expert Committee on Insecticides 7th Report. WHO Technical Report Series No. 125. ARGENTINE, J. A., K. Y. ZHU, S. H. LEE AND J. M. CLARK. 1994. Biochemical mechanisms of azinophosmethyl resistance in isogenic strains of Colorado potato beetle. Pestic. Biochem. Physiol. 48:63-68. ARONSTEIN, K., P. ODE AND R. J. FFRENCH-CONSTANT. 1994. Direct comparison of PCR-based monitoring for cyclodiene resistance in Drosophila populations with insecticide bioassay. Pestic. Biochem. Physiol. 48:229-233. BAKER, W. L., A. G. CLARK, G. FAULDS AND J. S. NIELSEN. 1994. Multiple glutathione S-transferases in Galleria mellonella; their detection with fluorigenic substrates. Insect Biochem. Molec. Biol. 24:301-307. BERG, D. E. 1989. Transposable elements in prokaryotes, pp. 99-137 in S. B. Levy and R. V. Miller [eds.] Gene Transfer in the Environment. McGraw-Hill, New York, New York. BOTHWELL, M. L., D. M. J. SHERBOT AND C. M. POLLOCK. 1994. Ecosystem response to solar ultraviolet-B radiation: influence of trophic-level interations. Science 265-97-100. BUSH, M. R., Y. A. I. ABDEL-AAL, K. SAITO AND G. C. ROCK. 1993. Azinophosmethyl resistance in the tufted apple bud moth (Lepidoptera: Torticidae): reversion, diagnostic concentrations, associated esterases, and glutathione transferases. J. Econ. Entomol. 86:213-225. CARIÑO, F. A., J. F. KOENER, F. W. PLAPP AND R. FEYEREISEN. 1994. Constitutive over-expression of the cytochrome P450 gene CYP6A1 in a house fly strain with metabolic resistance to insecticides. Insect Biochem. Molec. Biol. 24:411-418. Scott: Symposium on Pesticide Resistance 411

CARLINI, E. J., B. A. MCPHERON, C. M. FELLAND AND L. A. HULL. 1991. Elevated esterase activity in resistant tufted apple bud moth, Platynota idaeusalis (Walker) (Lepidoptera: Tortricidae). Comp. Biochem. Physiol. 99C:375-377. CHEN, W-L. AND C. N. SUN. 1994. Purification and characterization of carboxylesterases of a rice brown planthopper, Nilaparvata lugens Ståal. Insect. Bio- chem. Molec. Biol. 24:347-355. COCHRANE, B. J., M. HARGIS, P. CROCQUET-DE-BELLIGNY, F. HOLTSBERG, AND J. COR- ONELLA. 1992. Evolution of glutathione S-transferases associated with insecticide resistance in Drosophila. ACS Symp. Ser. American Chem. Soc. Washington, D.C.: The Society 505:53-70. DEVONSHIRE, A. L. AND G. D. MOORE. 1982. A carboxylesterase with broad substrate specificity causes organophosphorus, carbamate and pyrethroid resistance in peach-potato aphids (Myzus persicae). Pestic. Biochem. Physiol. 18:235-246. DEVONSHIRE, A. L. AND G. D. MOORE. 1984. Different forms of insensitive acetylcholinesterase in insecticide-resistant house flies (Musca domestica). Pestic. Bio- chem. Physiol. 21:336-340. DONG, K. AND J. G. SCOTT. 1994. Linkage of kdr-type resistance and the para-homologous sodium channel gene in German cockroaches (Blatella germanica). In- sect Biochem. Molec. Biol. 24:647-654. FEYEREISEN, R., J. F. KOENER, F. A. CARIÑO AND A. S. DAGGETT. 1990. Biochemistry and molecular biology of insect cytochrome P450., pp. 263-272 in H. H. Hage- dorn, J. G. Hildebrand, M. G. Kidwell, and J. H. Law [eds.] Molecular Insect Science, Plenum, New York. FFRENCH-CONSTANT, R. H., D. P. MORTLOCK, C. D. SHAFFER, R. J. MACINTYRE AND R. T. ROUSH. 1991. Molecular cloning and transformation of cyclodiene resistance in Drosophila: an invertebrate γ-aminobutyric acid subtype A receptor locus. Proc. Natl. Acad. Sci. USA 88:7209-7213. FFRENCH-CONSTANT, R. H., J. C. STEICHEN, T. A. ROCHELEAU, K. ARONSTEIN AND R. T. ROUSH. 1993. A single-amino acid substitution in a γ-aminobutyric acid subtype A receptor locus is associated with cyclodiene insecticide resistance in Drosophila populations. Proc. Natl. Acad. Sci. USA 90:1957-1961. FIELD, L. M., A. L. DEVONSHIRE AND B. G. FORDE. 1988. Molecular evidence that insecticide resistance in peach-potato aphids (Myzus persicae Sulz.) results from amplification of an esterase gene. Biochem. J. 251:309-312. FOURNIER, D., J. M. BRIDE, C. MOUCHÈS, M. RAYMOND, M. MAGNIN, J-B. BERGÉ, N. PASTEUR AND G. P. GEORGHIOU. 1987. Biochemical characterization of the esterases A1 and B1 associated with organophosphate resistance in Culex pipiens L. complex. Pestic. Biochem. Physiol. 27:211-217. FOURNIER, D., J. M. BRIDE, M. POIRIÉ, J-B. BERGÉ AND F. W. PLAPP. 1992. Insect glutathione S-transferases: biochemical characteristics of the major forms from houseflies susceptible and resistant to insecticides. J. Biol. Chem. 267:1840- 1845. GEORGHIOU, G. P. 1994. Principles of insecticide resistance management. Phytopro- tection 75 (suppl.): 51-59. HARITOS, V. S., J. R. J. FRENCH AND J. T. AHOKAS. 1994. Cytochrome P450 monooxy- genase and glutathione S-transferase activity of two Australian termites: Mas- totermes darwiniensis and Coptotermes acinaciformis. Insect Biochem. Molec. Biol. 24:929-935. HATANO, R. AND J. G. SCOTT. 1993. Anti-P450Lpr antiserum inhibits the activation of chlorpyrifos to chlorpyrifos oxon in house fly microsomes. Pestic. Biochem. Physiol. 45:228-233. HEMINGWAY, J., G. J. SMALL AND A. G. MONRO. 1993. Possible mechanisms of organophosphorus and carbamate insecticide resistance in german cockroaches (Dic- tyoptera: Blattelidae) from different geographical areas. J. Econ. Entomol. 86:1623-1630. HOFFMAN, E. R. AND S. W. FISHER. 1994. Comparison of a field and laboratory-derived population of Chironomus riparius (Diptera: Chironomidae): biochemical and fitness evidence for population divergence. J. Econ. Entomol. 87:318-325. 412 Florida Entomologist 78(3) September, 1995

HOY, M. A. 1995. Multitactic resistance management: An approach that is long overdue? Florida Entomol. 78:443-451. HOYER, R. F. AND F. W. PLAPP. 1968. Insecticide resistance in the house fly: identification of a gene that confers resistance to organotin insecticides and acts as an intensifier of parathion resistance. J. Econ. Entomol. 61:1269-1276. JEONG, T. C., H. G. JEONG AND K-H. YANG. 1992. Induction of cytochrome P-450 by dimethyl sulfoxide in primary cultures of adult rate hepatocytes. Toxicol. Let- ters 61:275-281. KAO, C. H. AND C. N. SUN. 1991. In vitro degradation of some organophosphorus insecticides by susceptible and resistant diamondback moth. Pestic. Biochem. Physiol. 41:132-141. KARUNARATNE, S. H. P. P., K. G. I. JAYAWARDENA, J. HEMINGWAY AND A. J. KETTER- MAN. 1993. Characterization of a B-type esterase involved in insecticide resistance from the mosquito Culex quinquefasciatus. Biochem. J. 294:575-579. KETTERMAN, A. J., K. G. I. JAYAWARDENA AND J. HEMINGWAY. 1992. Purification and characterization of a carboxylesterase involved in insecticide resistance from the mosquito Culex quinquefasciatus. Biochem. J. 287:355-360. KETTERMAN, A. J., S. H. P. P. KARUNARATNE, K. G. JAYAWARDENA AND J. HEMING- WAY. 1993. Qualitative differences between populations of Culex quinquefasciatus in both the esterases A2 and B2 which are involved in insecticide resistance. Pestic. Biochem. Physiol. 47:142-148. KNIPPLE, D. C., K. E. DOYLE, P. A. MARSELLA-HERRICK AND D. M. SODERLUND. 1994. Tight genetic linkage between the kdr insecticide resistance trait and a voltage-sensitive sodium channel gene in the house fly. Proc. Natl. Acad. Sci. USA 91:2483-2487. KOEHN, R. K. AND B. L. BAYNE. 1989. Towards a physiological and genetical understanding of the energetics of the stress response. Biological Journal of the Lin- nean Society 37:157-171. LAGADIC, L., A. CUANY, J. B. BERGÉ AND M. ECHAUBARD. 1993. Purification and partial characterization of glutathione S-tranferases from insecticide-resistant and lindane-induced susceptible Spodoptera littoralis (Boisd.) larvae. Insect Biochem. Mol. Biol. 23:467-474. LEIBEE, G. L. AND CAPINERA, J. L. 1995. Pesticide resistance in Florida insects limits management options. Florida Entomol. 78:386-399. LESZCZYNSKI, B., M. MATOK AND A. F. G. DIXON. 1994. Detoxification of cereal plant allelochemicals by aphids: activity and molecular weights of glutathione S- transferase in three species of cereal aphids. J. Chem. Ecol. 20:387-394. MOTOYAMA, N. AND W. C. DAUTERMAN. 1975. Interstrain comparison of glutathione- dependent reactions in susceptible and resistant houseflies. Pestic. Biochem. Physiol. 5:489-495. MOUCHÈS, C., N. PASTEUR, J. B. BERGÉ, O. HYRIEN, M. RAYMOND, B. R. DE SAINT VIN- CENT, M. DE SILVESTRI G. P. GEORGHIOU. 1990. Amplification of an esterase gene is responsible for insecticide resistance in a California Culex mosquito. Science 233:778-780. MUEHLEISEN, D. P., J. H. BENEDICT, F. W. PLAPP, JR. AND F. A. CARIÑO. 1989. Effects of cotton allelochemicals on toxicity of insecticides and induction of detoxifying enzymes in bollworm (Lepidoptera: Noctuidae). J. Econ. Entomol. 82:1554- 1558. MUTERO, A., M. PRALAVORIO, J-M. BRIDE AND D. FOURNIER. 1994. Resistance-associated point mutations in insecticide-insensitive acetylcholinesterase. Proc. Natl. Acad. Sci. USA 91:5922-5926. NOLAN, J., H. J. SCHNITZERLING AND C. A. SCHUNTNER. 1972. Multiple forms of acetylcholinesterase from resistant and susceptible strains of the cattle tick, Boo- philus microplus (Can.). Pestic. Biochem. Physiol. 2:85-94. OSBORNE, M. P. AND D. R. PEPPER. 1992. Mechanisms of kdr and super-kdr resistance. ACS Symp. Ser. American Chem. Soc. 505:71-89. Scott: Symposium on Pesticide Resistance 413

OTTEA, J. A. AND F. W. PLAPP. 1984. Glutathione-S-transferase in the house fly: biochemical and genetic changes associated with induction and insecticide resistance. Pestic. Biochem. Physiol. 22:203-208. PEPPER, D. R. AND M. P. OSBORNE. 1993. Electrophysiological identification of site-insensitive mechanisms in knockdown-resistant strains (kdr, super-kdr) of the housefly larva (Musca domestica). Pestic. Sci. 39:279-286. PLAPP, F. W. AND R. F. HOYER. 1968. Insecticide resistance in the house fly: decreased rate of absorption as the mechanism of action of a gene that acts as an intensifier of resistance. J. Econ. Entomol. 61:1298-1303. POIRIÉ, M., M. RAYMOND AND N. PASTEUR. 1992. Identification of two distinct amplifications of the esterase B locus in Culex pipiens (L.) mosquitoes from Mediter- ranean countries. Biochem. Genet. 30:13-26. PRALAVORIO, M. AND D. FOURNIER. 1992. Drosophila acetylcholinesterase: characterization of different mutants resistant to insecticides. Biochem. Genet. 30:77-83. PRAPANTHADARA, L. A., J. HEMINGWAY AND A. J. KETTERMAN. 1993. Partial purification and characterization of glutathione S-transferases involved in DDT resistance from the mosquito Anopheles gambiae. Pestic. Biochem. Physiol. 47:119- 133. RAYMOND, M., D. G. HECKEL AND J. G. SCOTT. 1989. Interactions between pesticide genes: model and experiment. Genetics 123:543-551. RAYMOND, M., A. CALLAGHAN, P. FORT AND N. PASTEUR. 1991. Worldwide migration of amplified insecticide resistance genes in mosquitoes. Nature 350:151-153. ROSE, R. L., F. GOULD, P. E. LEVI AND E. HODGSON. 1991. Differences in cytochrome P450 activities in tobacco budworm larvae as influenced by resistance to host plant allelochemicals and induction. Comp. Biochem. Physiol. 99B:535-540. ROSSIGNOL, D. P. 1991. Analysis of pyrethroid binding by use of a photoreactive analogue: possible role for GTP-binding proteins in pyrethroid activity. Pestic. Bio- chem. Physiol. 41:103-120. ROUSH, R. T. AND G. L. MILLER. 1986. Considerations for design of insecticide resistance monitoring programs. J. Econ. Entomol. 79:293-298. SCOTT, J. A., F. H. COLLINS AND R. FEYEREISEN. 1994. Diversity of cytochrome P450 genes in the mosquito, Anopheles albimanus. Biochem. Biophys. Res. Comm. 205:1452-1459. SHEMSHEDINI, L. AND T. G. WILSON. 1990. Resistance to juvenile hormone and an insect growth regulator in Drosophila is associated with an altered cytosolic juvenile hormone-binding protein. Proc. Natl. Acad. Sci. USA 87:2072-2076. SNYDER, M. J., E-L. HSU AND R. FEYEREISEN. 1993. Induction of cytochrome P-450 activities by nicotine in the tobacco hornworm, Manduca sexta. J. Chem. Ecol. 19:2903-2916. THOMPSON, M., J. C. STEICHEN AND R. H. FFRENCH-CONSTANT. 1993. Conservation of cyclodiene insecticide resistance-associated mutations in insects. Insect Molec. Biol. 2:149-154. VAUGHAN, A., M. RODRIGUEZ AND J. HEMINGWAY. 1995. The independent gene amplification of indistinguishable esterase B electromorphs in insecticide-resistant Culex quinquefasciatus. Biochem. J. 305:651-658. WANG, J-Y., S. MCCOMMAS AND M. SYVANEN. 1991. Molecular cloning of a glutathione S-transferase overproduced in an insecticide-resistant strain of the housefly (Musca domestica). Mol. Gen. Genet. 227:260-266. WATERS, L. C., A. C. ZELHOF, B. J. SHAW AND L-Y. CHANG. 1992. Possible involvement of the long terminal repeat of transposable element 17.6 in regulating expression of an insecticide resistance-associated P450 gene in Drosophila. Proc. Natl. Acad. Sci. USA 89:4855-4859. WAXMAN, D. J. AND L. AZAROFF. 1992. Phenobarbital induction of cytochrome P-450 gene expression. Biochem. J. 281:577-592. WHEELOCK, G. D. AND J. G. SCOTT. 1992. The role of cytochrome P450 Lpr in delta- methrin metabolism by pyrethroid-resistant and susceptible strains of house flies. Pestic. Biochem. Physiol. 43:67-77. 414 Florida Entomologist 78(3) September, 1995

WHYARD, S., R. J. RUSSELL AND V. K. WALKER. 1994a. Insecticide resistance and malathion carboxylesterase in the sheep blowﬂy, Lucilia cuprina. Biochem. Genet. 32:9-22. WHYARD, S., A. E. DOWNE AND V. K. WALKER. 1994b. Isolation of an esterase confer- ring insecticide resistance in the mosquito Culex tarsalis. Insect Biochem. Molec. Biol. 24:819-827. WILSON, T. G. AND C. TURNER. 1992. Molecular analysis of Methoprene-tolerant, a gene in Drosophila involved in resistance to juvenile hormone analog insect growth regulators, in C. Mullin and J. G. Scott [eds.], Molecular mechanisms of insecticide resistance. American Chemical Society Symp. Series 505:99-112. YU, S. J. 1992a. Plant-allelochemical-adapted glutathione transferases in Lepi- doptera. ACS Symp. Ser. American Chem. Soc. 505:174-190. YU, S. J. 1992b. Detection and biochemical characterization of insecticide resistance in fall armyworm (Lepidoptera: Noctuidae). J. Econ. Entomol. 85:675-682.

♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦

414 Florida Entomologist 78(3) September, 1995

RESISTANCE: A THREAT TO THE INSECTICIDAL CRYSTAL PROTEINS OF BACILLUS THURINGIENSIS

LEAH S. BAUER USDA Forest Service, North Central Forest Experiment Station Pesticide Research Center & Department of Entomology Michigan State University East Lansing, MI 48823-5290

ABSTRACT

Insecticidal crystal proteins (also known as δ-endotoxins) synthesized by the bacterium Bacillus thuringiensis Berliner (Bt) are the active ingredient of various environmentally friendly insecticides that are 1) highly compatible with natural enemies and other nontarget organisms due to narrow host specificity, 2) harmless to vertebrates, 3) biodegradable in the environment, and 4) highly amenable to genetic engineering. The use of transgenic plants expressing Bt δ-endotoxins has the potential to greatly reduce the environmental and health costs associated with the use of conventional insecticides. The complex mode of action of Bt is the subject of intensive research. When eaten by a susceptible insect δ-endotoxin crystals are solubilized in the midgut; proteases then cleave protoxin molecules into activated toxin which binds to receptors on the midgut brush border membrane. Part of the toxin molecule inserts into the membrane causing the midgut cells to leak, swell, and lyse; death results from bacterial septicemia. Insecticides formulated with Bt account for less than 1% of the total insecticides used each year worldwide because of high cost, narrow host range, and comparatively low efficacy. Environmental contamination, food safety concerns, and pest resistance to conventional insecticides have caused a steady increase in demand for Bt-based insecticides. The recent escalation of commercial interest in Bt has resulted in more persistent and efficacious formulations. For example, improved Bt-based insecticides have allowed management of the diamondback moth, Plutella xylostella (L.). Unfortunately this has resulted in the evolution of resistance to δ-endotoxins in P. xylostella populations worldwide. The recent appearance of Bt resistance in the field, corroborated by the results of laboratory selection experiments, demonstrates genetically-based resistance in several species of Lepidoptera, Diptera, and Coleoptera. The genetic capacity to evolve resistance to these toxins is probably

Bauer: Symposium on Pesticide Resistance 415 present in all insects, and the heritability, fitness costs, and stability of the resistance trait are documented in several insect populations. In two strains of Bt-resistant lepidopteran species, mechanisms of resistance involve reductions in the binding of toxin to midgut receptors. Research on other resistant strains suggests that other mechanisms are also involved. Unfortunately, the high stability of the resistance trait, as well as broad spectrum cross-resistance to other δ-endotoxins, undermines many potential options for resistance management. Genetically engineered plants, expressing δ-endotoxin continuously and at ultrahigh doses, ensure intense and rapid selection of the target insect population. The efficacy of transgenic plants can be preserved only by developing an integrated pest management program that is designed specifically to reduce selection pressure by minimizing exposure to Bt and increasing other mortality factors, thereby slowing the rate of pest adaptation to Bt.

Key Words: δ-endotoxin, cross-resistance, transgenic plants, resistance management

RESUMEN

Las proteínas de cristales insecticidas (también conocidas como δ-endotoxinas) sintetizadas por la bacteria Bacillus thuringiensis Berliner (Bt) son el ingrediente ac- tivo de varios insecticidas ambientalmente amistosos que son 1) altamente compati- bles con los enemigos naturales y otros organismos no objetos de control debido a su estrecha especificidad de hospedante, 2) inocuos a los vertebrados, 3) biodegradables en el ambiente, y 4) altamente dóciles para la ingeniería genética. El uso de plantas transgénicas expesando δ-endotoxinas de Bt tiene la posibilidad de reducir grande- mente los costos ambientales y de salud asociados con el uso de insecticidas convencionales. El modo complejo de acción de Bt es sujeto de investigación intensiva. Cuando son ingeridos por un insecto susceptible, los cristales de δ-endotoxina son di- sueltos en el intestino medio; las proteasas abren las moléculas de proteínas transfor- mándolas en toxinas activadas que se unen a receptores en el cepillo de la membrana del intestino medio. Parte de la molécula de la toxina se inserta en la membrana cau- sando que las células del intesitno medio pierdan el control de la permeabilidad, se hinchen y rompan, y la muerte ocurre luego por septicemia bacteriana. Los insecticidas formulados con Bt son menos del 1% del total de los insecticidas usados cada año en todo el mundo debido a su alto costo, estrecho rango de hospedantes, y eficacia com- parativamente baja. La contaminación abiental y la resistencia de las plagas a los insecticidas convencionales han causado un incremento constante en la demanda de insecticidas de Bt. La reciente escalada de interés comercial en Bt ha provocado la aparición de formulaciones más persistentes y eficaces. Sin embargo, el uso intensivo de insecticidas mejorados a base de Bt autorizados para el manejo de la polilla de la col, Plutella xyllostella (L.), ha traído como reusltado la evolución de resistencia a las δ-endotoxinas en las poblaciones de P. xyllostella en todo el mundo. La reciente apar- ción de reistencia a Bt en el campo, corroborada por los resultados de experimentos de selección de laboratorio, demuestra que existe resistencia genética en varias especies de Lepidoptera, Diptera y Coleoptera. La capacidad genética de evolucionar la resistencia hacia esas toxinas está probablemente presente en todos los insectos, y la he- redabilidad, costo de ajuste, y estabilidad de la resistencia son documentados en varias poblaciones de insectos. En dos especies de lepidópteros resistentes a Bt, los mecanismos de resistencia incluyen reducciones en la unión de la toxina a los receptores del intestino medio. La investigación sobre otras especies de insectos resistentes sugiere que otros mecanismos están también relacionados. Desafortunadamente, la alta estabilidad de la resistencia, así como la resistencia cruzada de amplio espectro a otras δ-endotoxinas, determina pocas opciones potenciales para el manejo de la resistencia. Las plantas transgénicas expesando δ-endotoxinas continuamente, y las dosis muy altas, aseguran la selección intensa y rápida de la población de insectos a controlar. La eficacia de las plantas transgénicas puede ser preservada sólo desarro- llando un programa de manejo integrado de plagas diseñado específcamente para re-

416 Florida Entomologist 78(3) September, 1995 ducir la presión de seleccción, minimizando la exposición a Bt e incrementando otros factores de mortalidad, para disminuir la velocidad de adaptación de la plaga a Bt.

The bacterium Bacillus thuringiensis Berliner (Bt) is a complex of subspecies characterized by their ability to synthesize crystalline inclusions during sporulation. These crystalline inclusions are comprised of relatively high quantities of one or more glyco- proteins known as δ-endotoxins or Cry toxins (Table 1). The toxins produced by Bt play a vital role in the pathogenicity of this bacterium to insects and other invertebrates. The Cry toxins have enormous commercial value as safe, biodegradable pesticides. The specificity of Bt toxicity is highly desirable in integrated pest management (IPM) programs, particularly in sensitive aquatic and forest ecosystems where other life forms, including many beneficial and nontarget insects, must be conserved (May 1993). The selective toxicity, rapid environmental degradation, and vertebrate safety of Bt-based insecticides provide growers and the public with environmentally friendly and effective alternatives to conventional insecticides (Meadows 1993). Advances in biotechnology and genetic engineering, as well as the proteinaceous nature of the Cry toxins, led to the selection of the cry genes as the primary insect-resistance genes transferred into, and expressed in, plants and microbes (Gasser & Fraley 1989, Adang 1991, Peferoen 1992, Ely 1993, Gelernter & Schwab 1993). Subspecies of Bt are distributed in a variety of diverse habitats worldwide (Martin & Travers 1989) and are typically isolated from soil, leaf surfaces, and environments rich with insects, such as grain bins and insectaries (Smith & Couche 1991, Burges & Hurst 1977). In fact, the first reports of Bt were from colonies of the silkworm Bombyx mori (L.) (Ishiwata 1901), and from the Mediterranean flour moth, Ephestia kueh-

TABLE 1. BT STRAINS, THEIR RESPECTIVE δ-ENDOTOXINS AND HOST RANGES, IN INSECT RESISTANCE STUDIES.

Bt Strain1 δ-Endotoxin2 Spectrum

kurstaki HD-1 (Btk) CryIA(a), CryIA(b), CryIA(c), CryIIA, Lepidoptera CryIIB Diptera3 kurstaki HD-73 CryIA(c) Lepidoptera aizawai HD-112 (Bta) CryIA(a), CryIA(b), CryIC, CryID, CryIG, CryII4 Lepidoptera aizawai HD-133 (Bta) CryIA(a), CryIA(b), CryIC, CryID Lepidoptera thuringiensis HD-2 CryIA, CryIB Lepidoptera Coleoptera5 entomocidus HD-198 (Bte) CryIA(a), CryIA(b), CryIC, CryID Lepidoptera sotto (Bts) CryIA(a) Lepidoptera israeliensis (Bti) CryIVA, CryIVB, CryIVC, CryIVD, CytA Diptera tenebrionis (Btt) CryIIIA Coleoptera

1Bt strains may contain multiple toxins, and composition may differ slightly from those reported here. 2Organized by amino acid sequence by Höfte & Whiteley (1989). 3CryIIA accounts for the dipteran activity of this strain. 4Presence of CryIG uncertain, and CryII type unknown (McGaughey & Johnson 1994). 5CryIB is toxic to some coleopterans (Bradley et al. 1995). Bauer: Symposium on Pesticide Resistance 417 niella (Zeller) (Berliner 1911). Ecological considerations of Bt as an insect pathogen and in the environment are discussed by Meadows (1993). In the United States, commercial development of Bt into a formulated insecticide did not begin until the late 1950s. Most Bt-based insecticides are formulated mixtures of δ-endotoxin crystals and Bt spores, which are known to synergize the toxicity of the crystals. Although the effectiveness of these early Bt-based insecticides was often erratic, progress was slow in research and development of improved Bt formulation, delivery, and application technologies, as well as in the discovery of more active strains. Until the mid-1970s, it was generally accepted that lepidopterans were the only target of Bt. The discovery of Bt subsp. israelensis, which is toxic to larval mosquitoes and black flies (Goldberg & Margalit 1977), and the discovery of Bt subsp. tenebrionis (Krieg et al. 1983), which is toxic to several beetle species, stimulated sudden and dramatic commercial interest in Bt. During the 1980s, new biotech companies and large agro- chemical and pharmaceutical corporations initiated research programs to isolate Bt from various environmental samples and to screen for toxicity in agriculturally and medically-important target organisms (Van Frankenhuyzen 1993). Lambert & Pefer- oen (1992) estimated that 40,000 strains of Bt are now stored, mainly in private collections, worldwide. The spectrum of activity of Bt toxins has expanded from species in three insect orders (Lepidoptera, Diptera, and Coleoptera) to species in eight insect orders (Homoptera, Orthoptera, Mallophaga, Hymenoptera, Siphanoptera) (Bauer, unpublished) and various mites, nematodes, flukes, mollusks, and protozoans (Feitel- son et al. 1992). Commercial products formulated with Bt are now registered for control of lepidopteran (Navon 1993), dipteran (Becker & Margalit 1993), and coleopteran pests (Keller & Langenbruch 1993). The short half-life of Bt, due to ultraviolet inactivation when topically applied, has stimulated considerable research into alternative delivery strategies. By far the most controversial strategy is the use of insect-resistant crops expressing Bt δ-endotoxin genes, which are already in the field in Asia and the United States where potato, cotton and corn are registered. As concerns over environmental quality and food safety increase, Bt-based insecticides will become increasingly important in the development of IPM strategies. Cur- rently, the use of Bt in insect control programs accounts for less than 1% of insecticides used worldwide each year. The comparatively high production cost of Bt- based insecticides is a primary impediment to more widespread usage. However, the major impetus for greater use of Bt in agriculture is the development of resistance to conventional insecticides (Georghiou 1994, Watkinson 1994). In fact, many growers typically add Bt to conventional sprays because of concerns about chemical control failure (Marrone & MacIntosh 1993). Because of their environmental safety, microbial insecticides are one of the few pesticides that can be developed and registered quickly and cheaply. In addition, resistance to conventional insecticides does not confer cross- resistance to Bt toxins due to the unique mode of action of δ-endotoxin (Stone et al. 1991, Tabashnik 1994a). Resistance is a major problem associated with the intensive use of pesticides in agriculture and human health protection, and hundreds of insect and mite species can no longer be controlled by one or more pesticides (Georgiou & Lagunes 1988). Resis- tance is documented in diverse groups of insecticides, including neurotoxins, chitin synthesis inhibitors, juvenile hormone analogues (National Research Council 1986), and, most recently, Bt (Tabashnik et al. 1990). Adaptation to individual insecticides is a consequence of their intensive and prophylactic use, and many conventional insecticides are being lost to resistance faster than industry can replace them. Many researchers now predict that the use of Bt cry genes in the genetic engineering of insect- 418 Florida Entomologist 78(3) September, 1995 resistant plants will expedite selection for resistance in the target organisms more rapidly than has occurred through conventional application methods (Gould 1988a, 1988b, Van Rie 1991, McGaughey & Whalon 1992, Marrone & MacIntosh 1993, May 1993, Whalon & McGaughey 1993, McGaughey 1994, Tabashnik 1994a). Our understanding of the nature of Bt insecticidal crystal proteins has advanced rapidly since the mid-1980s, thanks to highly collaborative research programs involving entomologists, microbiologists, physiologists, geneticists, protein biochemists, and molecular biologists. Research on Bt is fast-paced, often involving worldwide liai- sons between researchers in industry, university, and government, leading to an abundance of new research results, methodologies, and discoveries. In this paper, my objectives are to provide researchers, even those only peripherally involved or knowledgeable of Bt δ-endotoxins, with a greater understanding of 1) Bt mode of action, 2) resistance and cross-resistance to Cry toxins, and 3) resistance management, as they relate to the use of both improved Bt insecticides and genetically-engineered plants expressing δ-endotoxin genes. The research results presented here will compliment and update those reported in the comprehensive review by Tabashnik (1994a). In presenting this information, I hope to attract more researchers into the complex area of IPM and management of Bt resistance. The design, validation, and implementation of effective and sound IPM practices are some of the greatest challenges facing researchers in agriculture, agroforestry, and vector control today.

BT MODE OF ACTION

A generalized ﬂow chart of the events leading to Bt intoxication by Cry toxins in a susceptible host reveals various levels at which resistance might evolve in an insect population (Fig. 1). The high degree of host speciﬁcity, as well as the complexity of Bt mode of action, results from the interaction of the toxin within the complex environment of the insect’s midgut lumen and on the surface of the midgut epithelial cells (English & Slatin 1992). Although researchers discovered relatively early that the midgut was the primary site of δ-endotoxin activity (Heimpel & Angus 1959), the molecular mechanisms of Bt intoxication continue to be the subject of intensive research (for reviews see Gill et al. 1992, English & Slatin 1992, Lambert & Peferoen 1992, Aronson 1993, Honée & Visser 1993, Knowles & Dow 1993, Yamamoto & Powell 1993, Federici 1993, Visser et al. 1993). A small family of Bt δ-endotoxins known as the cytolytic or Cyt toxins, an important component of Bt subsp. israelensis (Table 1), is not covered in this review (see Chilcott et al. 1990, Koni & Ellar 1994, Wu et al. 1994).

Ingestion

Feeding stimulants are known to greatly enhance Bt performance since most susceptible insects stop feeding after consuming food treated with δ-endotoxin. Detection and behavioral avoidance of food treated with Cry toxins have also been reported for many target species (Gould & Anderson 1991, Gould et al. 1991, Ramachandran et al. 1993). Formulation and application technology has improved the rate of toxin ingestion, increasing the probability that the target insect will consume a lethal dose after treatment.

Crystal Solubilization

Following ingestion, solubilization of crystalline δ-endotoxin is a prerequisite to all subsequent events in the intoxication pathway (Tojo & Aizawai 1983, Du et al. 1994). Bauer: Symposium on Pesticide Resistance 419

High midgut pH (>9.5) was once thought to be essential to crystal solubility, but coleopteran-speciﬁc toxins were found to function at much lower pH (Koller et al. 1992). Midgut detergency and redox potential also affect solubilization. The rate and extent of crystal solubilization greatly inﬂuence toxicity levels in different hosts (Aronson et al. 1991, Bradley et al. 1995), as well as the rate of intoxication (Bauer & Pankratz 1992, Koller et al. 1992).

Enzymatic Processing

The Cry proteins are synthesized as protoxins that require processing by midgut enzymes to generate activated toxin (Ogiwara et al. 1992). The larger protoxins of about 130-140 kDa (e.g. CryI) are proteolytically cleaved, exposing the activated toxin which is a protease-resistant core of about 55-65 kDa (Höfte & Whiteley 1989). Many other toxins (e.g. CryII, CryIII, CryIVD) are synthesized as 70 kDa proteins and are similar to the N-terminal half of the larger protoxins. Enzymatic processing of these smaller 70 kDa toxins also occurs, with amino acids cleaved from the N terminus (Carroll et al. 1989).

Receptor Binding

The action of Cry δ-endotoxins on the midgut epithelium begins with binding of the activated toxin to receptors (Hoffman et al. 1988a, 1988b). Much of the host specificity of Bt toxins results from their ability to bind to specific receptors on the brush border membrane, although most toxins bind to more than one receptor (Van Rie et al. 1989, 1990a, Denolf et al. 1993, Estada & Ferré 1994). Amino acid sequence similarity in the receptor-binding domain of the toxin molecule is a useful predictor of overall host specificity (Van Rie et al. 1990, Cummings & Ellar 1994). The binding domain is also the most variable region of the toxin molecule (Li et al. 1991). The function of these receptors in midgut physiology is an elusive research ques- tion. Recently, aminopeptidase N, a 120 kDa glycoprotein, was purified from the lepidopteran Manduca sexta and identified as the receptor for CryIA(c) (Knight et al. 1994, Sangadala et al. 1994). Aminopeptidase N is an abundant Zn+-dependent ec- toenzyme present in the brush border of membranes of the alimentary tracts of most animals (Ellar 1994, Garczinski & Adang in press). Binding, while essential, is not sufficient to produce mortal damage, as shown by several studies that found specific binding of toxins to receptors on brush border preparations is not correlated with in vivo toxicity (Van Rie 1990a, Wolfersberger 1990, Ferré et al. 1991, Garczynski et al. 1991, Gould et al. 1992, Escriche et al. 1994, San- chis et al. 1994, Masson et al. 1995). Other researchers showed that in vivo toxicity is, however, strongly correlated with measures of membrane disruption (Wolfersberger 1991) or membrane permeability (Carroll & Ellar 1993). They stressed that, although receptors play an essential role, post-binding factors are required for successful intoxication by Bt δ-endotoxins.

Intercalation, Pore Formation, and Cell Lysis

After binding to a receptor on the cell surface, the toxin then inserts or intercalates into the plasma membrane (English & Slatin 1992, Knowles & Dow 1993). Evidence from electrical conductance and ion leakage studies suggests that several toxin/receptor complexes aggregate to form lesions or leaky regions in the brush border membrane (Walters et al. 1993, 1994). Using a very different method, Masson et al. (1995) 420 Florida Entomologist 78(3) September, 1995

Figure 1. Generalized ﬂow chart of events leading to Bt intoxication in a susceptible host. Bauer: Symposium on Pesticide Resistance 421 also detected toxin-toxin aggregation, supporting evidence that the toxin acts as a multimer on the cell surface. Pores in the plasma membrane are estimated at 1-2 nm diam, disrupting the actively maintained osmotic balance, causing the cells to swell and burst by a process known as colloid-osmotic lysis (Knowles & Ellar 1987). Carroll & Ellar (1993) demonstrated that δ-endotoxin-induced changes in cell permeability is non-selective to cations, anions, neutral solutes, and water. A three-dimensional model of the CryIIIA protein structure supports the hypothesis that the toxin causes pores or channels to form in the lipid bilayer (Li et al. 1991). However, the role of midgut receptors in this toxin-induced leakage is still unclear (Knowles & Dow 1993, Parenti et al. 1995, Garczynski & Adang, in press).

Bacterial Septicemia and Death The synergistic effect of Bt spores, in the presence of δ-endotoxin, on insect mortality leads to speculation that δ-endotoxins facilitate bacterial exploitation of the nu- trients present in the insect hemolymph (Ali et al. 1985, Wilson & Benoit 1990, Borgonie 1995). Death occurs when lysis of midgut cells causes irreparable break- down of the midgut integrity, allowing Bt and other bacteria present in the lumen to gain access to the body cavity. The insect hemolymph provides an excellent medium for bacterial growth. Death caused by bacterial septicemia usually occurs within 2-3 days post-ingestion. However, the immediate cessation of feeding observed in most insects after ingestion of Bt (Angus 1954), as well as the rapid regenerative capability of midgut epithelial cells, can allow damaged regions of the midgut to heal. The actual recovery of treated insects is dependent on many intrinsic and extrinsic factors including host genetics, age, and vigor; dosage and potency of toxin ingested; various environmental factors including host plant species (Meade & Hare 1994, Moldenke et al. 1994); and the presence of Bt spores and other bacteria in the insect gut (Miyasono et al. 1994).

RESISTANCE

Over several decades of commercial use, the continued efficacy of Bt-based insecticides led to considerable skepticism that resistance to Bt was possible (Burges 1971, Krieg & Langenbruch 1981). However, recent field and laboratory evidence suggest otherwise. The slow development of field resistance in the past may have resulted from low selection pressure exerted by early formulations and usage patterns (Stone et al. 1991). Other researchers believed that the complex mode of action of Bt, often involving multiple toxins and Bt spores, provided protection against resistance because a single mutation in the insect would be unlikely to affect susceptibility (Boman 1981, Briese 1981, de Barjac 1987). McGaughey & Whalon (1992), however, suggested that at high levels of selection, the multicomponent-toxicity pathway merely expands behavioral and/or physiological opportunities for adaptation to Bt. Technological ad- vancements in Bt toxicity, host range, stability, formulation, application, and ultimately the expression in transgenic plants, are greatly improving biopesticide potency and efficacy (Feitelson et al. 1992, Stone & Sims 1993, Carlton & Gawron- Burke 1993). Unfortunately, while providing high levels of pest suppression, improved efficacy will rapidly help select for the segment of the population that is capable of withstanding Bt intoxication. Possible shifts in susceptibility to Bt were first reported by Kinsinger & Mc- Gaughey (1979), with a 42-fold difference among “natural” populations of Plodia interpunctella (Hübner), the Indianmeal moth, and up to a 15-fold difference in Cadra cautella (Walker), the almond moth. However, the underlying cause of this variation 422 Florida Entomologist 78(3) September, 1995 is unknown because the history of Bt applications in these grain storage facilities was not reported. In a subsequent study over a five-state area, the mean median lethal concentration (LC50) for populations of P. interpunctella from grain bins treated with

Bt was 1.2-fold higher than the mean LC50 for populations from untreated bins (Mc- Gaughey 1985). When these Bt-exposed populations were selected further in the laboratory, 30-fold resistance developed in 2 generations, and 100-fold resistance in 15 generations. Initially, resistance to Bt in the Indianmeal moth was considered somewhat unique because exposure to natural infestations of Bt in stored grains may increase the genetic variability in Bt susceptibility (Kinsinger & McGaughey 1979). In addition, the dark, stable and closed environment of grain bins favored selection for resistance by long toxin residual times because of no UV exposure and by minimizing the potential of outbreeding with susceptibles from other populations. It was not until Tabashnik et al. (1990) first reported field resistance to Bt in Ha- waiian populations of Plutella xylostella (L.), the diamondback moth, that the potential for widespread resistance to Bt was generally acknowledged. Resistant populations of P. xylostella have also been documented in field populations from Flor- ida (Jansson & Lecrone 1990), New York (Shelton et al. 1993), the Philippines (Kirsch & Schmutterer 1988, Ferré et al. 1991), Japan (Hama et al. 1992), Thailand, and Ma- laysia (Georghiou 1994). These recent reports of resistance to Bt in the field have provided credibility to the results of laboratory selections for Bt resistance (for reviews, see Briese 1981, Georghiou 1990, Stone et al. 1991, McGaughey & Whalon 1992, Tabashnik 1994a). As laboratory and field data accumulate, concern is growing that these unique bacterial toxins may be rendered useless as pest management tools, particularly with the imminent commercialization of transgenic plants expressing single activated toxic fragments (May 1993, Whalon & McGaughey 1993).

Laboratory Selections The increased effort researchers are now devoting to Bt resistance is reflected in the many experiments designed to select for Bt resistance in the laboratory (Table 2). The results summarized in Table 2 represent selection experiments, performed and ongoing during the last decade, which achieved significant levels of resistance to Bt preparations containing either a mixture of spores and native δ-endotoxin crystals, or individual Cry toxins in various forms. These results represent only a few of the more than 50 laboratory selection experiments performed with at least 16 insect species (Tabashnik 1994a). Significant levels of resistance have been documented in nine species of Lepidoptera and two species each of Diptera and Coleoptera, with resistance ratios ranging from 1.1 to >1000. Concern about rapid adaptation of insects to δ-endotoxin-based transgenic plants has led to a steady increase in the number of researchers performing laboratory selection experiments with single toxins and in some cases with the specific gene products that insects will ingest when consuming the foliage from bioengineered plants (Estada & Ferré 1994). Researchers found that resistance ratios were consistently higher, and resistance developed more rapidly, in insect populations selected with individual toxins than in populations selected with Bt-insecticides that contain live spores and multi-toxin crystals, such as Bt subsp. kurstaki (Btk) (Tables 1 and 2). The results of these experiments suggest that deployment of the high dose, single toxin strategy in the design of transgenic plants will quickly generate resistant populations of the target pest. The selection experiment reported by Moar et al. (1994) best illustrates the differential response of Spodoptera exigua populations subjected to selection pressure from either Btk spore/crystal preparations or purified CryIC toxin (Table 2). After 20 gen- Bauer: Symposium on Pesticide Resistance 423 erations, larvae selected with Btk spore/crystal preparations were only 3- to 4-fold resistant. However, the cohort selected with pure CryIC protoxin was 1000-fold resistant after 21 generations. This confirms that the complex mode of action, present in most commercially Bt-based insecticides available today, is important in the low prevalence of field resistance for pest populations treated with Bt. It may also explain the lack of Bt-resistance generated in laboratory experiments which used crude or formulated Bt preparations containing both δ-endotoxins and spores for selections (Tabashnik 1994a). Before 1985, large quantities of a single, pure Cry toxin, made possible through advances in molecular biology, were simply not available. Although laboratory-generated resistance within a susceptible species may not reflect the rates and mechanisms generated in field populations, laboratory populations are invaluable tools for the study of potential risk of resistance, physiological and behavioral mechanisms of resistance, cross-resistance, genetics, stability, fitness costs, and, ultimately, for the development of methods for monitoring, managing, and delaying resistance.

Heritability It is apparent that the genetic capacity to evolve resistance to Bt δ-endotoxin is widespread in insects. Although some efforts to select for resistance have failed, this may reflect either insufficient selection pressure, a genetic bottleneck due to lack or loss of genetic diversity in the laboratory colony, or both (McGaughey & Whalon 1992, Whalon & McGaughey 1993). Selection experiments often provide the necessary information to estimate heritability (h2), the proportion of the observed variability that is caused by additive genetic variation (Falconer 1989). Estimated heritability was used by Tabashnik (1992, 1994a) to estimate the ability of populations to develop resistance in 27 selection experiments. He showed that P. interpunctella has a relatively high h2 compared to other moths. This reflects low phenotypic variation, perhaps resulting from its stable environment, and high additive genetic variation for the resistance trait, perhaps resulting from exposure to Bt in its environment. Tabashnik (1994a) also discussed the potential usefulness and limitation of the heritability estimates in assessing resistance risk, i.e., predicting the rate at which a pest will evolve resistance (Tabashnik 1992, Keiding 1986, Tabashnik & McGaughey 1994). Plodia interpunctella has a high estimated h2 and adapts readily to Bt in the laboratory, but high levels of field resistance are unknown. McGaughey (1985) suggested that the infrequent applications of Bt in grain bins, as well as Bt’s limited efficacy, helped preserve the susceptible individuals within the treated population, thereby generating only low levels of resistance. In contrast, susceptible populations of P. xylostella with a comparatively low h2 do not achieve significant levels of resistance to Bt in laboratory selections (Devriendt & Martouret 1976, Krieg & Langenbruch 1981). However, moderately resistant field populations of P. xylostella quickly reach high levels of resistance during laboratory selections (Tabashnik et al. 1991). This shows that selection is occurring in pest populations that are being intensively managed with Bt-based insecticides. Similar results of rapid laboratory adaptation to Bt after intensive field exposure were found in populations of mosquitoes (Gill et al. 1992), P. interpunctella (McGaughey & Johnson 1992), and Leptinotarsa decemlineata (Say) (Whalon et al. 1993) (Table 2).

Intraspeciﬁc Susceptibility

In most species, phenotypic variations in Bt tolerance have a strong genetic basis, and species with high variability in this trait will develop resistance more quickly un- 424 Florida Entomologist 78(3) September, 1995 Bt Reference(s) . 1 3 NDOTOXINS Resistance Ratios 10,000 Gould et al. in press -E δ

T spores, and many were cloned expressed in acrystalliferous strains of B

4 Bt WITH

Generations Selected XPERIMENTS E ELECTION Populations Selected 2 14 10-50 Müller et al. 1994 6 12-22 8-11311 McGaughey & Beeman 1988 20 8 3 1.4 Moar et al. 1994 Salama & Matter 1991 11615 313 24 11 5-9 3-8 1.7 15-66 McGaughey & Beeman 1988 et al. 1991 Tabashnik Brewer 1991 S ﬁeld applications for 1 to 2 years. prior to selection with CryIVD. Btt 2 Bti of the susceptible colony. Range reﬂects combined results of more than one resistant population. of the susceptible colony. . ABORATORY 5 50 Bta Bti L HD-112 HD-133 1 1 na na 28 94 1992, 1994 HD-1 1 + na 305 McGaughey & Johnson HD-198 1 na 32 Strain or -Endotoxin Bt δ Btk Bta Bta Bte Btk CryICCryIC 1 32 >1000 CryIA(b)CryIA(c)CryIA(c) 1 1 1 18 17 >19 episodes 65 50 Stone et al. 1989, Sims & 1991 Gould et al. 1992 IGNIFICANT S of resistant colony/LC OF 50

T spores, and many were cloned expressed in acrystalliferous strains of B

Bt WITH

Generations Selected 32232 70 15 >900 Gill et al. 1992 Whalon et al. 1993 XPERIMENTS E 7 8 9 ELECTION Populations Selected 1131 7 81 14 8 31 3.8 1.1 1.8 Estada & Ferré 1994 van Frankenhuyzen et al. 1995 Goldman et al. 1986 Saleh 1987 S ﬁeld applications for 1 to 2 years. prior to selection with CryIVD. Btt 2 Bti 6 of the susceptible colony. Range reﬂects combined results of more than one resistant population. of the susceptible colony. . ABORATORY 50 Bti L Strain or -Endotoxin CryIVDCryIIIA 1 8 Bt δ CryIIIA 1 35 >3000 Bauer et al. 1994 IGNIFICANT S of resistant colony/LC OF 50

UMMARY 2. S preparations may contain spores and one or more toxins; Cry usually do not or other species of bacteria. Includes unpublished data. Summarized, in part, from Tabashnik (1994a). Summarized, in part, from Tabashnik Bt Resistance ratio = LC Selections were not performed every generation. Larvae were selected with CryIC crystalline protoxin through generation 14. Continued selections done toxin (requiring activation trypsin). Activated toxin. Laboratory population exposed for 98 generations to Field population with no prior exposure to Laboratory population included individuals surviving 1 2 3 4 5 6 7 8 9 10 ABLE Choristoneura fumiferanaDIPTERA Bts Aedes aegyptiCulex quinquefasciatus COLEOPTERA Leptinotarsa decemlineata Bti Species niTrichoplusia CryIA(b) Chrysomela scripta T 426 Florida Entomologist 78(3) September, 1995 der intensive selection pressure (Tabashnik 1994a). Baseline data on target pest sensitivity to Bt are essential in assessing the risk of resistance. Tabashnik (1994a) estimated variations in Bt-susceptibility among populations of 15 insect species within three orders and found minimal variation except in three species of moths. Variation in two stored product moths (up to 42-fold) was not associated with Bt treatments, but presumably resulted from natural exposure to Bt in grain bins (Kinsinger & McGaughey 1979). Variation in the third moth, P. xylostella, however, was attribut- able to repeated exposure to Bt foliar sprays. Tabashnik et al. (1990) determined that the susceptibility of intensively treated populations varied up to 40-fold, whereas the susceptibility for populations receiving minimal Bt exposure in the field and in laboratory colonies varied no more than 7-fold. Increasing reliance on Btk-based products for the suppression of spruce budworm, Choristoneura fumiferana (Clemens), an important defoliator of coniferous forests in North America, stimulated interest in assessing the risk of field resistance (van Fran- kenhuyzen et al. 1995). Studies on the variation in Bt tolerance within and among nine spruce budworm field populations from Ontario with no previous Bt exposure re- veal substantial familial differences in sensitivity to Bt, whereas differences between populations are minimal. This finding suggests that Bt tolerance is genetically based, and that spruce budworm has the potential to adapt to Bt. One laboratory selection experiment (Table 2) illustrates an increase in frequency of the resistance trait (van Frankenhuyzen et al. 1995). To gain baseline data on the Bt susceptibility of two pests targeted by transgenic cotton, Heliothis virescens (F.) and Helicoverpa zea (Boddie), Stone & Sims (1993) bioassayed populations from 14 states with activated CryIA(c) toxin and a commercial Btk-based insecticide. Although previous exposure to Bt was not identified, significant differences within populations of both species were detected. Moreover, the variability in Bt tolerance was consistently higher for the activated toxin in both species. This again suggests that an individual toxin as expressed in transgenic cotton would stimulate adaptation more rapidly than the commercial preparations that contain a mixture of spores and crystals. Stone & Sims (1993) acknowledged that monitoring and managing for resistance will be an exciting challenge when insect-resistant cotton, expressing Bt δ-endotoxin, is deployed.

Mechanisms

An understanding of the mechanisms of resistance to Bt toxins will prove essential in the future design and management of transgenic plants containing Bt toxin genes. At present, the primary mechanism of resistance reported for P. interpunctella (Van Rie et al. 1990b) and P. xylostella (Ferré et al. 1991) is a reduction in the binding of toxin to receptors on the midgut brush border membrane. This is the same mechanism known to account for much of the host specificity to various δ-endotoxins in Lep- idoptera (Hoffman et al. 1988b). Plodia interpunctella, selected for resistance to Btk (see Table 1), shows a 50-fold reduction in midgut brushborder membrane receptor-binding affinity to CryIA(b) (Van Rie et al. 1990b). However, in vivo toxicity of CryIC (not present in Btk) increased when bioassayed in the Btk-resistant P. interpunctella. This increase in toxicity resulted from an increase in the CryIC binding sites. These results show that at least two distinct molecular changes occurred in the midgut receptor population in response to Bt selection. Although the role of these receptor molecules in the insect midgut are, as yet, poorly characterized, Van Rie et al. (1990b) suggest the increase in one receptor population may compensate for diminished function in the other. In a strain Bauer: Symposium on Pesticide Resistance 427 of P. interpunctella selected for resistance to Bt subsp. entomocidus (Bte), Oppert et al. (1994) reported reduced proteolytic activation of CryI(A)c. Using similar methodologies, resistance mechanisms studies of Btk-resistant field populations of P. xylostella demonstrated greatly reduced or lack of toxin binding to midgut receptors, suggesting a change or complete loss of the receptor (Ferré et al. 1991, Tabashnik et al. 1994). Using a different method for quantifying receptor binding, Masson et al. (1995) reported a loss of receptors in a strain of resistant P. xylostella, although not adequate to explain the high level of resistance. The results from studies of other resistant insect species suggest that factors other than receptor binding can contribute to resistance. For example, two separate studies of resistance in H. virescens found no relation between resistance to CryIA(b) or CryIA(c) and toxin-receptor binding (MacIntosh et al. 1991, Gould et al. 1992). In addition, studies of Trichoplusia ni (Hübner) determined that CryIA(b) and CryIA(c) share the same receptor, but a strain of T. ni selected for resistance to CryIA(b) showed cross-resistance to CryIA(c) (Estada and Ferré 1994). Resistance mechanisms in these insects may involve changes in post-binding events such as channel formation, leakage, and repair rate.

Cross-resistance The apparent specificity, diversity, and genetic versatility of Bt δ-endotoxins suggest that resistance might be managed by deploying toxins in mixtures or sequences (Georghiou 1990, Stone et al. 1991, Van Rie 1991). Although at least 12 lepidopteran- active δ-endotoxins are available (Adang 1991), evidence is mounting that selection for resistance to one or more δ-endotoxins causes resistance to others (Tables 3 and 4). This phenomenon, known as cross-resistance, typically occurs when mechanisms of toxicity are similar. Many of the resistant insect populations generated in the laboratory and the field were selected with Bt-based insecticides containing multiple toxins; the most studied products are formulated with Btk strain HD-1, which is comprised of live spores and a mixture of five toxins (Table 1). In many studies, cross-resistance is referred to as an increase in tolerance of a population, selected with one Bt isolate, to an isolate containing a different mixture of toxins. For example, efforts to find other Bt isolates to control P. interpunctella resistant to Btk (140-fold) in grain bins showed that these insects were also resistant to 32 of the 57 Bt isolates assayed (McGaughey & Johnson 1987). Resistance was highest among various Btk isolates, suggesting some degree of specificity. Many Bt isolates overlap considerably in their δ-endotoxin composition. In the context of the following discussion, cross-resistance is defined as an increase in the tolerance of a population to a toxin absent in the preparation used for selection. Resistance, on the other hand, refers to increasing tolerance to a toxin that is present in the Bt isolate used in selection. Strains of P. interpunctella, selected for resistance to Btk, Bte, Bt subsp. aizawai (Bta) strain HD-112, Bta strain HD-133, or a mixture of Btk and Bta HD-133, showed some level of resistance and cross-resistance to six δ-endotoxins tested (McGaughey & Johnson 1994) (Table 3). Btk tended to select for high levels of resistance to the entire complex of CryIA toxins which are 82 to 90% homologous in their amino acid sequences (Höfte & Whitley 1989). Resistance was highest to CryIA(b) and CryIA(c). Evidence suggesting that they share the same binding site (Wolfersberger 1990) is also supported by the high levels of cross-resistance to CryIA(c) in populations selected with Bta HD133 and Bte, which is likely derived from the presence of CryIA(b). The specificity of the target receptor appears to be greater for CryIA(c) than for CryIA(b) because cross-resistance is significantly greater than the selected resis- 428 Florida Entomologist 78(3) September, 1995 5 y 24 20 XYLOSTELLA P.

AND 4

na 1,2 5na 32 na 10 1419 na na 5 INTERPUNCTELLA -Endotoxins P. δ

5 1 >240 17 13 2 . NDOTOXINS OXINS -E δ T isolate used in selection (see Table 1). isolate used in selection (see Table

Bt 27 7 789150 44 9 ULTIPLE M NDIVIDUAL I Resistance Ratios for Individual TO

ONTAIN C >750 >6,800 HAT T ESISTANCE -R 6 ROSS SOLATES I C T B CryIA(a) CryIA(b) CryIA(c) CryIB CryIC CryIF CryIIA

of the susceptible colony. AND 50 TO

) 2 ESISTANCE ESISTANCE R R 3 OF of resistant colony/LC FOR

50 5 HD-133 4 253 2,267 Bta ATTERNS ELECTED S (Resistance Ratio Strain Used in Selection 3. P Bt HD-112 (28) HD-112 HD-133 (94) 4 17 24 226 HD-1 + HD-1 (70) HD-1 (3,300) 6 263 y 2,816 HD-198 (32) 10 27 (164) and (100) Resistance ratios printed in bold type represent cross-resistance (i.e. toxins are not present the Resistance ratio = LC Results from McGaughey & Johnson (1994). na, data not available. et al. (1993, 1994b). Results compiled from Tabashnik significant level of resistance, but ratio was not reported. y, 1 2 3 4 5 6 ABLE Bte Btk Plutella xylostella Bta Bta Plodia interpunctella Btk Btk T Bauer: Symposium on Pesticide Resistance 429 tance. Cross-resistance to CryIB and CryIC was significant, although very low, which is consistent with their relatively low amino acid sequence similarity to the CryIA toxins (ranging from 55 to 67%) (Tabashnik et al. 1994b). As expected, Bt isolates containing more dissimilar toxins, such as found in Bta and Bte, selected for a broader spectrum of cross-resistance in the treated insects than did isolates producing similar Cry toxins such as Btk HD-1. The complexity in the pattern of resistance results, in part, from differential selective pressure exerted by different levels of toxicity and the amounts of each toxin in each Bt strain. Overall, the results of this selection experiment clearly contradict the claim that toxin mixtures will prevent or retard the development of resistance (Tabashnik & McGaughey 1994). Field-resistant P. xylostella from Hawaii, further selected with Btk in the laboratory, show a similar pattern of broad, but highly variable, resistance and cross-resistance among the Cry toxins tested (Tabashnik et al. 1993, 1994b) (Table 3). Again, resistance was highest to the CryIA toxins, and cross-resistance to CryIB and CryIC was low. Intermediate cross-resistance to CryIF is supported by its amino acid sequence homology (70 to 72% to the CryIA toxins). A different field population of P. xylostella from the Philippines, also resistant to CryIA(b) (resistance ratio 236), showed no resistance to Btk, CryIA(a), CryIA(c) and no cross-resistance to CryIB, and CryIC (Ferré et al. 1991, Ballester et al., in press). These researchers determined that CryIA(b) has a single binding site that is also rec- ognized by CryIA(a) and CryIA(c). Loss or modification of the CryIA(b) binding site results in the loss of its toxicity. Ballester et al. (in press) noted that this narrow spectrum of resistance is somewhat unique and suggested that this population may represent a biotype present in the Phillipines, not a case of field selection by Btk. To understand the complexity of these emerging cross-resistance patterns, we must select insect populations with individual toxins. In the case of most Bt isolates, which produce several toxins, this requires cloning and transferring the cry gene into an acrystalliferous strain of Bt, or into another species of bacteria, such as Escheri- chia coli or Pseudomonas fluorescens. In some studies, the gene is modified to simulate the specific Cry products being expressed in transgenic plants or microorganisms. The first such study that also included cross-resistance data was the selection of H. virescens with purified CryIA(c) (Gould et al. 1992) (Table 4). After 17 generations, the population was 50-fold resistant to CryIA(c) and, as reported for other lepidopterans, was 13-fold cross-resistant to CryIA(b). This population, however, was also cross- resistant to CryIIA, CryIB, and CryIC. In a subsequent paper, Gould et al. (in press) also reported high levels of cross-resistance to CryIF. No measurable differences in the concentration of CryIA(c), or CryIA(b) binding sites or binding affinities, were detected between selected and unselected H. virescens, suggesting other mechanisms of resistance are involved. Similar broad-spectrum cross-resistance was also reported for two species of Spodoptera and two coleopterans, Chrysomela scripta F. (Table 4) and L. decemlineata (Whalon, unpublished results). In contrast, cross-resistance in Trichoplusia ni (Hübner) selected with CryIA(b) had a higher degree of specificity within the CryIA group of toxins (Estada & Ferre 1994). This specificity was similar to that reported for P. xylostella from the Phil- lipines (Ballester et al. in press), although no data were presented on unrelated toxins for T. ni. As determined for several other lepidopterans, receptor-binding assays re- veal that CryIA(b) and CryIA(c) share the same high-affinity binding sites, and resistance to one would imply cross-resistance to the other. However, no cross-resistance to CryIA(c) was detected, suggesting CryIA(c) toxicity results from other binding sites or alternative mechanisms. Overall, cross-resistance patterns and their underlying physiological mechanism are very complex and somewhat unpredictable, even within a closely related group of 430 Florida Entomologist 78(3) September, 1995 . NDOTOXIN -E δ

na 12 73 2,3 URIFIED P 8 NE na na na 53 O -Endotoxins δ WITH

5 ELECTED S or other species of bacteria. Bt NSECTS I IN

Cross-Resistance Ratios for Speciﬁc Cross-Resistance Ratios for Speciﬁc NDOTOXINS -E δ

13 — y y na ns ns 328 na na na na na ns ESISTANCE -R 10 4 na 2,364 — y 2.5 na na 3,678 na 15 of the susceptible colony. 50 ROSS C ON

spores, and many were cloned produced in acrystalliferous strains of TUDIES Bt 6 S of resistant colony/LC OF 9 50

7 ) CryIA(a) CryIA(b) CryIA(c) CryIB CryIC CryID CryIE CryIF CryIH CryIIA 3 Used in Selection 2 UMMARY 1 4. S -endotoxins do not contain ns, not susceptible. See references listed in Table 2. See references listed in Table δ Resistance ratio = LC signiﬁcant level of resistance, but ratio was not reported. y, na, data not available. Cross-resistance bioassays performed over time as selections continued and resistance to CryIA(c) was increasing (Gould et al. in press). personal communication. J. Moar, W. CryIE/CryIC fusion protein. Trypsinized unpublished. Bauer, 1 2 3 4 5 6 7 8 9 10 CryIA(c) (50) y CryIC (>1000)CryIC (50)CryIA(b) (31) na (>3000)CryIIIA >93 na na 0 na ns na — na — 0 na na — na 80 na y na y na na na na na na na CryIA(c) (558-10,000) ABLE -Endotoxin Heliothis virescens Species Spodoptera exigua δ Spodoptera littoralis ni Trichoplusia Chrysomela scripta (Resistance Ratio T Bauer: Symposium on Pesticide Resistance 431 toxins and susceptible insects. A more complete understanding of how each toxin interacts within a particular target species at the molecular level is critical to selecting δ-endotoxins for design of transgenic plants that do not favor broad-spectrum cross- resistance.

Adaptation to δ-Endotoxin

The recent appearance of resistance to conventional Bt-based insecticides was correlated with improved formulation and more intensive usage patterns (Tabashnik et al. 1990). In addition to operational uses of the Bt toxins, pest genetics, behavior, physiology, and ecology are critical factors in predicting resistance risk. The best studied examples of Bt-resistance, P. interpunctella and P. xylostella, share many attributes that leave their populations particularly vulnerable to rapid selection (Table 5). These attributes include multivoltinism, short generation time, and populations that tend to be isolated and stable. These attributes are typical of many agricultural and medical pests, and they temper enthusiasm for bioengineered plants which produce continuous and high levels of δ-endotoxin. It is apparent, however, that strategies are needed to delay or avoid resistance in pests that are intensively managed with Bt, regardless of the deployment method.

Inheritance

The development of strategies to manage resistance requires some understanding of the inheritance of a resistance trait in the pest population (Gould 1986). This was demonstrated by Tabashnik (1994b) using a population genetics model to simulate the response of a pest population to different resistance management strategies. The best strategy, i.e., one that delays resistance the longest, must be customized to the number of alleles and inheritance of the trait within the population. Studies of the genetics of resistance typically involve determining the susceptibility of progeny from crosses between individuals from the selected and unselected pop-

TABLE 5. FACTORS ASSOCIATED WITH DEVELOPMENT OF RESISTANCE TO BT IN TWO LEPIDOPTERANS.

Factors P. interpunctella P. xylostella

Pest Attributes and Management Short generation yes yes Generations/year 5-6 8-10 Population isolation yes yes Realized heritability high low Crop rotation no no Operational Use of Bt Multiple toxins yes yes Applications/generation 5-6 5-8 Frequency/generation 2-4 days 1.5-2 days Selection pressure high high 432 Florida Entomologist 78(3) September, 1995 ulations. In both P. interpunctella and P. xylostella, resistance is autosomally inherited (no maternal effects or sex linkage), partially recessive (progeny susceptibility more similar to unselected parent), and apparently due to one or a few major loci (McGaughey 1985, McGaughey & Beeman 1988, Hama et al. 1992, Tabashnik et al. 1992). The genetic basis of resistance to the CryIA toxin-complex in H. virescens is partially recessive and due to a single locus or set of tightly linked loci (Gould et al. in press). In L. decemlineata, resistance to CryIIIA is also autosomally inherited, and conferred by one incompletely dominant gene (Rahardja & Whalon 1995). Knowledge of the genetic basis of resistance is critical to our understanding of the stability of the trait within the selected population.

Stability Perhaps one of the simplest resistance management strategies involves providing the pest population with intermittent time periods in which Bt is not used for control. The success of such temporal refuges is dependent on the stability of the resistance trait after Bt exposure ceases. The rate of reversion is dependent on the inheritance of the resistance trait and the fitness costs associated with resistance. Different populations of resistant P. xylostella revert at different rates when selection with Bt is relaxed. Typically, the decline is slow and incomplete; for example, one Btk-resistant population declined from 29-fold to 1-fold after 32 generations without selection (Tabashnik et al. 1991). Similar results were reported in other Btk-resistant populations of P. xylostella (Tabashnik et al. 1991, 1994, Hamma et al. 1992) and in other resistant species, including P. interpunctella (McGaughey & Beeman 1988), H. virescens (Sims & Stone 1991), and L. decemlineata (Rahardja & Whalon 1995). In a recent study, rapid and complete reversal of resistance occurred after 13 generations without selection in a population of P. xylostella with 2800-fold resistance to Btk (Tabashnik et al. 1994a). This result suggests that resistance is achieved with significant loss in fitness, perhaps related to the alteration in the midgut binding sites documented in the study. Groeters et al. (1993, 1994) quantified reduced egg hatch, survival to the adult stage, fecundity, and mating success in male moths as significant fitness costs of resistance. Despite the rapid reversion to the susceptible genotype, the population responded rapidly to reselection (Tabashnik et al. 1994a). Rapid resur- gence of resistance in relaxed populations is typical, indicating the persistence of a low number of highly resistant individuals. If the alleles for resistance become fixed in the population, or other alleles compensate for losses in fitness, resistance becomes stable and reversion to susceptibility is unlikely.

RESISTANCE MANAGEMENT In an effort to preserve the utility of these unique insecticidal proteins, knowledge of resistance management to conventional pesticides is useful (Gould 1988a, 1988b, Stone et al. 1991, McGaughey & Whalon 1992, Whalon & McGaughey 1993, Mc- Gaughey 1994, Tabashnik 1994a). Unfortunately, selection for the resistance trait in a pest population is probably the inevitable consequence of insecticide use (Denholm & Rowland 1992). The goal then becomes how to design and manipulate operational strategies that best conserve susceptibility, thereby delaying resistance. The implementation of integrated pest management (IPM) strategies that optimize the goals of resistance management involves 1) diversifying the sources of mortality to avoid selection for a single mechanism, 2) reducing selection pressure for the major mortality factors, 3) maintaining susceptible individuals by providing refuges and encouraging immigration, 4) monitoring for increasing resistance to any one of the mortality agents, and 5) responding to resistance through management strategies designed to Bauer: Symposium on Pesticide Resistance 433 reduce the frequency of the resistance trait (Whalon & McGaughey 1993). Unfortu- nately, IPM is rarely implemented before a resistance crisis occurs, and generally another insecticide is available to replace the old one. Conventionally applied Bt-based insecticides are more amenable to such IPM strategies, because the short residual time and host specificity help reduce selection pressure associated with the Bt toxins. Insects surviving Bt exposure are generally in a weakened condition, facilitating their exploitation by other mortality factors such as beneficial insects (Tabashnik 1986, 1994a) and pathogens (Krieg 1971, Jacques & Morris 1981). Other stressing agents, such as adverse weather conditions and low plant nutritional quality, will also cause higher mortality in insects recovering from Bt exposure. The cumulative suppression exerted by these factors also reduces selection pressure by reducing the frequency of pesticide applications. However, recent improvements in Bt formulations involve increasing toxicity, increasing residual times, and in some cases broadening host range, thereby bringing selection pressure for resistance more in line with conventional insecticides (Tabashnik 1994a). The expression of δ-endotoxins in transgenic plants is considered analogous, in some respects, to intrinsic host plant defenses selected for by classical plant breeders (Gould 1988a). Unfortunately, pests adapt to resistant cultivars, and without appropriate deployment strategies, they are rendered ineffective (Gould 1986, Cox & Hatchett 1986). At present, bioengineering insect-resistant plants involves the incorporation of Bt δ-endotoxin genes into plants. There was considerable optimism several years ago that these plants would remain durable (Gould 1986,Gould 1988b, Rousch 1989). Early optimism and excitement over genetically-engineered plants expressing δ- endotoxin, perhaps bolstered by the presumption of an unlimited variety of these proteins, have given way to serious concern over the durability of these plants, and with it, conventional uses of Bt. Resistance management strategies in the context of Bt were recently reviewed in some detail by McGaughey & Whalon (1992), Whalon & McGaughey (1993), Robison et al. (1994), and Tabashnik (1994a). These strategies include 1) mixtures of toxins with different mechanisms, either within the same plant or in different plants, or expressed serially over time (Gould 1986); 2) synergists to increase toxicity (MacIntosh et al. 1990); 3) rotations to alternative toxins tempo- rally to reduce the frequency of resistant individuals (Tabashnik 1989); 4) refuges, temporal and spatial, to facilitate survival of susceptible individuals (Gould & Ander- son 1991); 5) low doses of toxin that produce sublethal effects, such as reduced fecundity and slowed development, favoring other mortality factors; 6) ultrahigh doses of toxin that kill resistant heterozygotes and homozygotes (Denholm & Row- land 1992, Tabashnik 1994a); and 7) gene regulation of toxin titre, location, and induction (Whalon & McGaughey 1993). Tabashnik (1994b) used a population genetics model (Mallet & Porter 1992) to simulate the effect of several resistance management strategies on resistance (single locus with two alleles) in a pest such as Heliothis. Transgenic plants expressing δ-endotoxin were planted as 1) pure stands of toxic plants, 2) seed mixtures with varying proportions of toxic and toxin-free plants, 3) toxin-free plants in refugia, and 4) seed mixtures + refugia. Across a range of conditions, seed mixtures always delayed the onset of insect resistance to the toxin, when compared to pure stands of toxic plants. Ref- ugia will delay resistance as long, or in some cases longer, than seed mixtures because refugia reduce selection without altering dominance (Mallet & Porter 1992). Refugia will delay resistance longer than mixtures + refugia only under specific conditions. However, the implementation of pest control within refugia limits their ability to delay resistance in proportion to the efficacy of the controls. 434 Florida Entomologist 78(3) September, 1995

The distinction between seed mixtures and refugia is the spatial distribution of toxin-free plants relative to the dispersal capability of pest larvae. Due to close proximity, larvae may move easily between toxic and toxin-free plants in fields planted with a seed mixture, whereas movement from toxic stands to toxic-free refugia is less likely. The tissue-specific expression of toxin through gene regulation is analogous to seed mixtures, because larvae can move freely between toxic and toxin-free tissue (Gould 1988a). Tabashnik (1994b) recommends maximizing spatial refuges, as well as temporal refuges, such as alternative crops and controls. The only method known to prolong the efficacy of any insecticide is to minimize the exposure of the target pests to the toxin in space and time (Denholm & Rowland 1992). Resistance management tactics must be validated in the field, under the inevitable practical, economic, and po- litical constraints imposed by agriculture today (May 1993).

CONCLUSIONS

The genetic capacity of insect populations to evolve resistance to Bt δ-endotoxins is now well documented in many species within eight different insect orders. Although high-level field resistance is known only in P. xylostella, much has been learned from studying the many insect populations selected for resistance to Bt in the laboratory. We now know that 1) Bt resistance alleles are present at varying levels in different insect species and populations, 2) within a single species, the genetics, mechanisms, level, and stability of resistance vary between selected populations, 3) selection with a blend of toxins can select for resistance to each toxin in the blend, 4) resistance occurs more rapidly with purified toxins than with spore/crystal preparations, 5) cross-resistance to δ-endotoxins is almost ubiquitous and often unpredictable, and 6) reselection of revertant populations is rapid. Today, Bt-based insecticides are frequently used in intensive agriculture, either in conjunction with conventional insecticides as a backup for control failure, or, as a last resort once resistance to other registered insecticides has occurred. Many insect pests, therefore, are already adapted to mixtures of δ-endotoxins. It is probable that the deployment of transgenic plants will precede the development of resistance management strategies, because advances in resistance management technology have not kept pace with those made in biotechnology. Developing and validating realistic resistance management plans that preserve the durability of δ-endotoxins deployed in transgenic plants may prove far more complex than the theory and techniques that actually generated the plants. Unfortunately, experimentally developed tactics to delay resistance have often been too naive or unrealistic for large-scale field implementation (Hoy 1995). Fortunately, many researchers with experience and knowledge of resistance management with conventional insecticides have shifted the emphasis of their research into the development of strategies designed to prolong the durability of these unique bacterial toxins in transgenic plants. In general, resistance management seeks to minimize the exposure of the target pest to a toxin in time and space. This can be accomplished by developing IPM plans for these crops that include synergists, seed mixtures and refugia, tissue-specific and inducible toxin expression, and alternating crops or control measures (Hoy 1995). Substantial benefits to the environment will be gained if insects can be successfully managed through the careful deployment of genetically-engineered insect-resistant plants. However, only few Bt δ-endotoxins, all with similar modes of action, are now available to plant molecular biologists. Deploy- ment of these plants before the management tactics are validated will result in the loss of Bt-based insecticides for many of the pests they are targeted to control. Bauer: Symposium on Pesticide Resistance 435

ACKNOWLEDGMENTS

An overview of this paper was presented in a symposium entitled “The Myths of Managing Resistance” at the Florida Entomological Society held 8-11 August 1994 in Stuart, Florida. I am grateful to M. Hoy for inviting me to participate in that symposium. I wish to thank M. Adang, N. Dubois, F. Gould, W. McGaughey, W. Moar, K. van Frankenhuyzen, B. Tabashnik, and M. Whalon for their help with preprints and various discussion, and D. Bradley for providing CryIB for laboratory bioassay. I am ap- preciative of the critical editorial comments provided by R. Haack, M. Hoy, N. Koller, and M. Whalon. I also want to give special acknowledgment to the excellent technical support provided by D. Miller in my laboratory with selection and bioassay of Chry- somela scripta.

REFERENCES CITED

ADANG, M. J. 1991. Bacillus thuringiensis insecticidal crystal proteins: gene structure, action, and utilization, pp. 3-24 in K. Maramorosch [ed], Biotechnology for biological control of pests and vectors. Boca Raton, Florida:CRC Press. ALY, C., M. S. MULLA, AND B. A. FEDERICI. 1985. Sporulation and toxin production by Bacillus thuringiensis var. israelensis in cadavers of mosquito larvae (Diptera: Culicidae). J. Invertebr. Pathol. 46: 251-258. ANGUS, T. A. 1954. A bacterial toxin paralyzing silkworm larvae. Nature (London) 173: 545-546. ARONSON, A. I. 1993. Insecticidal toxins, pp. 953-963 in A. L. Sonenshein, J. A. Hoch, and R. Losick [eds], Bacillus subtilus and other gram-positive bacteria: biochemistry, physiology, and molecular biology. Washington, DC:American Soci- ety for Microbiology. ARONSON, A. I., E. S. HAN, W. MCGAUGHEY, AND D. JOHNSON. 1991. The solubility of inclusion proteins from Bacillus thuringiensis is dependent upon protoxin composition and is a factor in toxicity to insects. Appl. Environ. Microbiol. 57: 981-986. BALLESTER, V., B. ESCRICHE, J. L. MÉNSUA, G. W. RIETHMACHER, AND JUAN FERRÉ. Lack of cross-resistance to other Bacillus thuringiensis crystal proteins in a population of Plutella xylostella highly resistant to CryIA(b). Biocontrol Sci. Technol. In Press. BAUER, L. S., AND H. S. PANKRATZ. 1992. Ultrastructural effects of Bacillus thuringiensis var san diego on midgut cells of the cottonwood leaf beetle. J. Invertebr. Pathol. 60: 15-25. BAUER, L. S., C. N. KOLLER, D. L. MILLER, AND R. M. HOLLINGWORTH. 1994. Labora- tory selection of the cottonwood leaf beetles, Chrysomela scripta, for resistance to Bacillus thuringiensis var. tenebrionis δ-endotoxin. p. 68 in Abstracts VIth In- tern. Coll. Invertebr. Pathol. BECKER, N., AND J. MARGOLIT. 1993. Use of Bacillus thuringiensis israeliensis against mosquitoes, pp. 147-170 in P. F. Entwistle, J. S. Cory, M. J. Bailey, and S. Higgs [eds], Bacillus thuringiensis, an environmental biopesticide: theory and practice. New York:Wiley. BERLINER, E. 1911. Über die Schlaffsucht der Mehlmottenraupe (Ephestia kuehniella Zell), und ihren Erreger Bacillus thuringiensis n. sp. Zeitschrift für an- gewandtes. Entomol. 2: 29-56. BOMAN, H. G. 1981. Insect Responses to Microbial Infections, pp. 769-784 in H. D. Burges and N. W. Hussey [eds], Microbial control of pests and plant diseases 1970-1980. New York:Academic Press. BORGONIE, G., R. VAN DRIESSCHE, F. LEYNS, G. ARNAUT, D. DE WAELE, AND A. COOMANS. 1995. Germination of Bacillus thuringiensis spores in bacteriopha- gous nematodes (Nematoda: Rhabditida). J. Invertebr. Pathol. 65: 61-67. 436 Florida Entomologist 78(3) September, 1995

BRADLEY, D., M. A. HARKEY, M. K. KIM, K. D. BIEVER, AND L. S. BAUER. 1995. Insec- ticidal CryIB crystal protein of Bacillus thuringiensis has dual specificity to coleopteran and lepidopteran larvae. J. Invertebr. Pathol. 65: 162-173. BRAVO, A., S. JANSENS, AND M. PEFEROEN. 1992. Immunocytochemical localization of Bacillus thuringiensis insecticidal crystal proteins in intoxicated insects. J. In- vertebr. Pathol. 60: 237-246. BREWER, G. J. 1991. Resistance to Bacillus thuringiensis subsp. kurstaki in the sun- flower moth (Lepidoptera: Pyralidae). Environ. Entomol. 20:316-322. BRIESE, G. J. 1981. Resistance of insect species to microbial pathogens, pp. 511-545 in E. W. Davidson [ed], Pathogenesis of Invertebrate Microbial Diseases. Totowa, New Jersey:Allanheld Osmun. BURGES, H. D. 1971. Possibilities for pest resistance to microbial control agents, pp. 445-457 in H. D. Burges and N. W. Hussey [eds], Microbial Control of Insects and Mites. New York:Academic Press. BURGES, H. D., AND J. A. HURST. 1977. Ecology of Bacillus thuringiensis in storage moths. J. Invertebr. Pathol. 30: 131-139. CARLTON, C. C., AND C. GAWRON-BURKE. 1993. Genetic improvement of Bacillus thuringiensis, pp. 43-61 in L. Kim [ed], Advanced Engineered Pesticides. New York:Marcel Dekker. CARROLL, J., AND D. J. ELLAR. 1993. Analysis of Bacillus thuringiensis δ-endotoxin action on insect-membrane permeability using a light-scattering assay. Eur. J. Biochem. 214: 771-778. CARROLL, J., J. LI, AND D. J. ELLAR. 1989. Proteolytic processing of a coleopteran-specific delta-endotoxin produced by Bacillus thuringiensis var. tenebrionis. Biochem. J. 261:99-105. CHILCOTT, C. N., B. H. KNOWLES, D. J. ELLAR, AND F. A. DROBNIEWSKI. 1990. Mech- anism of action of Bacillus thuringiensis parasporal body, pp. 44-65 in H. de Barjac and D. Sutherland [eds], Bacterial Control of Mosquitoes and Blackflies: Biochemistry, Genetics, and Applications of Bacillus thuringiensis and Bacillus sphaericus. New Brunswick:Rutgers University Press. COX, T. S., AND J. H. HATCHETT. 1986. Genetic model for wheat/Hessian fly (Diptera: Cecidomyiidae) interaction: strategies for deployment of resistance genes in wheat cultivars. Environ. Entomol. 15: 24-31. CUMMINGS, C. E., AND D. J. ELLAR. 1994. Chemical modification of Bacillus thuringiensis activated δ-endotoxin and its effect on toxicity and binding to Manduca sexta midgut membranes. Microbiol. 140: 2737-2747. DE BARJAC, H. 1987. Operational bacterial insecticides and their potential for future improvements, pp. 63-73 in K. Maramorosch [ed], Biotechnology in Inverte- brate Pathology and Cell Culture. San Diego, California:Academic Press. DENHOLM, I., AND M. W. ROWLAND. 1992. Tactics for managing pesticide resistance in arthropods: theory and practice. Annu. Rev. Entomol. 37: 91-112. DENOLF, P., S, JANSENS, M. PEFEROEN, D. DEGHEELE, AND J. VAN RIE. 1993. Two different Bacillus thuringiensis delta-endotoxin receptors in the midgut brush border membrane of the european corn borer, Ostrinia nubilalis (Hübner) (Lep- idoptera: Pyralidae). Appl. Environ. Microbiol. 59: 1828-1837. DEVRIENDT, M., AND D. MARTOURET. 1976. Absence de résistance à Bacillus thuringiensis chez la teigne des crucifères, Plutella maculipennis. Entomophaga 21: 189-199. DU, C., P. A. MARTIN, AND K. W. NICKERSON. 1994. Comparison of disulfide contents and solubility at alkaline pH of insecticidal and noninsecticidal Bacillus thuringiensis protein crystals. Appl. Environ. Microbiol. 60: 3847-3853. ELLAR, D. J. 1994. Structure and mechanism of action of Bacillus thuringiensis endotoxins and their receptors. Biocon. Sci. Technol. 4: 445-447. ELY, S. 1993. The engineering of plants to express Bacillus thuringiensis δ-endotoxins, pp. 105-124 in P. F. Entwistle, J. S. Cory, M. J. Bailey, and S. Higgs [eds], Bacillus thuringiensis, an environmental biopesticide: theory and practice. New York:Wiley, 311 pp. Bauer: Symposium on Pesticide Resistance 437

ENGLISH, L., AND S. L. SLATIN. 1992. Mode of action of delta-endotoxin from Bacillus thuringiensis: a comparison with other bacterial toxins. Insect Biochem. Molec. Biol. 22: 1-7. ESCRICHE, B., A. C. MARTINEZ-RAMIREZ, M. D. REAL, F. J. SILVA, AND J. FERRÉ. 1994. Occurrence of three different binding sites for Bacillus thuringiensis δ-endotoxins in the midgut brush border membrane of potato tuber moth, Phthorimaea operculella (Zeller). Arch. Insect Biochem. Physiol. 26: 315-327. ESTADA, U., AND J. FERRÉ. 1994. Binding of insecticidal crystal proteins of Bacillus thuringiensis to the midgut brush border of the cabbage looper, Trichoplusia ni (Hübner) (Lepidoptera: Noctuidae), and selection for resistance to one of the crystal proteins. Appl. Environ. Microbiol. 60: 3840-3846. FALCONER, D. S. 1989. Introduction to Quantitative Genetics. New York:Longman. FEDERICI, B. A. 1993. Insecticidal bacterial proteins identify the midgut epithelium as a source of novel target sites for insect control. Arch. Insect Biochem. Physiol. 22: 357-371. FERRÉ, J., M. D. REAL, J. VAN RIE, S. JANSENS, AND M. PERFEROEN. 1991. Resistance to the Bacillus thuringiensis bioinsecticide in a field population of Plutella xylostella is due to a change in a midgut membrane receptor. Proc. Natl. Acad. Sci. USA 88: 5119-5123. FEITELSON, J. S. 1993. The Bacillis thuringiensis family tree, pp. 63-71 in L. Kim [ed.], Advanced Engeered Pesticides, New York: Marcel Dekker. FEITELSON, J. S., J. PAYNE, AND L. KIM. 1992. Bacillus thuringiensis: insects and be- yond. Bio/technology 10: 271-276. GARCZYNSKI, S. F., AND M. J. ADANG. Bacillus thuringiensis CryIA(c) δ-endotoxin binding aminopeptidase in the Manduca sexta midgut has a glycosyl-phos- phatidylinositol anchor. Insect Biochem. Molec. Biol. In Press. GARCZYNSKI, S. F., J. W. CRIM, AND M. J. ADANG. 1991. Identification of putative insect brush border membrane-binding molecules specific to Bacillus thuringiensis δ-endotoxin by protein blot analysis. Appl. Environ. Microbiol. 57: 2816- 2820. GASSER, C. S., AND R. T. FRALEY. 1989. Genetically engineering plants for crop improvement. Science 244: 1293-1299. GELERNTER, W., AND G. E. SCHWAB. 1993. Transgenic bacteria, viruses, algae and other microorganisms as Bacillus thuringiensis delivery systems, pp. 89-104 in P. F. Entwistle, J. S. Cory, M. J. Bailey, and S. Higgs [eds], Bacillus thuringiensis, an environmental biopesticide: theory and practice. New York:Wiley, 311 pp. GEORGHIOU, G. P. 1990. Resistance potential to biopesticides and consideration of countermeasures, pp. 409-420 in J. E. Casida [ed], Pesticides and Alternatives. New York:Elsevier. GEORGHIOU, G. P. 1994. Mechanisms and microbial characteristics of invertebrate resistance to bacterial toxins, pp. 48-50. in Proc. VIth Intern. Coll. Invertebr. Pathol., Montpellier, France:Soc. Invertebr. Pathol. GEORGHIOU, G. P., AND A. LAGUNES. 1988. The occurrence of resistance to pesticides: cases of resistance reported worldwide through 1988. Food and Agricultural Organization, Rome, 325 pp. GILL, S. S., E. A. COWLES, AND P. V. PIETRANTONIO. 1992. The mode of action of Ba- cillus thuringiensis endotoxins. Annu. Rev. Entomol. 37: 615-36. GOLDBERG, L. J., AND J. MARGALIT. 1977. A bacterial spore demonstrating rapid lar- vicidal activity against Anopheles serengetii, Uranotaenia unguiculata, Culex univittatus, Aedes aegypti and Culex pipiens. Mosq. News. 37: 355-358. GOLDMAN, I. F., J. ARNOLD, AND B. C. CARLTON. 1986. Selection for resistance to Ba- cillus thuringiensis subspecies israelensis in field and laboratory populations of the mosquito Aedes aegypti. J. Invertebr. Pathol. 47: 317-324. GOULD, F. 1986. Simulation models for predicting durability of insect-resistant germ- plasm: Hessian fly (Diptera: Cecidomyiidae) resistant winter wheat. Environ. Entomol. 15: 1-10. 438 Florida Entomologist 78(3) September, 1995

GOULD, F. 1988a. Evolutionary biology and genetically engineered crops: Consider- ation of evolutionary theory can aid in crop design. BioScience 38: 26-33. GOULD, F. 1988b. Genetic engineering, integrated pest management and the evolution of pests, pp. 15-18 in Planned Release of Genetically Engineered Organ- isms. Cambridge (UK):Elsevier. GOULD, F., AND A. ANDERSON. 1991. Effects of Bacillus thuringiensis and HD-73 delta-endotoxin on growth, behavior, and fitness of susceptible and toxin-adapted strains Heliothis virescens (Lepidoptera: Noctuidae). Environ. Entomol. 20: 30-38. GOULD, F., A. ANDERSON, D. LANDIS, AND H. VAN MELLAERT. 1991. Feeding behavior and growth of Heliothis virescens larvae on diets containing Bacillus thuringiensis formulations or endotoxins. Entomol. Exp. Appl. 58: 199-210. GOULD, F., A. ANDERSON, A. REYNOLDS, L. BUMGARNER, AND W. MOAR. Selection and genetic analysis of a Heliothis virescens (Lepidoptera: Noctuidae) strain with high levels of resistance to some Bacillus thuringiensis toxins. J. Econ. Ento- mol. In Press. GOULD, F., A. MARTINEZ-RAMIREZ, A. ANDERSON, J. FERRE, F. J. SILVA, AND W. J. MOAR. 1992. Broad-spectrum resistance to Bacillus thuringiensis toxins in He- liothis virescens. Proc. Natl. Acad. Sci. 88: 7986-7990. GROETERS, F. R., B. E. TABASHNIK, N. FINSON, AND M. W. JOHNSON. 1993. Resistance to Bacillus thuringiensis affects mating success of the diamondback moth (Lep- idoptera: Plutellidae). J. Econ. Entomol. 86: 1035-1039. GROETERS, F. R., B. E. TABASHNIK, N. FINSON, AND M. W. JOHNSON. 1994. Fitness costs of resistance to Bacillus thuringiensis in the diamondback moth (Plutella xylostella). Evol. 48: 197-201. HAMA, H., K. SUZUKI, AND H. TANAKA. 1992. Inheritance and stability of resistance to Bacillus thuringiensis formulations of the diamondback moth, Plutella xylostella L. (Lepidoptera: Yponomeutidae). Appl. Entomol. Zool. 27: 355-362. HEIMPEL, A. M., AND T. A. ANGUS. 1959. The site of action of crystalliferous bacteria in Lepidoptera larvae. J. Insect Pathol. 1: 152-170. HOFFMAN, C., P. LUTHY, R. HUTTER, AND V. PLESKA. 1988a. Binding of the delta-endotoxin from Bacillus thuringiensis to brush border membrane vesicles of the cabbage butterfly (Pieris brassicae). Eur. J. Biochem. 173:85-91. HOFFMAN, C., H. VANDERBRUGGEN, H. HÖFTE, J. VAN RIE, AND S. JANSENS. 1988b. Specificity of Bacillus thuringiensis delta-endotoxin is correlated with the presence of high-affinity binding sites in the brush border membrane of target insect midguts. Proc. Natl. Acad. Sci. USA 85: 7844-7848. HÖFTE, H., AND H. R. WHITELEY. 1989. Insecticidal crystal proteins of Bacillus thuringiensis. Microbiol. Rev. 53: 242-255. HONÉE, G., AND B. VISSER. 1993. The mode of action of Bacillus thuringiensis crystal protein. Entomol. exp. appl. 69: 145-155. HOY, M. A. 1995. Multitactic resistance management: an approach that is long overdue? Florida Entomol. 78:443-451. ISHIWATA, S. 1901. On a kind of severe flasherie (sotto disease). Dainihan Sanbshi Kaiho 9: 1-5. JAQUES, R. P., AND O. N. MORRIS. 1981. Compatibility of pathogens with other methods of pest control and with different crops, pp. 695-715 in H. D. Burges [ed], Microbial control of pests and plant diseases 1970-1980. New York:Academic Press. JANSSON, R. K., AND S. H. LECRONE. 1990. Management of diamondback moth, Plu- tella xylostella, with nonconventional chemical and biological insecticides. Proc. Florida State Hortic. Soc. 103: 122-126. KEIDING, J. 1986. Prediction of resistance risk assessment, pp. 279-297 in National Research Council. Pesticide Resistance: Strategies and Tactics for Manage- ment. Washington, DC:National Academy. KELLER, B., AND G. LANGENBRUCH. 1993. Control of coleopteran pests by Bacillus thuringiensis and resistance management, pp. 171-191 in P. F. Entwistle, J. S. Bauer: Symposium on Pesticide Resistance 439

Cory, M. J. Bailey, and S. Higgs [eds], Bacillus thuringiensis, an environmental biopesticide: theory and practice. New York:Wiley. KINSINGER, R. A., AND W. H. MCGAUGHEY. 1979. Susceptibility of populations of In- dianmeal moth and almond moth to Bacillus thuringiensis isolates (Lepi- doptera: Pyralidae). J. Econ. Entomol. 72: 346-349. KIRSCH, K., AND H. SCHMUTTERER. 1988. Low efﬁcacy of a Bacillus thuringiensis (Berl.) formulation in controlling the diamondback moth, Plutella xylostella (L.), in the Philippines. J. Appl. Entomol. 105: 249-255. KNIGHT, P. J. K., N. CRICKMORE, AND D. J. ELLAR. 1994. The receptor for Bacillus thuringiensis CryIA(c) delta-endotoxin in the brush border membrane of the lepidopteran Manduca sexta is aminopeptidase N. Mol. Microbiol. 11: 429-436. KNOWLES, B. H., AND J. A. T. DOW. 1993. The crystal delta-endotoxins of Bacillus thuringiensis: models for their mechanism of action on the insect gut. Bioessays 15: 469-475. KNOWLES, B. H., AND D. J. ELLAR. 1987. Colloid-osmotic lysis is a general feature of the mechanism of action of Bacillus thuringiensis delta-endotoxin with different insect speciﬁcity. Biochem. Biophys. Acta 924: 509-518. KOLLER, C. N., L. S. BAUER, AND R. M. HOLLINGWORTH. 1992. Characterization of the pH-mediated solubility of Bacillus thuringiensis var. san diego native delta-endotoxin crystals. Biochem. Biophys. Res. Comm. 184:692-699. KONI, P. A., AND D. J. ELLAR. 1994. Biochemical characterization of Bacillus thuringiensis cytolytic δ-endotoxin. Microbiol. 140: 1869-1880. KRIEG, A. 1971. Interactions between pathogens, pp. 459-468 in H. D. Burges and N. W. Hussey [eds], Microbial control of insects and mites. New York:Academic Press. KRIEG, A., AND G. A. LANGENBRUCH. 1981. Susceptibility of arthropod species to Ba- cillus thuringiensis, pp. 837-898 in H. D. Burges and N. W. Hussey [eds], Mi- crobial Control of Pests and Plant Diseases 1970-1980. New York: Academic Press. KRIEG, A., A. M. HUGER, G. A. LANGENBROOK, AND W. SCHNETTER. 1983. Bacillus thuringiensis var. tenebrionis: ein neuer gegenuber larven von Coleopteren wirk- samer. Pathotyp. Z. Ang. Entomol. 96: 500-508. LAMBERT, B., AND M. PEFEROEN. 1992. Insecticidal promise of Bacillus thuringiensis. Bioscience 42: 112-122. LI. J., J. CARROLL, AND D. J. ELLAR. 1991. Crystal structure of insecticidal delta-endotoxin from Bacillus thuringiensis at 2.5 Å resolution. Nature 353: 815-821. MACINTOSH, S. C., G. M. KISHORE, F. J. PERLAK, P. G. MARRONE, AND T. B. STONE. 1990. Potentiation of Bacillus thuringiensis insecticidal activity by serine protease inhibitors. J. Agric. Food Chem. 38: 1145-1152. MACINTOSH, S. C., T. B. STONE, R. S. JOKERST, AND R. L. FUCHS. 1991. Binding of Ba- cillus thuringiensis proteins to a laboratory-selected line of Heliothis virescens. Proc. Natl. Acad. Sci. USA 88: 8930-8933. MALLET, J., AND P. PORTER. 1992. Preventing insect adaptation to insect-resistant crops: are seed mixtures or refugia the best strategy? Proc. R. Soc. London B 250: 165-169. MARRONE, P. G., AND S. C. MACINTOSH. 1993. Resistance to Bacillus thuringiensis and resistance management, pp. 221-235 in P. F. Entwistle, J. S. Cory, M. J. Bailey, and S. Higgs [eds], Bacillus thuringiensis, an environmental biopesticide: theory and practice. New York:Wiley. MARTIN, P. A. W., AND R. S. TRAVERS. 1989. Worldwide abundance and distribution of Bacillus thuringiensis isolates. Appl. Environ. Microbiol. 55: 2437-2442. MASSON, L., A. MAZZA, R. BROUSSEAU, AND B. TABASHNIK. 1995. Kinetics of Bacillus thuringiensis toxin binding with brush border membrane vesicles from susceptible and resistant larvae of Plutella xylostella. J. Biol. Chem. 270: 1-10. MAY, R. M. 1993. Resisting resistance. Nature 361: 593-594. 440 Florida Entomologist 78(3) September, 1995

MCGAUGHEY, W. H. 1985. Insect resistance to the biological insecticide Bacillus thuringiensis. Science 229:193-195. MCGAUGHEY, W. H. 1994. Problems of insect resistance to Bacillus thuringiensis. Agr. Ecosystem. Environ. 49: 95-102. MCGAUGHEY, W. H., AND R. W. BEEMAN. 1988. Resistance to Bacillus thuringiensis in colonies of Indianmeal moth and almond moth (Lepidoptera:Pyralidae). J. Econ. Entomol. 81: 28-33. MCGAUGHEY, W. H., AND D. E. JOHNSON. 1987. Toxicity of different serotypes and toxins of Bacillus thuringiensis to resistant and susceptible Indianmeal moths. J. Econ. Entomol. 80: 1122-1126. MCGAUGHEY, W. H., AND D. E. JOHNSON. 1992. Indianmeal moth (Lepidoptera: Pyral- idae) resistance to different strains and mixtures of Bacillus thuringiensis. J. Econ. Entomol. 85: 1594-1600. MCGAUGHEY, W. H., AND D. E. JOHNSON. 1994. Inﬂuence of crystal protein composition of Bacillus thuringiensis strains on cross-resistance in Indianmeal moths (Lepidoptera: Pyralidae). J. Econ. Entomol. 87: 535-540. MCGAUGHEY, W. H., AND M. E. WHALON. 1992. Managing insect resistance to Bacil- lus thuringiensis toxins. Science 258: 1451-1455. MEADE, T., AND J. D. HARE. 1994. Effects of genetic and environmental host plant variation on the susceptibility of two noctuids to Bacillus thuringiensis. Ento- mol. exp. appl. 70: 165-178. MEADOWS, M. P. 1993. Bacillus thuringiensis in the environment: ecology and risk assessment, pp. 193-220 in P. F. Entwistle, J. S. Cory, M. J. Bailey, and S. Higgs [eds], Bacillus thuringiensis, an environmental biopesticide: theory and practice. New York:Wiley. MIYASONO, M, M. YAMAMOTO, S. INAGAKI, K. OHBA, T. ISHIGURO, Y. HAYASHI, AND R. TAKEDA. 1994. Effects of several microorganisms on the insecticidal activity of δ-endotoxin of Bacillus thuringiensis serovar kurstaki HD-1. Jpn. J. Appl. En- tomol. Zool. 38: 101-108. MOAR, W., M. PUSZTAI-CAREU, H. VAN FAASSEN, R. MCCOLLUM, AND S. CLEMENT. 1994. Resistance to Bacillus thuringiensis CryIC in Spodoptera exigua. p. 65 in Abstracts VIth Intern. Coll. Invertebr. Pathol. MOLDENKE, A. F., R. E. BERRY, J. C. MILLER, J. G. WERNZ, X. H. LI. 1994. Toxicity of Bacillus thuringiensis subsp. kurstaki to gypsy moth, Lymantria dispar, fed with alder or Douglas ﬁr. J. Invertebr. Pathol. 64: 143-145. MÜLLER-COHN, J., J. CHAUFAUX, C. BUISSON, N. GILOIS, V. SANCHIS, AND D. LERE- CLUS. 1994. Spodoptera littoralis resistance to the Bacillus thuringiensis cryIC toxin and cross-resistance to other toxins. p. 66 in Abstracts VIth Intern. Coll. Invertebr. Pathol. NATIONAL RESEARCH COUNCIL. 1986. Pesticide Resistance: strategies and tactics for management. National Academy, Washington, DC. 471 pp. NAVON, A. 1993. Control of lepidopteran pests with Bacillus thuringiensis, pp. 125- 146 in P. F. Entwistle, J. S. Cory, M. J. Bailey, and S. Higgs [eds], Bacillus thuringiensis, an environmental biopesticide: theory and practice. New York:Wiley. OGIWARA, K., L. S. INDRASITH, S. ASANO, AND H. HORI. 1992. Processing of delta-endotoxin from Bacillus thuringiensis subsp. kurstaki HD-1 and HD-73 by gut juices of various insect larvae. J. Invertebr. Pathol. 60: 121-126. OPPERT, B., K. J. KRAMER, D. E. JOHNSON, S. C. MACINTOSH, AND W. H. MCGAUGHEY. 1994. Altered protoxin activation by midgut enzymes from a Bacillus thuringiensis resistant strain of Plodia interpunctella. Biochem. Biophys. Res. Comm. 198: 940-947. PARENTI, P, V. MANUELA, G. M. HANOZET, M. TASCA, AND B. GIORDANA. 1995. Inter- action of the insecticidal crystal protein CryIA from Bacillus thurigiensis with amino acid transport into brush border membranes from Bombyx mori larval midgut. J. Invertebr. Pathol. 65: 35-42. Bauer: Symposium on Pesticide Resistance 441

PEFEROEN, M. 1992. Engineering of insect-resistant plants with Bacillus thuringiensis crystal protein genes, pp. 135-153 in A. M. R. Gatehouse, V. A. Hilder, and D. Boulter [eds], Plant genetic manipulation for crop protection. Wallingford, United Kingdom:CAB International. RAHARDJA, U., AND M. E. WHALON. 1995. Inheritance of resistance to Bacillus thuringiensis subsp. tenebrionis δ-endotoxin in the Colorado potato beetle (Co- leoptera: Chrysomelidae) J. Econ. Entomol. 88: 21-26. RAMACHANDRIN R., K. F. RAFFA, M. J. MILLER, D. D. ELLIS, AND B. H. MCCOWN. 1993. Behavioral responses and sublethal effects of spruce budworm (Lepidoptera: Tortricidae) and fall webworm (Lepidoptera: Arctiidae) to Bacillus thuringiensis CryIA(s) toxin in diet. Environ. Entomol. 22: 197-211. ROBISON, D. J., B. H. MCCOWN, AND K. F. RAFFA. 1994. Responses of gypsy moth (Lep- idoptera: Lymanitriidae) and forest tent caterpillar (Lepidoptera: Lasiocampi- dae) to transgenic poplar, Populus spp., containing a Bacillus thuringiensis δ- endotoxin gene. Environ. Entomol. 23: 1030-1041. ROUSCH, R. T. 1989. Designing resistance management programs: how can you choose? Pestic. Sci. 26: 423-441. SALAMA, H. S., AND M. M. MATTER. 1991. Tolerance level to Bacillus thuringiensis Berliner in the cotton leafworm Spodoptera littoralis Biosduval (Lepidoptera: Noctuidae). J. Appl. Entomol. 111: 225-230. SALEH, M. S. 1987. Effect of larval selection with two bioinsecticides on susceptibility levels and reproductive capacity of Aedes aegypti. Anz. Schadingskd. Pflanz. Umweltschutz 60: 55-57. SANCHIS, V., J. CHAUFAUX, AND D. PAURON. 1994. A comparison and analysis of the toxicity and receptor binding properties of Bacillus thuringiensis CryIC δ-endotoxin on Spodoptera littoralis and Bombyx mori. FEBS Letters 353: 259-263. SANGADALA, S., F. W. WALTERS, L. H. ENGLISH, AND M. J. ADANG. 1994. A mixture of Manduca sexta aminopeptidase and phosphatase enhances Bacillus thuringiensis insecticidal CryIA(c) toxin binding and 86Rb+-K+ efflux in vitro. J. Biol. Chem. 269: 10088-10092. SCHWARTZ, J. M., B. E. TABASHNIK, AND M. W. JOHNSON. 1991. Behavioral and physiological responses of susceptible and resistant diamondback moth larvae to Bacillus thuringiensis. Entomol. Exp. Appl. 61: 179-187. SHELTON, A. M., J. L. ROBERTSON, J. D. TANG, C. PEREZ, AND S. E. EIGENBRODE. 1993. Resistance of diamondback moth to Bacillus thuringiensis subspecies in the field. J. Econ. Entomol. 86: 697-705. SIMS, S. R., AND T. B. STONE. 1991. Genetic basis of tobacco budworm resistance to an engineered Pseudomonas fluorescens expressing the δ-endotoxin of Bacillus thuringiensis var. kurstaki. J. Invertebr. Pathol. 57: 206-210. SMITH, R. A., AND G. A. COUCHE. 1991. The phylloplane as a source of Bacillus thuringiensis variants. Appl. Environ. Microbiol. 57: 311-315. STONE, T. B., AND S. R. SIMS. 1993. Geographic susceptibility of Heliothis virescens and Helicoverpa zea (Lepidoptera: Noctuidae) to Bacillus thuringiensis. J. Econ. Entomol. 86: 989-994. STONE, T. B., S. R. SIMS, S. C. MACINTOSH, R. L. FUCHS, AND P. G. MARRONE. 1991. Insect resistance to Bacillus thuringiensis, pp. 53-66 in K. Maramorosch [ed], Biotechnology for biological control of pests and vectors. Boca Raton, Flor- ida:CRC Press. STONE, T. B., S. R. SIMS, AND P. G. MARRONE. 1989. Selection of tobacco budworm for resistance to a genetically engineered Pseudomonas fluorescens containing the delta-endotoxin of Bacillus thuringiensis subsp. kurstaki. J. Invertebr. Pathol. 53: 228-234. TABASHNIK, B. E. 1986. Evolution of pesticide resistance in predator-prey systems. Bull. Entomol. Soc. America 32: 156-161. TABASHNIK, B. E. 1989. Managing resistance with multiple pesticide tactics: theory, evidence and recommendations. J. Econ. Entomol. 82: 1263-1269. 442 Florida Entomologist 78(3) September, 1995

TABASHNIK, B. E. 1992. Resistance risk management: realized heritability of resistance to Bacillus thuringiensis in diamondback moth (Lepidopteran: Plutel- lidae), tobacco budworm (Lepidopteran: Noctuidae), and Colorado potato beetle (Coloeptera: Chrysomelidae). J. Econ. Entomol. 85: 1551-1559. TABASHNIK, B. E. 1994a. Evolution of resistance to Bacillus thuringiensis. Annu. Rev. Entomol. 39: 47-79. TABASHNIK, B. E. 1994b. Delaying insect adaptation to transgenic plants: seed mixtures and refugia reconsidered. Proc. R. Soc. London B 255: 7-12. TABASHNIK, B. E., AND W. H. MCGAUGHEY. 1994. Resistance risk assessment for single and multiple insecticides: responses of Indianmeal moth (Lepidoptera: Pyralidae) to Bacillus thuringiensis. J. Econ. Entomol. 87:834-841. TABASHNIK, B. E., N. L. CUSHING, N. FINSON, AND M. W. JOHNSON. 1990. Field development of resistance to Bacillus thuringiensis in diamondback moth (Lepi- doptera: Plutellidae). J. Econ. Entomol. 83: 1671-1676. TABASHNIK, B. E., N. FINSON, C. F. CHILCUTT, N. L. CUSHING, AND M. W. JOHNSON. 1993. Increasing efficiency of bioassays: evaluation of resistance to Bacillus thuringiensis in the diamondback moth (Lepidoptera: Plutellidae). J. Econ. En- tomol. 86: 635-644. TABASHNIK, B. E., N. FINSON, F. R. GROETERS, W. J. MOAR, M. W. JOHNSON, K. LUO, AND M. J. ADANG. 1994a. Reversal of resistance of Bacillus thuringiensis in Plutella xylostella. Proc. Natl. Acad. Sci. USA 91: 4120-4124. TABASHNIK, B. E., N. FINSON, AND M. W. JOHNSON. 1991. Managing resistance to Ba- cillus thuringiensis: lessons from the diamondback moth (Lepidoptera: Plutel- lidae). J. Econ. Entomol. 84: 49-55. TABASHNIK, B. E., N. FINSON, M. W. JOHNSON, AND D. G. HECKEL. 1994b. Cross-resistance to Bacillus thuringiensis toxin CryIF in the diamondback moth (Plu- tella xylostella). Appl. Environ. Microbiol. 60: 4627-4629. TABASHNIK, B. E., J. M. SCHWARTZ, N. FINSON, AND M. W. JOHNSON. 1992. Inherit- ance of resistance to Bacillus thuringiensis in the diamondback moth (Lepi- doptera: Plutellidae). J. Econ. Entomol. 85: 1046-1055. TOJO, A., AND K. AIZAWA. 1983. Dissolution and degradation of Bacillus thuringiensis delta-endotoxin by gut juice protease of the silkworm, Bombyx mori. Environ. Microbiol. 45: 576-580. VAN FRANKENHUYZEN. 1993. The challenge of Bacillus thuringiensis, pp. 1-35 in P. F. Entwistle, J. S. Cory, M. J. Bailey, and S. Higgs [eds], Bacillus thuringiensis, an environmental biopesticide: theory and practice. New York:Wiley. VAN FRANKENHUYZEN, K., C. W. NYSTROM, AND B. E. TABASHNIK. 1995. Variation in tolerance to Bacillus thuringiensis among and within populations of the spruce budworm (Lepidoptera: Tortricidae) in Ontario. J. Econ. Entomol. 88: 97-105. VAN RIE, J. S. 1991. Insect control with transgenic plants: resistance proof ? Trends Biotechnol. 9: 177-179. VAN RIE, J., S. JANSENS, H. HÖFTE, D. DEGHEELE, AND H. VAN MELLAERT. 1989. Spec- ificity of Bacillus thuringiensis delta-endotoxins. Eur. J. Biochem. 186: 239- 247. VAN RIE, J., S. JANSENS, H. HÖFTE, D. DEGHEELE, AND H. VAN MELLAERT. 1990a. Re- ceptors on the brush border membrane of the insect midgut as determinants of the specificity of Bacillus thuringiensis delta-endotoxins. Appl. Environ. Micro- biol. 56: 1378-1385. VAN RIE, J., W. H. MCGAUGHEY, D. E. JOHNSON, B. D. BARNETT, AND H. VAN MEL- LAERT. 1990b. Mechanisms of insect resistance to the microbial insecticide Ba- cillus thuringiensis. Science 247: 72-74. VISSER, B., D. BOSCH, AND G. HONÉE. 1993. Domain-function studies of Bacillus thuringiensis crystal protein, pp. 71-88 in P. F. Entwistle, J. S. Cory, M. J. Bailey, and S. Higgs [eds], Bacillus thuringiensis, an environmental biopesticide: theory and practice. New York:Wiley. Bauer: Symposium on Pesticide Resistance 443

WALTERS, F. S., C. A. KULESZE, A. T. PHILLIPS, AND L. H. ENGLISH. 1994. A stable oli- gomer of Bacillus thuringiensis delta-endotoxin, CryIIIA. Insect Biochem. Molec. Biol. 24: 963-968. WALTERS, F. S., S. L. SLATIN, C. A. KULESZE, AND L. H. ENGLISH. 1993. Ion channel activity of n-terminal fragments from cryIA(c) delta-endotoxin. Biochem. Bio- phys. Res. Comm. 196: 921-926. WATKINSON, I. 1994. Global view of present and future markets for Bt products. Agri. Ecosys. Environ. 49: 3-7. WHALON, M. E., AND W. H. MCGAUGHEY. 1993. Insect resistance to Bacillus thuringiensis, pp. 215-232 in L. Kim [ed], Advanced Engineered Pesticides, New York:Marcel Dekker. WHALON, M. E., D. L. MILLER, R. M. HOLLINGWORTH. E. J. GRAFIUS, AND J. R. MILLER. 1993. Selection of a Colorado potato beetle (Coleoptera: Chrysomel- idae) strain resistant to Bacillus thuringiensis. J. Econ. Entomol. 86: 226-233. WILSON, G. R., AND T. G. BENOIT. 1990. Activation and germination of Bacillus thuringiensis spores in Manduca sexta larval gut ﬂuid. J. Invertebr. Pathol. 56: 233- 236. WOLFERSBERGER, M. G. 1990. The toxicity of two Bacillus thuringiensis δ-endotoxins to gypsy moth larvae is inversely related to the afﬁnity of binding sites on the midgut brush border membranes for the toxins. Experientia 46: 475-477. WOLFERSBERGER, M. G. 1991. Inhibition of potassium-gradient-driven phenylalanine uptake in larval Lymantria dispar midgut by two Bacillus thuringiensis δ-endotoxins correlates with the activity of the toxins as gypsy moth larvicides. J. Exp. Biol. 161: 519-525. WU, D., J. J. JOHNSON, AND B. A. FEDERICI. 1994. Synergism of mosquitocidal toxicity between CytA and CryIVD proteins using inclusions produced from cloned genes of Bacillus thuringiensis. Mol. Microbiol. 13: 965-972. YAMAMOTO, T., AND G. K. POWELL. 1993. Bacillus thuringiensis crystal proteins: recent advances in understanding insecticidal activity, pp. 215-232 in L. Kim [ed], Advanced Engineered Pesticides, New York:Marcel Dekker.

♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦

Hoy: Symposium on Managing Resistance 443

MULTITACTIC RESISTANCE MANAGEMENT: AN APPROACH THAT IS LONG OVERDUE?

MARJORIE A. HOY Department of Entomology and Nematology University of Florida, Gainesville, FL 32611-0620

ABSTRACT

“Resistance management” tactics have been much discussed, but such tactics have provided surprisingly limited practical results for pest management programs to date. We have learned a great deal about pesticide resistance mechanisms, the mode of inheritance of resistances, the molecular basis of resistance and cross resistance mechanisms, and how to evaluate the impact of resistance on ﬁtness. However, it remains difﬁcult to “manage” resistance once resistant individuals make up more than 5 to 10% of the population. Generally, the best that can be achieved is to delay the development of high levels of resistance for a few years, most often by using the product less often. A more effective resistance management strategy will combine a variety of effective pest management tactics along with a reduction in numbers and rates of pesticides applied. Effective pest management tactics include monitoring, evaluating

444 Florida Entomologist 78(3) September, 1995 economic injury levels so that pesticides are applied only when needed, biological control, host plant resistance, cultural controls, biorational pest controls, and genetic control methods. As a part of this multi-tactic strategy, it is crucial to evaluate the impact of pesticides on natural enemies. Sometimes, pesticide-resistant natural enemies can be effective components of a program to delay the development of resistance in pest arthropods.

Key Words: Pesticide resistance, integrated pest management, biological control, pesticide selectivity, resistance management.

RESUMEN Las tácticas de manejo de la resistencia han sido muy discutidas pero sorprenden- temente, hasta la fecha, tales tácticas han dado resultados prácticos limitados en programas de manejo de plagas. Hemos aprendido mucho sobre los mecanismos de resistencia, el modo de heredarse, sus bases moleculares y los mecanismos de resistencia cruzada, así como a evaluar el impacto de la resistencia en el ajuste genético. Sin embargo, es aún difícil manejar la resistencia cuando los individuos resistentes integran más del 5-10% de la población. Generalmente, lo mejor que se ha logrado ha sido retrasar en varios años el desarrollo de altos niveles de resistencia en la pobla- ción, más a menudo usando menos f recuentemente el producto que la provoca. Una estrategia más efectiva de manejo de la resisitencia combinaría una variedad de tác- ticas eﬁcaces de manejo de plagas con la de reducir el número y dosis de los pesticidas aplicados. Las tácticas efectivas de manejo inculyen el monitoreo, la evaluación del daño económico de modo que los pesticidas sean aplicados solamente cuando es nece- sario, el control biológico, la resistencia de las plantas hospedantes, el control cultural, el control biorracional de plagas, y los métodos de control genético. Como parte de esta estrategia multitáctica, es crucial evaluar el impacto de los pesticidas en los enemigos naturales. A veces, los enemigos naturales resistentes pueden ser conponentes efectivos de un programa para retrasar el desarrollo de la resistencia en los artrópo- dos plagas.

This commentary will make the following argument: “integrated pest management” (IPM) and “management of pesticide resistance in pest arthropods” (MPR), which are usually perceived to be distinct topics for research, should be considered to have equivalent goals and methods. When we accept that the goals and tactics are similar, we will develop effective resistance management programs for arthropods. Effective management of resistance and effective IPM programs require an holistic and multitactic strategy. A key component of this holistic and multitactic approach includes enhancing the compatibility of pesticides and biological control agents (Hoy 1992). Resistance to pesticides is an extremely signiﬁcant problem internationally, na- tionally (Georghiou 1986, Roush & Tabashnik 1990), and within Florida (Leibee & Capinera, this volume). At least 440 arthropod species have become resistant to insecticides and acaricides, with many species having become resistant to all the major classes of such products (Georghiou & Saito 1983, Georghiou 1986, Roush & Tabash- nik 1990). Resistance to pesticides in weeds, plant pathogens, and nematodes also is increasing, although somewhat more slowly (National Academy of Sciences 1986, Denholm et al. 1992). While my commentary focuses on resistance to insecticides and acaricides, it will probably be applicable to fungicides, herbicides, and nematicides. Developing and registering a new pesticide is an elaborate, and increasingly expensive, business in the USA (Georghiou 1986) with costs estimated to be more than

Hoy: Symposium on Managing Resistance 445

$60 million per successful compound. Thus, pesticide producers should be increasingly interested in extending the economic life of their products in order to maximize a return on their investment. Likewise, most pest management specialists want to preserve registered pesticides. This is especially true for products that are effective against arthropod pests in minor crops, which are increasingly being ignored by pesticide companies because they are such a small market. These so-called ‘minor’ crops of fruits and vegetables are a major component of Florida’s agriculture. Registration of new pesticides is likely to be more difﬁcult and expensive in the future, which could leave some pest management specialists with extremely limited options for managing certain recalcitrant pests. A few environmentalists have argued that we do not need pesticides, that they will soon be outlawed, and that pesticide resistance will no longer be an important issue. However, it is unrealistic to eliminate all pesticides from agriculture; there are significant arthropod pests for which we have no other effective control tactic. Pesticides are the most effective tools for ﬁghting outbreaks and emergency arthropod pest problems, and they are often required to control plant pathogens, weeds, or nematodes that cannot be controlled by alternative methods.

RESEARCH APPROACHES TO RESISTANCE MANAGEMENT Scientists have approached the problem of pesticide resistance in a variety of ways. Fundamental research over the past 40 years has produced insights into resistance mechanisms (Corbett et al. 1984, Scott 1990, Soderlund & Bloomquist 1990) and the mode of inheritance of resistance in arthropods (Georghiou & Saito 1983, Scott 1990, Soderlund & Bloomquist 1990). Simulation models have been developed to evaluate different options for managing resistance (for a recent review, see Tabash- nik 1990), but the debate over whether to recommend (1) alternation of different pesticides or (2) mixtures of different pesticides for slowing the development of resistance remains controversial and field-tested experimental data available are not strong enough to support either model (Roush & Daly 1990, Tabashnik et al. 1992). The hypothesis that reduced fitness, which is often associated with resistance alleles, could be used in management programs continues to be controversial and may have limited application (Tabashnik 1990). Not all resistance alleles confer lowered fitness (for example, Hoy & Conley 1989, Hoy 1990) and natural selection can select for modifying genes that restore fitness to individuals carrying resistance alleles. Various monitoring techniques have been developed to identify resistant individuals and detect their establishment and spread (ffrench-Constant & Roush 1990). These methods are particularly useful for documenting that resistance has occurred. However, monitoring methods that would allow us to detect rare resistant individuals in populations in sufficient time that operational programs could be altered remain difficult and expensive to execute (Brent 1986). Resistance management research programs and IPM research programs have had fairly distinct identities to date (Denholm et al. 1992, Croft 1990b, Hoy 1992). Because they have been distinct, an effective paradigm for resistance management has not been adopted in US agriculture. The current scenario usually goes something like this: A pesticide is registered and used, resistant individuals are detected in populations, people begin to discuss developing and implementing a resistance management program. With this short-sighted approach, it is exceedingly difficult to develop and execute a program in sufficient time to have the desired results. Developing a resistance management program can take several years; studies typically are conducted to develop an appropriate monitoring method, estimate the fre-

446 Florida Entomologist 78(3) September, 1995 quency of resistant individuals in populations, detect cross resistances, and evaluate mode of inheritance and stability of the resistance (Roush & Daly 1990). Meanwhile, unless pesticide applications are discontinued, selection for resistance continues. Ini- tial detection of resistance usually requires that resistant individuals comprise at least 5% of the population (Brent 1986). Thus, by the time resistant individuals are detected, selection by additional pesticide applications is likely to increase their frequency in the population. This scenario is particularly familiar with multivoltine and highly fecund pests such as aphids, spider mites, whiteflies, and leafminers. Although we can learn from the experiences of scientists studying pesticide resistance in ubiquitous pests in other geographic regions and thus be alerted to a potential problem, this seems to be an inefficient method for managing resistance in arthropods. Furthermore, this approach may be misleading because geneticists recognize that different species or different geographic populations may develop resistance to a particular toxic chemical by a variety of mechanisms. The mechanisms, their mode of inheritance, and the degree of reduction in fitness associated with them may vary, because the resistance alleles at each site are different. Because it is difficult to sample for rare individuals in natural populations, monitoring programs may not be cost effective if employed other than as a method to document a problem once it has developed. Waiting until the pest becomes resistant before instituting a resistance management program is ineffectual (Hoy 1992).

RESISTANCE AND IPM A better paradigm for managing resistance in arthropods involves altering pesticide use patterns, and nearly everyone will agree that reducing pesticide use is an effective resistance management tactic (Croft 1990a, Tabashnik 1990, Leeper et al. 1986). What has not been widely acknowledged is that resistance management programs should include: 1) Altering the way pesticides are developed and registered, and 2) Recognizing that resistance management must be a broad-based, multitactic endeavor (Hoy 1992, Fig. 1). It seems reasonable, conservative, and fiscally-responsible to assume that nearly all major insect and mite pests will eventually become resistant to all classes of pesticides given sufficient selection pressure over sufficient time. There may be some exceptions, but this generalization is reasonable given the documented record of resistance development in arthropod pests during the past 40 years. Resistance to stress is a fundamental and natural response by living organisms (Scott, this volume). On an evolutionary time scale, it is apparent that insects will develop multiple and diverse mechanisms to survive extreme temperatures, allelochemicals, and other environmental stresses. Thus, we should expect most insects to develop resistance to most pesticides, if subjected to appropriate and sustained selection. While new pesticide classes have been proclaimed to be potential ‘silver bullets’, and not amenable to resistance development, these hopes have been misplaced to date. It seems appropriate to assume that the development of resistance is nearly inevitable and the issue is not whether resistance will develop, but when. With this assumption, resistance management programs have the goal of delaying rather than preventing resistance (Hoy 1992). Growers and pest management experts can not afford to rely on pesticides as their primary management tool, as has been done for the past 30 years. There are increasing social, economic, and ecological pressures to reduce pesticide use and to increase the use of nonchemical control tactics such as host plant resistance, biorational methods, cultural controls, and biological controls (National Research Council 1989, Office of Technology Assessment 1992). There is an increasing priority by research scien-

Hoy: Symposium on Managing Resistance 447

Figure 1. Effective resistance management programs will incorporate multiple tactics as part of a fully integrated IPM program. Pesticides should be used only when needed, at the lowest rates possible, and in a manner to reduce negative impacts on arthropod natural enemies. Pesticide labels should include information about its impact on natural enemies. tists, regulatory agencies, legislators, and the public on using pesticides that are nontoxic to biological control agents and have minimal impact on the environment. Compatibility of pesticides with natural enemies and other nonchemical tactics is critical for improving pest management and environmental quality, and for managing resistance to pesticides. Enhancing the compatibility of pesticides and biological control agents is complex and sometimes difﬁcult (Croft 1990a, Hoy 1985a, 1990), but can reward us with handsome dividends in improved pest control (Metcalf 1994) and pesticide resistance management (Tabashnik & Croft 1985). The way pesticides are registered should be changed as part of an effective resistance management strategy (Hoy 1992). These changes also are essential in achieving improved integrated pest management. For example, some pesticides are relatively nontoxic to important natural enemies in cropping systems at low rates, but the recommended application rates are too high (Hoy 1985b). These high rates disrupt effective biological control, leading to additional pesticide applications, which exert unnecessary selection for resistance in the pest. Under these circumstances, it may be appropriate for the label to contain two different directions for use; one rate could be recommended for the traditional strategy of relying on pesticides to provide control (although this is becoming a less and less viable option). A lower rate could be recommended for use in an IPM program that employs effective natural enemies. This approach to labeling could reduce the number of pesticide applications and reduce rates, resulting in reduced selection for resistance in both target and nontarget pests. Another innovation in pesticide registration would require that the toxicity of the pesticide to a selected list of biological control agents be determined for each cropping system. This information should be provided, either on the label or in readily-available computerized data bases, perhaps via the internet. Without such information, use of these pesticides could disrupt beneﬁts derived from effective biological control agents.

448 Florida Entomologist 78(3) September, 1995

This often results in unnecessary use of pesticides, leading to more rapid evolution of pesticide resistance. Enhancing biological control not only leads to improved pest management, but is also an essential tool in managing pesticide resistance. How could information about the toxicity of pesticides to biological control agents be made available to the end user most effectively? How should bioassays be conducted to evaluate pesticide selectivity? There are no simple answers to these questions. Theiling & Croft (1988) and Croft (1990b) compiled an extensive set of data on the impact of pesticides on natural enemies, but additional data are also buried in many publications or reports that are difficult to find. Unfortunately, even if the data can be found, it is not always easy to interpret bioassay data obtained by different scientists using different methods. Thus, different conclusions about the toxicity of pesticides to natural enemies can be drawn. Also, it is often difficult to predict the impact of pesticides under field conditions based on laboratory assays (Hoy 1990, Hassan et al. 1991, Robertson & Preisler 1992). Consequently, the recommendation that labels or data bases be developed with information on the impact of pesticides on natural enemies requires considerable discussion and additional research. Should pesticide companies conduct the research using standard bioassay methods? Should a consor- tium of pest management scientists conduct the assays? Who should pay for the research? What species of natural enemies should be tested? These questions are not new, and in Europe standardized bioassays already are being conducted on selected natural enemy species by a scientific working group (Hassan et al. 1991). Whether this concept can be imported to the USA and Florida should be explored. Increased international consultation and cooperation between scientists, regulatory agencies, and pesticide companies could resolve many of the questions raised above. Evolution of pesticide resistance has been shown by computer simulations of predator and prey systems to be slowed by reduced pesticide use (Tabashnik & Croft 1985, Tabashnik 1990). There is general agreement that reduced pesticide use is one of the essential elements of any resistance management program (Croft 1990b, Tabashnik 1990, Metcalf 1994). Thus, the compatibility of pesticides and biological control agents is a crucial issue in pesticide resistance management. Attempts to manage pesticide resistance generally has involved making relatively minor tactical shifts in use patterns. What we need is a major shift in thinking about pesticide development and use, if we are to develop effective resistance management tactics. It is time to recognize that effective resistance management begins before the product is even registered (Figure 1). The strategy thus should be to manage the pesticide, even before it is fully developed and registered, with the goal of delaying resistance development. If this farsighted strategy is adopted, decisions on application rates and the numbers of applications per growing season will be made with the understanding that they affect the speed with which resistance will develop. In some cases, new products may not be developed because they are toxic to biological control agents and thus could disrupt effective IPM programs already in place. This would only speed up the development of resistance in specific pests. Some people suggest that financial incentives may have to be provided to induce pesticide companies to develop and register products that are harmless to biological control agents (= physiological selectivity). Others argue that many pesticides can be applied in a manner that affords substantial selectivity (= ecological selectivity) if the timing, location, rates, and methods of application are altered (Hull & Beers 1985). Relying on ecological selectivity is more likely to be cost effective than developing large numbers of special use products with physiological selectivity. However, a complete financial analysis may indicate that, over the long term, selective pesticides (based both on physiological and ecological selectivity) are among the most cost effective approaches to managing resistance to pesticides.

Hoy: Symposium on Managing Resistance 449

THE ROLE OF PESTICIDE-RESISTANT NATURAL ENEMIES Pesticide-resistant natural enemies are a special category of pesticide selectivity. Relatively few natural enemies have developed resistance to pesticides through natural selection, but several have been employed in effective IPM programs (Croft 1990a, Hoy 1990). Artiﬁcial selection of phytoseiid predators for pesticide resistance can be a practical and cost effective tactic for the biological control of spider mites (Hoy 1985a, b, 1990). Field tests have been conducted with several laboratory-selected phytoseiid species and some are being used in IPM programs. Laboratory selection of resistant strains of parasitoids and insect predators currently lags behind efforts with predatory mites, but several laboratory-selected insect natural enemies are being evaluated for incorporation into integrated pest management programs (Hoy 1994). The use of mu- tagenesis and recombinant DNA techniques could improve the efﬁciency of genetic improvement projects. However, development of pesticide-resistant natural enemies is time consuming and expensive and should not be considered before exploring other, less expensive, options for IPM and pesticide resistance management.

CONCLUSIONS We need to maintain a source of pesticides because they are powerful and effective pest management tools. Pesticides can be highly selective, rapid in their impact, adaptable to many situations, and relatively economical. Thus, preserving pesticides is essential. Effective paradigms for resistance management are not yet deployed in US and Florida agriculture. This is because resistance management and IPM have been considered separate issues. We need to recognize that effective resistance management is based on the development of effective, fully-integrated, multitactic IPM programs. Such programs will acknowledge the role of biological control, host plant resistance, cultural controls, and biorational controls, such as mating disruption, insect growth regulators, and mass trapping. A key issue should always be whether pesticides can be used in a precise and selective manner without disrupting the impact of natural enemies. While altering the way in which pesticides are registered and labeled is difﬁcult, the potential beneﬁts are great for both IPM and resistance management programs. The dialogue should begin on how best to change pesticide labeling and develop data- bases on the selectivity of pesticides to natural enemies.

ACKNOWLEDGMENTS This is University of Florida, Institute of Food and Agricultural Sciences, journal series number R-04510. I thank D. C. Herzog for comments on the manuscript and the Florida Entomological Society for the opportunity to organize the symposium, “Myths of Managing Resistance”. A brief essay containing many of the ideas presented here has been published (Hoy 1992). The references cited provide an entre to the extensive literature on resistance and are not complete.

REFERENCES CITED

BRENT, K. J. 1986. Detection and monitoring of resistant forms: an overview, pp. 298-312 in Pesticide Resistance: Strategies and Tactics for Management. Na- tional Academy Press, Washington, D.C. CORBETT, J. R., K. WRIGHT AND A. C. BAILLIE. 1984. The Biochemical Mode of Action of Pesticides. 2nd ed., Academic Press, New York.

450 Florida Entomologist 78(3) September, 1995

CROFT, B. A. 1990a. Arthropod Biological Control Agents and Pesticides. Wiley-Interscience Publ., New York. CROFT, B. A. 1990b. Developing a philosophy and program of pesticide resistance management, pp. 277-296 in R. T. Roush and B. E. Tabashnik [eds.] Pesticide Resistance in Arthropods, Chapman and Hall, New York. DENHOLM, I., A. L. DEVONSHIRE AND D. W. HOLLOMON [eds.]. 1992.Resistance 91: Achievements and Developments in Combating Pesticide Resistance, Elsevier Applied Science, London. FFRENCH-CONSTANT, R. H. AND R. T. ROUSH. 1990. Resistance detection and documentation: the relative roles of pesticidal and biochemical assays. pp. 4-38. GEORGHIOU, G. P. 1986. The magnitude of the resistance problem, pp. 14-43 in R. T. Roush and B. E. Tabashnik [eds.] Pesticide Resistance in Arthropods, Chap- man and Hall, New York. GEORGHIOU, G. P. AND T. SAITO [eds.] 1983. Pest Resistance to Pesticides, Plenum Press, New York. HASSAN, S. A., F. BIGLER, H. BOGENSCHUTZ, E. BOLLER, J. BRUN, J. N. M. CALIS, P. CHIVERTON, J. COREMANS-PELSENEER, C. DUSO, G. B. LEWIS, F. MANSOUR, L. MORETH, P. A. OOMEN, W. P. J. OVERMEER, L. POLGAR, W. RIECKMANN, L. SAMSOE-PETERSEN, A. STAUBLI, G. STERK, K. TAVARES, J. J. TUSET AND G. VIG- GIANI. 1991. Results of the fifth joint pesticide testing programme carried out by the IOBC/WPRS-Working Group “Pesticides and Beneficial Organisms”. Entomophaga 36: 55-67. HOY, M. A. 1985a. Almonds: integrated mite management for California almond orchards, pp. 299-310 in W. Helle and M. W. Sabelis [eds.]. Spider Mites, Their Bi- ology, Natural Enemies, and Control, Vol. 1B. Elsevier, Amsterdam. HOY, M. A. 1985b. Recent advances in genetics and genetic improvement of the Phy- toseiidae. Annu. Rev. Entomol. 30: 345-370. HOY, M. A. 1990. Pesticide resistance in arthropod natural enemies: variability and selection responses, pp. 203-236 in R. T. Roush and B. E. Tabashnik [eds.] Pes- ticide Resistance in Arthropods, Chapman and Hall, New York. HOY, M. A. 1992. Guest editorial: Proactive management of pesticide resistance in agricultural pests. Phytoparasitica 20(2): 93-97. HOY, M. A. 1994. Transgenic pest and beneficial arthropods for pest management programs: an assessment of their practicality and risks, pp. 641-670 in D. Rosen, F. D. Bennett and J. L. Capinera [eds.] Pest Management in the Subtropics: Bi- ological Control a Florida Perspective. Intercept Ltd., Andover, U.K. HOY, M. A. AND J. CONLEY. 1989. Propargite resistance in Pacific spider mite (Acari: Tetranychidae): stability and mode of inheritance. J. Econ. Entomol. 82: 11-16. HULL, L. A. AND E. H. BEERS. 1985. Ecological selectivity: modifying chemical control practices to preserve natural enemies, pp. 103-122 in M. A. Hoy and D. C. Her- zog [eds.], Biological Control in Agricultural IPM Systems. Academic, Orlando. LEEPER, J. R., R. T. ROUSH AND H. T. REYNOLDS. 1986. Preventing or managing resistance in arthropods, pp. 335-346 in Pesticide Resistance: Strategies and Tactics for Management, National Academy Press, Washington, D.C. METCALF, R. L. 1994. Insecticides in pest management, pp. 245-284 in R. L. Metcalf and William H. Luckmann [eds.], Introduction to Insect Pest Management, third ed., John Wiley and Sons, New York. NATIONAL ACADEMY OF SCIENCES. 1986. Pesticide Resistance: Strategies and Tactics for Management. National Academy Press, Washington, D.C. NATIONAL RESEARCH COUNCIL. 1989. Alternative Agriculture, National Academy Press, Washington, D. C. OFFICE OF TECHNOLOGY ASSESSMENT. 1992. A New Technological Era for American Agriculture, U. S. Congress, U.S. Government Printing Office, Washington, D.C. ROBERTSON, J. L. AND H. K. PREISLER. 1992. Pesticide Bioassays with Arthropods. CRC Press, Boca Raton, Florida. Hoy: Symposium on Managing Resistance 451

ROUSH, R. T. AND J. C. DALY. 1990. The role of population genetics in resistance research and management, pp. 97-152 in R. T. Roush and B. E. Tabashnik [eds.] Pesticide Resistance in Arthropods, Chapman and Hall, New York. ROUSH, R. T. AND B. E. TABASHNIK [eds.] 1990. Pesticide Resistance in Arthropods. Chapman and Hall, New York, NY. SCOTT, J. G. 1990. Investigating mechanisms of insecticide resistance: methods, strategies, and pitfalls, pp. 39-57 in R. T. Roush and B. E. Tabashnik [eds.] Pesticide Resistance in Arthropods, Chapman and Hall, New York. SODERLUND, D. M. AND J. R. BLOOMQUIST. 1990. Molecular mechanisms of insecticide resistance, pp. 58-96 in R. T. Roush and B. E. Tabashnik [eds.] Pesticide Resis- tance in Arthropods, Chapman and Hall, New York. TABASHNIK, B. E. 1990. Modeling and evaluation of resistance management tactics, pp. 153-182 in R. T. Roush and B. E. Tabashnik [eds.] Pesticide Resistance in Arthropods, Chapman and Hall, New York. TABASHNIK, B. E. AND B. A. CROFT. 1985. Evolution of pesticide resistance in apple pests and their natural enemies. Entomophaga 30: 37-49. TABASHNIK, B. E., J. A. ROSENHEIM AND M. A. CAPRIO. 1992. What do we really know about management of insecticide resistance? pp. 124-135 in I. Denholm, A. L. Devonshire and D. W. Hollomon [eds.]. 1992. Resistance 91: Achievements and Developments in Combating Pesticide Resistance, Elsevier Applied Science, London. THEILING, K. M. AND B. A. CROFT. 1988. Pesticide side-effects on arthropod natural enemies: a database summary. Agric. Ecosys. Environm. 21: 191-218.

452 Florida Entomologist 78(3) September, 1995

IMMUNOLOGICAL STUDY OF JUVENILE HORMONE BINDING PROTEIN FROM HEMOLYMPH OF THE FALL WEBWORM (LEPIDOPTERA: ARCTIIDAE)

I. H. LEE1, HAK R. KIM1 AND RICHARD T. MAYER2 1Department of Biology, Korea University, Seoul, Korea

2USDA, ARS, U.S. Horticultural Research Laboratory, 2120 Camden Road, Orlando, FL 32803-1419

ABSTRACT The juvenile hormone binding protein (JHBP) was puriﬁed from the hemolymph of Hyphantria cunea (Lepidoptera: Arctiidae) using anion exchange, gel ﬁltration, and Mono P FPLC chromatofocusing chromatography. The protein is a single polypeptide µ (Mr 32,000) with an apparent dissociation constant of 0.43 M for juvenile hormone III. An antibody developed against hemolymph JHBP (hJHBP) was prepared, and use of it showed that a protein immunologically identical with hJHBP occurs in fat body and ovary. The hJHBP of H. cunea was neither immunologically related to the lipophorin from H. cunea nor similar to the hJHBP of Bombyx mori and Periplaneta americana.

Key Words: Hemolymph, hemolymph proteins, insect, immunodetection, juvenile hormone III, lipoprotein, protein puriﬁcation.

RESUMEN

La proteína de unión de la hormona juvenil (JHBP) fue purificada a partir de la hemolinfa de Hyphantria cunea. Para purificar la JHBP fueron usados el intercambio de aniones, la filtración en gel y cromatografías cromatofocales Mono P FPLC. La pro- teína es un polipéptido simple (Mr 32,000) con una constante de disociación aparente de 0.43 µM para la hormona juvenil III. Fue preparado un antibiótico desarrollado contra la hemolinfa JHBP (hJHBP) y su uso demostró que una proteína inmunológi- camente idéntica a la hJHBP está presente en el cuerpo graso y el ovario. La hJHBP de H. cunea no estuvo inmunológicamente relacionada con la lipoforina de H. cunea ni fue similar a la hJHBP de Bombyx mori y Periplaneta americana.

Juvenile hormone is synthesized by the corpora allata, released into hemolymph, and transported to target tissues by the juvenile hormone binding protein (JHBP) (Whitmore & Gilbert 1972; Rudnica et al. 1979; Ozyhar et al. 1983). JHBP is synthesized by fat body and released into the hemolymph (Nowock et al. 1976; Ferkovich et al. 1977). Generally, high Mr (over 200,000) and low Mr (20,000-40,000) JHBPs are present in the hemolymph (de Kort & Granger 1981; de Kort et al. 1983). JHBP from hemolymph was first purified from Manduca sexta (Kramer et al. 1976) and then from Diatraea grandiosella (Dillwith et al. 1985), Diploptera punctata (King & Tobe 1988), and Platyprepia virginalis (Prestwich & Atkinson 1990). It has been reported that hemolymph juvenile hormone binding protein (hJHBP) is closely related to apoprotein-I of lipophorin (Rayne & Koeppe 1988; Koopmanschap & Dekort 1988). Proteins with JH affinity were found to be present in ovary and fat body and their characteristics were also reported (van Mellaert et al. 1985; Koeppe et al. 1987; Shem-

Lee et al.: JHBP of H. cunea 453 shedini & Wilson 1993). However, information on the relationship between hJHBP and JH binding protein in these tissues is still limited, thus additional information is needed for a better understanding of the role of JHBP. Here, we describe the presence of a JHBP in ovary and fat body which is immunologically identical with hJHBP by using an antibody against low Mr hJHBP from the fall webworm, Hyphantria cunea Drury (Lepidoptera: Arctiidae). In addition, the physicochemical characteristics of hJHBP are described.

MATERIALS AND METHODS

Insects Fall webworms, Hyphantria cunea, were reared on artiﬁcial diet (Dong Bang Oil Co., Ltd., Seoul, Korea) at 28 ± 1°C and 70 ± 5% relative humidity and with a photoperiod of 16:8 (L:D).

Collection of Hemolymph and Extraction of Ovary and Fat Body Hemolymph was collected into a cold Eppendorf microcentrifuge tube from last instar larvae by cutting the forelegs. A few crystals of phenylthiourea were added to prevent melanization, and the hemolymph was centrifuged at 10,000 g for 10 min at 4°C to remove hemocytes and cell debris. The supernatant was stored at -70°C. Ovaries were dissected from 5-day-old pupae in cold Ringer’s solution (128 mM

NaCl, 1.8 mM CaCl2, 1.3 mM KCl; pH 7.4). They were used directly for electron microscopic observation or homogenized and centrifuged at 10,000 g for 10 min and the supernatant used as the sample. Fat body was dissected from last instar larvae in Ringer’s solution. These tissues were stored at -70°C until used.

Dextran Coated Charcoal (DCC) Binding Assay

A DCC suspension (0.5% active charcoal and 0.05% dextran, Mr 80,000) in buffer

(10 mM Tris, 5 mM MgCl2, 50 mM KCl, pH 7.4) was used to separate protein-bound JH from unbound hormone (Engelmann 1981). [3H]JH III (NEN Corp., Wilmington, DE; speciﬁc activity: 440 GBq/mmol; 50,000 cpm) was put into a microcentrifuge tube and the solvent evaporated with a gentle stream of nitrogen gas. Buffer (100 µl; 10 mM Tris, 5 mM MgCl2, 50 mM KCl, pH 7.4), and the sample used for the binding assay, were added to the hormone and incubated for 1 h at 4°C. DCC suspension (100 µl) was added to the tube, the contents incubated for 2 min and then centrifuged at 10,000 g to remove free hormone. An aliquot (10 µl) of supernatant was transferred to 10 ml of scintillation cocktail solution (toluene 2 liters, triton X-100 1 liter, 12 g Om- niﬂuor, NEN) and the radioactivity was measured by liquid scintillation counter (Beckman LS 100°C, Palo Alto, CA).

Puriﬁcation of JHBP

Anion exchange chromatography. Hemolymph (10 ml) was diluted 1:1 with TPN buffer [10 mM Tris, 0.1 mM phenylmethylsulfonyl ﬂuoride (PMSF), 0.01% NaN3, pH 8.2] and dialyzed against the same buffer for 4 h. Dialyzed hemolymph was applied to a DEAE cellulose (DE 52, Whatman, Hillsboro, OR) column (1.8 × 20 cm) and eluted with 1 column volume of TPN buffer. This was followed by a NaCl concentration gradient (0-0.5 M) in buffer (10 mM Tris, 0.01% NaN3, 0.1 mM PMSF, pH 8.2, total 300

454 Florida Entomologist 78(3) September, 1995 ml) at a flow rate of 18 ml per h and then monitored for the presence of JHBP by the DCC binding assay. Fractions containing JHBP were pooled, dialyzed against 10 mM Tris buffer (0.01% NaN 0.1 mM PMSF, pH 8.2) and applied to a Sepharose CL-6B (Pharmacia, Piscataway, NJ) column (0.9 × 15 cm). The column was washed with the same buffer using a gradient of 0.05-0.35 M NaCl. Each fraction (2.6 ml) was monitored for JHBP by the DCC binding assay. Fractions containing JHBP were lyophilized before the next purification step. Gel Filtration. The lyophilized hemolymph sample was dissolved in 2 ml water and applied to a Sephadex G-100 column (1.8 × 55 cm) and eluted with buffer (25 mM bis-Tris, pH 7.2) at a flow rate of 12 ml per h. Each fraction (2.6 ml) was monitored for JHBP by the DCC binding assay. Fractions containing JHBP were pooled, concentrated by lyophilization, and applied to a Sephadex G-75 column (0.9 × 20 cm) and eluted with buffer (25 mM bis-Tris, pH 7.2) at a flow rate of 8 ml per h. Fractions of 1.5 ml were collected. Mono P FPLC Chromatofocusing Chromatography. Fractions containing JHBP from the Sephadex G-75 column were subjected to FPLC chromatofocusing chromatography on a Mono P column (Mono P HR 5120, 5 × 200 mm, Pharmacia). Free proteins were eluted from the column with 25 mM bis-Tris buffer (pH 7.2, 10 ml), and then Polybuffer (pH 7.4, 52 ml) was used to elute resin-bound proteins. Each fraction was monitored for JHBP by the DCC binding assay.

Determination of Molecular Weight and Isoelectric Point The molecular weights were determined on sodium dodecyl sulfate-polyacrylamide gel (10%) as described by Weber & Osborn (1969). Standard molecular weight markers were: bovine serum albumin, Mr 66,200; hen egg albumin, Mr 45,000; bovine carbonic anhydrase, Mr 31,000; soybean trypsin inhibitor, Mr 21,500; hen egg white lysozyme,

Mr 14,400. Isoelectric focusing was performed on 5% polyacrylamide gel according to Klages & Emmerich (1979) by using ampholytes in the pH range 3-10 (Sigma, St. Louis, MO).

Production of Antibody Against JHBP Fractions (0.5 ml) containing JHBP after Mono P FPLC chromatofocusing were mixed thoroughly with Freund’s complete adjuvant (0.5 ml) and injected subcutane- ously into a rabbit. Injections were repeated every other day for the ﬁrst week. The fourth injection was conducted 1 week after the third injection, and a booster injection was given 2 weeks later. JHBP-containing fractions (0.5 ml) were mixed with incomplete adjuvant (0.5 ml) for the booster injection. Blood was taken 1 week after the booster injection and centrifuged at 10,000 g for 10 min. Antibody against JHBP was puriﬁed by immunoprecipitation of extraneous antibodies using JHBP-free fractions.

Immunodiffusion Immunodiffusion was conducted using 1% agarose gel containing 0.1% (w/v) sodium azide in veronal buffer (pH 8.6) as described by Ouchterlony (1949). The plates were stained in 1% amido black 10B and destained in 2% acetic acid.

Electrophoresis and Immunodetection Fractions from each puriﬁcation step were electrophoresed on 10% SDS-polyacrylamide gel as described by Laemmli (1970) to determine purity. Gels were stained with silver nitrate according to Wray et al. (1981).

Lee et al.: JHBP of H. cunea 455

Western blotting was used to detect JHBP on SDS-polyacrylamide gels. Samples were electrophoresed as above and transferred to nitrocellulose sheets in Tris-glycine buffer (25 mM Tris, 92 mM glycine, 30% methanol, pH 8.3) at 100 V for 2 h (Towbin et al. 1979). After transfer, the nitrocellulose sheets were equilibrated in TBS (20 mM Tris, 500 mM NaCl, pH 7.5) for 10 min and incubated in blocking solution (3% gelatin in TBS) for 30 min. The sheet was then washed with TTBS (0.05% tween 20 in TBS) twice, each time for 5 min, and incubated for 1 h in a solution containing a 300-fold dilution of the primary antibody against JHBP. This sheet was again washed with TTBS twice and incubated for 1 h in a solution containing a 3000-fold diluted secondary antibody (goat antirabbit-horseradish peroxidase conjugated IgG). After incubation, the sheet was again washed with TTBS twice and submerged in horseradish peroxidase color development reagent containing 4-chloro-1-naphthol in 20 ml ice-cold methanol + 0.015% H2O2 in 100 ml TBS for development of purple color.

Electron Microscopic Observation by Immunogold Labeling Ovaries and fat body were dissected from pupae in Ringer’s solution and preﬁxed in 2.5% glutaraldehyde for 2 h at 4C. Tissues were washed with 0.1 M phosphate buffer (pH 7.2) 3 times at 15 min intervals and dehydrated in an ethanol series. De- hydrated tissues were then put into propylene oxide and embedded in an Epon-Araldite mixture. Embedded tissues were semithin-sectioned using an ultrami- crotome (Sorvall MT-II, Wilmington, DE) and stained in 1% toluidine blue and attached to a grid. The thin section-attached grid was washed with TBS. This grid was reacted with the primary antibody solution 30-fold diluted with antibody buffer and again washed with TBS 3 times to remove nonspeciﬁcally attached antibody. This grid was then reacted with antirabbit IgG combined with protein-A gold particles (30 nm in diam) for 40 min, washed with TBS and distilled water 3 times each, stained in 2% uranyl acetate and observed under JEOL JEM 100 CX-II electron microscope at 80 kV.

RESULTS

Purification of JHBP from Hemolymph The DCC binding assay for each fraction showed that JHBP was eluted from the DE-52 column with a linear NaCl gradient (0.06-0.1 M) (Fig. 1A). Fractions containing JHBP were subjected to a second anion exchange chromatography treatment (Sepharose CL-6B). JHBP was found to be present in the backside of the first peak (0.1 M NaCl) (Fig. 1B). Fractions containing JHBP from the second ion exchange chromatography run were subjected to gel filtration (Sephadex G-100) and fractions containing radioactivity were located (Fig. 1C) and pooled. The pooled fractions were applied to Mono P FPLC chromatofocusing columns and 4 peaks isolated by using A280 (Fig. 1D). The fourth peak exhibited high binding activity for [3H]JH III (Fig. 1D, inset). Because the JHBP fractions were contaminated with other hemolymph proteins after Mono P chromatofocusing, gel filtration on Sephadex G-75 was necessary (data not shown). Fractions were applied to a 10% SDS gel and completely purified JHBP was confirmed (Fig. 2A).

Characterization of hJHBP JHBP was electrophoresed with a low molecular weight standard marker to determine the Mr of the JHBP subunit. The Mr was estimated to be 32,000 (Fig. 2B). Also,

456 Florida Entomologist 78(3) September, 1995

Figure 1. Chromatography elution profiles for purification of H. cunea hemolymph JHBP. A, the first anion exchange chromatography of dialyzed hemolymph proteins on DE52. Hatched region (35-46) indicates fractions containing JHBP; B, the second anion exchange chromatography on Sepharose CL-6B. Fractions (47-55, hatched region) showing radioactivity in DCC binding assay were pooled and lyophilized for the next purification step; C, gel (Sephadex G-100) filtration of partially purified JHBP. JHBP containing fractions were confirmed (hatched region, 31-37); D, Mono P FPLC chromatofocusing column. Four major protein peaks were resolved and collected (fractions 26-29, 32, 35, 37-38). Each peak was also analyzed by DCC binding assay (inset). Arrow indicates peak containing JHBP. the pI was determined to be 5.3 (data not shown). Saturation analysis of the purified µ hJHBP revealed a high affinity for JH III (KD of 0.43 M) (Fig. 3).

Immunological Studies Antibodies were made against fractions containing JHBPs (Fig. 4A, lane b) that had been separated by Mono P FPLC chromatofocusing. Extraneous antibodies, i.e., other than antibody against hJHBP, were precipitated with JHBP-free fractions (Fig. 4A, lane a). The resulting supernatant from this precipitation showed pure hJHBP antiserum (Fig. 4B). A protein that is immunologically identical with hJHBP was found in ovaries by immunodiffusion and western blotting (Fig. 5). However, the hJHBP of H. cunea was neither immunologically related to the lipophorin from H. cunea nor was it similar to the hJHBP of Bombyx mori and Periplaneta americana (Fig. 5). The western blots

Lee et al.: JHBP of H. cunea 457

Figure 2. A, electrophoretic profile of fractions from final purification step (gel filtration on Sephadex G 75). Anterior portions of the chromatographic profile still contain other proteins in addition to JHBP (A: a, b, c, d). Completely purified JHBP from the posterior portion of the chromatography was confirmed on lane e. The gel (10%) was stained with silver nitrate. B, SDS-polyacrylamide gel (15%) electrophoresis to determine the molecular weight of JHBP. The gel was stained with coomassie bril- liant blue. Right lane contains the low molecular weight standard markers and the arrow in left lane indicates JHBP. shown in Figure 5 (left) show a crossreaction for the H. cunea hemolymph, ovarian extracts, and JHBP with the antibody against hJHBP. Likewise, the immunodiffusion test (Fig. 5, right) showed an immunopreciptation line extending from the JHBP well through the ovarian extract and ending at the H. cunea hemolymph well, thus corrob- orating the western blot data. Electron microscopic observations showed that gold particles accumulated in ovarian protein bodies as well as the fat body (Fig. 6). These results clearly indicate protein that is immunologically identical with hJHBP is present in protein bodies of ovary.

DISCUSSION

The DCC binding assay was employed to purify a low Mr JHBP. Fractions showing radioactivity according to the DCC binding assay were electrophoresed and silver stained and found to be a single polypeptide with an Mr 32,000. The protocols employed to purify hemolymph JHBP from H. cunea are similar to those used to purify JHBP of Diatraea grandiosella (Dillwith et al. 1985) and Manduca sexta (Kramer et al. 1976). The hJHBP of H. cunea has a pI of 5.3 (data not shown) and can be effectively separated from other hemolymph proteins with chromatofocusing. Low Mr

458 Florida Entomologist 78(3) September, 1995

Figure 3. The saturation analysis (Scatchard plot) of the binding of JH III to the purified hJHBP. •••, total binding; °°°, nonspecific binding; ▲▲▲, specific binding. The specific binding was used for the Scatchard analysis (inset).

JHBP has an Mr range of 20,000-40,000 in several insects (de Kort & Granger 1981; de Kort et al. 1983). The dissociation constant of H. cunea hJHBP was estimated to be 0.43 µM. This value is similar to those of pyralid moth’s 0.08-0.28 µM (Lenz et al. 1986), D. grandiosella 0.31 µM, (Dillwith et al. 1985), and M. sexta, 0.44 µM (Kramer et al. 1976). The

JHBP puriﬁed from hemolymph of H. cunea is of low Mr and high afﬁnity.

Thus far, most JHBP antibody production employed large Mr proteins. Prepara- tion of antibodies for the low Mr JHBP was difﬁcult because of the small amount of Lee et al.: JHBP of H. cunea 459

Figure 4. Western blot of puriﬁed JHBP. The JHBP band was detected by enzyme immunoassay using antibody against JHBP (B,b). Left panel (A) shows the electrophoretic patterns of fractions from the Mono P FPLC chromatofocusing chromatography. The JHBP-containing fractions (the fourth peak that the arrow indicates in Fig. 1D, lane b) were used for the production of antibody and fractions (anterior three peaks in Fig. 1D, lane a) that lack JHBP were used for the immunoprecipitation to remove other antibodies except antibody against JHBP. 460 Florida Entomologist 78(3) September, 1995

Figure 5. Left panel: Western blots of proteins from the hemolymph (H), JHBP (J), ovarian extracts (O) of H. cunea, the hemolymph (B) of B. mori, the hemolymph (P) of P. americana. The JHBP bands were detected by enzyme immunoassay using antibody against the hemolymph JHBP. In three lanes (H,J,O), a single Mr 32,000 protein corresponding to the JHBP was detected by the antibody. Right panel: Immunodiffu- sion test to determine the relationship of several antigens (J, JHBP, 6 µg protein; Ov, ovarian extracts, 1.8 mg protein; H, hemolymph of H. cunea, 1 mg protein; B, hemolymph of B. mori 1.4 mg protein; L, lipophorin of H. cunea, 20 µg protein; P, hemolymph of P. americana, 1.2 mg protein; Ab, antibody against JHBP in hemolymph of H. cunea). material in hemolymph. In the present work with H. cunea, immunoprecipitation was very effective in purifying antibody against low Mr JHBP from other antibodies. hJHBP was confirmed to be lipophorin in Leucophaea maderae (Koeppe et al. 1988) and Locusta migratoria (Koopmanschap & de Kort, 1988). Immunodiffusion studies conducted here indicate that the hJHBP was neither immunologically related to lipophorin in H. cunea nor the hJHBP of Bombyx mori and Periplaneta americana. The hJHBP of H. cunea appeared to be species-specific. hJHBP is synthesized by fat body and released into hemolymph. Protein with JH affinity (JH binding protein) is also present in cytosol and nucleus of fat body (Shem- shedini & Wilson 1993). A JH binding protein has been reported in the ovary (van Mellaert et al. 1985), but it is not known how it is related to hJHBP. We found a protein that is immunologically related to hJHBP that appears in fat body and the cytosol of ovarian protein body of H. cunea. This protein is assumed to be a JHBP. More information on the relationship between hJHBP and JHBP in the cytosol is required to better understand the transport of JH from secretory cells to target tissues and receptors.

ENDNOTE Mention of a trademark, warranty, proprietary product, or vendor does not constitute a guarantee by the U.S. Department of Agriculture and does not imply its approval to the exclusion of other products or vendors that may also be suitable. Lee et al.: JHBP of H. cunea 461

Figure 6. Immunoelectron micrographs of fat body (A) and ovary (B) from H. cunea. Immunogold particles were observed in large protein bodies (PB). Bar length = 0.2 µm (A), 0.5 µm (B).

REFERENCES CITED

DE KORT, C. A. D., AND N. A. GRAGER. 1981. Regulation of the juvenile hormone titer. Annu. Rev. Entomol. 26: 1-28. 462 Florida Entomologist 78(3) September, 1995

DE KORT, C. A. D., M. G. PETER, AND A. B. KOOPMANSCHAP. 1983. Binding and degradation of juvenile hormone III by hemolymph proteins of the Colorado potato beetle: a re-examination. Insect Biochem. 13: 481-487. DILLWITH, J. W., S. D. MANE, AND G. M. CHIPPENDALE. 1985. High affinity juvenile hormone binding protein of the hemolymph of the southwestern corn borer, Di- atraea grandiosella. Characteristics and relation to diapause. Insect Biochem. 15: 233-246. ENGELMANN, F. 1981. The identification of juvenile hormone binding protein in the fat bodies of Leucophaea maderae, pp. 263-270 in G. E. Pratt and G. T. Brooks [eds.], Juvenile Hormone Biochemistry, Elsevier North Horland Biomedical, New York. FERKOVICH, S. M., H. OBERLANDER, AND R. R. RUTTER. 1977. Release of juvenile hormone binding protein by fat body of the Indian meal moth, Plodia interpunctella, in vitro. J. Insect Physiol. 23: 297-302. KING, L. E., AND S. S. TOBE. 1988. The identification of an enantioselective JH III binding protein from the hemolymph of the cockroach, Diploptera punctata. In- sect Biochem. 18: 793-805. KLAGES, G., AND H. EMMERICH. 1979. Juvenile hormone binding protein in the hemolymph of third instar larvae of Drosophila hydei. Nature 286: 282-285. KOEPPE, J. K., R. C. RAYNE, M. D. SHEARIN, D. J. CARVER, E. A. WHITSEL, AND R. C. VOGLER. 1988. Synthesis and secretion of a precursor hemolymph juvenile hormone-binding protein in the adult female cockroach Leucophaea maderae. Insect Biochem. 18: 661-666. KOEPPE, J. K., R. C. RAYNE, E. A. WHITSEL, AND M. E. POOLER. 1987. Synthesis and secretion of juvenile hormone-binding protein by fat body from the cockroach Leucophaea maderae: protein a immunoassay. Insect Biochem. 17: 1027-1032. KOOPMANSCHAP, A. B., AND C. A. D. DE KORT. 1988. Isolation and characterization of a high molecular weight JH III transport protein in the hemolymph of Locusta migratoria. Arch. Insect Biochem. Physiol. 7: 105-118. KRAMER, K. J., P. E. DUNN, R. C. PETERSON, H. L. SEBALLOS, L. SANBURG, AND J. H. LAW. 1976. Purification and characterization of the carrier protein for juvenile hormone from the hemolymph of the tobacco hornworm Manduca sexta Jo- hannson (Lepidoptera: Sphingidae). J. Biol. Chem. 251: 4979-4985. LAEMMLI, U. K. 1970. Cleavage of structural proteins during assembly of head of bac- teriophage T4. Nature 227: 680-685. LENZ, C. J., J. W. DILLWITH, AND G. M. CHIPPENDALE. 1986. Comparison of some properties of the high juvenile hormone binding protein from the larval hemolymph of pyralid moth. Arch. Insect Biochem. Physiol. 3: 61-73. NOWOCK, J., B. HAMMOCK, AND L. I. GILBERT. 1976. The binding protein as a modu- lator of juvenile hormone stability and uptake, pp. 354-372 in L. I. Gilbert [ed.], The Juvenile Hormone, Plenum Press, New York. OUCHTERLONY, O. 1949. Antigen-antibody reactions in gels. Acta Path. Microbiol. Scan. 26: 507-515. OZYHAR, A., J. R. WISNIEWSKI, F. SEHNAL, AND M. KOCHMAN. 1983. Age dependent changes in the binding and hydrolysis of juvenile hormone in the hemolymph of last instar larvae of Galleria mellonella. Insect Biochem. 13: 435-441. PRESTWICH, G. D., AND J. K. ATKINSON. 1990. Rapid purification and N-terminal amino acid sequence of a photoaffinity-labeled juvenile hormone binding protein from an arctiid moth larva, Platyprepia virginalis. Insect Biochem. 20: 801-807. RAYNE, R. C., AND J. K. KOEPPE. 1988. Relationship of hemolymph juvenile hormone binding protein to lipophorin in Leucophaea maderae. Insect Biochem. 18: 667-673. RUDNICA, M., F. SEHNAL, V. JAROLIN, AND M. KOCHMAN. 1979. Hydrolysis and binding of the juvenile hormone in the hemolymph of Galleria mellonella. Insect Biochem. 9: 569-575. Lee et al.: JHBP of H. cunea 463

SHEMSHEDINI, L., AND T. G. WILSON. 1993. Juvenile hormone binding proteins in larval fat body nuclei of Drosophila melanogaster. J. Insect Physiol. 39: 563-569. TOWBIN, H., T. STAEHELIN, AND J. GORDON. 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA 76: 4350-4355. VAN MELLAERT, H., S. THEUNIS, AND A. DE LOOF. 1985. Juvenile hormone binding proteins in Sarcophaga bullata hemolymph and vitellogenic ovaries. Insect Biochem. 15: 655-661. WEBER, K. AND M. OSBORN. 1969. The reliability of molecular weight determination by dodecyl sulfate-polyacrylamide gel electrophoresis. J. Biol. Chem. 244:4406- 4412. WHITMORE, E., AND L. I. GILBERT. 1972. Haemolymph lipoprotein transport of juvenile hormone. J. Insect Physiol. 18: 1153-1167. WRAY, W., T. BOULIKAS, V. P. WRAY, AND R. HANCOCK. 1981. Silver staining of proteins in polyacrylamide gels. Anal. Biochem. 118: 197-203.

♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦

BA & Phillips: Fire Ant Yolk Sphere Degradation 463

DEGRADATION OF RED IMPORTED FIRE ANT (HYMENOPTERA: FORMICIDAE) YOLK SPHERES

AMADOU S. BA AND SHERMAN A. PHILLIPS, JR. Department of Plant and Soil Science Texas Tech University Lubbock, TX 79409-2134

ABSTRACT

Transmission electron microscopy reveals that the eggs of the red imported ﬁre ant, Solenopsis invicta Buren, contain yolk spheres that are evenly distributed throughout the center of the egg. Yolk degradation occurs sequentially as evidenced by the simultaneous observation of three stages of yolk platelet degradation within the eggs.

Key Words: Insecta, Solenopsis invicta, egg, oogenesis.

RESUMEN

La microscopía electrónica de transmisión revela que los huevos de la hormiga de fuego Solenopsis invicta Buren, contienen esferas de yema que se distribuyen pareja- mente a lo largo del centro del huevo. La observación simultánea de tres etapas de de- gradación en las esferas de yema dentro de los huevos evidencia que la degradación de la yema ocurre secuencialmente.

The most abundant subcellular organelles of mature oocytes in arthropods are yolk granules, constituting approximately 80% of the total protein present (Medina et al. 1988). Arthropod yolk spheres (granules or platelets) are phosphoglycolipoproteins with varying amounts of phosphorus, lipids, and carbohydrates (Yamashito & Indra-

464 Florida Entomologist 78(3) September, 1995 sith 1988). Much of the yolk material consists of vitellin, a protein that is derived from vitellogenin (Kunkel & Nordin 1985; Tata & Smith 1979; Wallace 1985). Vitellin, which comprises 60-90% of the total mass of yolk proteins (Kunkel & Nordin 1985), is transported through insect haemolymph and accumulates as yolk granules in the developing oocyte by pinocytosis (Postlethwait & Giorgi 1985). The yolk protein composition of numerous arthropods, including many orders of the class Insecta, have been determined (Bownes & Hames 1977; Yamashita & Indrasith 1988). For example, yolk spheres of the fruit fly (Drosophila) constitute about 80% of the total protein of the mature oocyte (Vallejo et al. 1981). However, the degradation of yolk in arthropods is not a process that is well understood. The red imported fire ant, Solenopsis invicta Buren, was introduced into the United States from South America around 1940 and has rapidly become a major pest in the southern U.S.A. Although much is known of its basic biology and behavior, relatively little attention has been given to oogenesis, vitellogenesis, and embryogenesis. However, one study indirectly addressing these subjects was initiated to understand the mode of action of the insect growth regulator fenoxycarb on the red imported fire ant (Glancey & Banks 1988). Fenoxycarb caused a lack of cell differentiation involving oocytes, trophocytes, and follicular epithelial cells. In addition, because nurse cells did not develop, oocytes did not contain the necessary yolk spheres. Therefore, eggs did not develop, and the colony died due to lack of worker replacement (Glancey & Banks 1988). In that study, the sequestration of yolk in untreated eggs and the lack of yolk spheres in treated eggs with concomitant egg reabsorption were clearly displayed. A review of the literature reveals that study as the only one portraying yolk sphere formation in oocytes of the red imported fire ant. Our paper, herein, is the first to describe the appearance of naturally occurring yolk degradation in fertilized eggs of ants, and in particular, of the red imported fire ant.

MATERIALS AND METHODS

Red imported fire ant colonies were collected from Abilene and Victoria, Texas, in the spring of 1991 and maintained at approximately 28°C. Oviposited eggs were removed from colonies with a camel hair brush after the ants were immobilized by car- bon dioxide. Samples were fixed for three hours at room temperature in 2% glutaraldehyde, 0.1 M phosphate buffer (pH 7.0) and several drops of the wetting agent Triton-X. Eggs were subsequently washed in 0.1 M phosphate buffer (pH 7.0) and post-fixed with osmium tetroxide. Eggs were next washed in 0.1 M phosphate buffer (pH 7.0) and dehydrated through a graded ethanol series. To remove the ethanol, specimens were rinsed twice with 100% acetone and stored over a dehydrating material (CuSO4). Eggs were embedded in Spurr’s low viscosity medium (Spurr 1969). Sorvall MT-2B and glass knives were used for ultramicrotomy. Sections were stained with methanolic uranyl acetate and lead citrate, and micrographs were taken using a Hitachi HS-9 transmission electron microscope.

RESULTS AND DISCUSSION

Electron micrographs showed the eggs to be ﬁlled with large and dense storage organelles (yolk spheres). Yolk platelets were distributed evenly throughout the center of eggs, whereas the periphery of eggs were relatively free of yolk granules (Fig. 1B). However, early embryonic cells were located at the periphery of the egg near the vi- telline membrane (Fig. 1A and B). The yolk is thus somewhat initially separated from tissue of the developing embryo. We observed three categories of yolk granules: 1)

BA & Phillips: Fire Ant Yolk Sphere Degradation 465

Figure 1. Transmission electron micrographs of cross sections of red imported fire ant eggs showing sequential degradation of yolk spheres: us = undegraded yolk spheres; ps = partially degraded yolk spheres; ds = degraded yolk spheres; ex = exo- chorion; en = endochorion; vtm = vitellin membrane (Fig. 1A, bar = 36 µm; Fig. 1B, bar = 50 µm; Fig 1C, bar = 52 µm; Fig. 1D, bar = 67 µm). large, membrane bound, undegraded dense yolk spheres; and 2) partially degraded and 3) degraded yolk spheres (Fig. 1C). Not all yolk platelets in red imported fire ant eggs are degraded at the same time, but rather, yolk sphere utilization proceeds sequentially. Our cytological observations of yolk degradation (Fig. 1C) in red imported fire ant eggs support earlier studies showing sequential yolk degradation in other arthropods. For example, proteinase activity is involved in yolk granule organelle degradation during embryogenesis in Drosophila, and this activity is developmentally regulated (Medina et al. 1988). Fagotto (1991) states that yolk spheres of the eggs of an African soft tick (Ornithodoros moubata) are dense and neutral and, that during later stages, these spheres are degraded through acidification. Furthermore, acidic protease activity is responsible for yolk degradation in the migratory locust, Locusta migratoria (McGregor & Loughton 1974). Although Glancey & Banks (1988) observed less yolk during reabsorption of developing oocytes in red imported fire ant queens after treat-

466 Florida Entomologist 78(3) September, 1995 ment with an insect growth regulator, proteinase is inactive during those stages of egg formation. Therefore, normal yolk degradation would not be observable. Proteinase is only activated in the egg subsequent to fertilization (Medina et al. 1988). Although common mechanisms of yolk degradation may exist among evolution- arily distant species (Fagotto 1991), the assumption should not be made that these degradation mechanisms are applicable to other systems (Yamashita & Indrasith 1988). However, the general mechanism for yolk degradation in many insects has been shown to be sequentially controlled (Yamamoto & Takahashi 1993), and proteins within these yolk spheres are differentially hydrolyzed.

ACKNOWLEDGMENT

This study was supported by the Texas State Line Item for Fire Ant Research. We thank H. Thorvilson, E. White, and C. Landry for reviewing our manuscript.

REFERENCES CITED

BOWNES, M., AND B. D. HAMES. 1977. Accumulation and degradation of three major yolk proteins in Drosophila melanogaster. J. Exp. Zool. 200: 149-156. FAGOTTO, F. 1991. Yolk degradation in tick eggs: III. Developmentally regulated acid- iﬁcation of the yolk spheres. Develop. Growth and Differ. 33: 57-66. GLANCEY, B. M., AND W. A. BANKS. 1988. Effect of the insect growth regulator fenoxycarb on the ovaries of queens of the red imported ﬁre ant (Hymenoptera: For- micidae). Ann. Entomol. Soc. America 81: 642-648. KUNKEL, J. G., AND J. H. NORDIN. 1985. Yolk proteins, pp. 83-111 in G. A. Kerkut and L. I. Gilbert [eds.] Comprehensive Insect Physiology, Biochemistry and Phar- macology, Vol. 1. Pergamon Press, Oxford. MCGREGOR, D. A., AND B. G. LOUGHTON. 1974. Yolk protein degradation during embryogenesis of the African migratory locust. Canadian J. Zool. 52: 907-917. MEDINA, M., P. LEON, AND C. G. VALLEJO. 1988. Drosophila cathepsin B-like proteinase: A suggested role in yolk degradation. Arch. Biochem Biophysics 263: 355- 363. POSTLETHWAIT, J. H., AND F. GIORGI. 1985. Vitellogenesis in insects, pp. 85-126 in W. L. Browder [ed.]. Developmental Biology: A Comprehensive Synthesis, Vol. 1: Oogenesis. Plenum Press, New York. SPURR, A. R. 1969. A low-viscosity epoxy resin medium for electron microscopy. J. Ul- trastructure Res. 26: 31-43. TATA, J. R., AND D. F. SMITH. 1979. Vitellogenesis: A versatile model for hormonal regulation of gene expression. Rec. Prog. Horm. Res. 35: 49-95. VALLEJO, C. G., R. PERONA, R. GARESSE, AND R. MAROCO. 1981. The stability of the yolk granules of Artenia. An improved method for their isolation and study. Cell Diff. 10: 345-356. WALLACE, R. W. 1985. Vitellogenesis and oocyte growth in nonmammalian vertebrates, pp. 127-177 in W. L. Browder [ed.]. Developmental Biology: A comprehensive synthesis, Vol. 1: Oogenesis. Plenum Press, New York. YAMASHITA, O., AND L. S. INDRASITH. 1988. Metabolic fates of yolk proteins during embryogenesis in arthropods. Develop. Growth and Differ. 30: 337-346. YAMAMOTO, Y., AND S. Y. TAKAHASHI. 1993. Cysteine proteinase from Bombyx eggs: role in programmed degradation of yolk proteins during embryogenesis. Comp. Biochem. Physiol. 106B: 35-45.

Moraes et al.: Salivary Gland Cultures 467

ORGAN CULTURE OF SALIVARY GLANDS OF MALE ANASTREPHA SUSPENSA (DIPTERA:TEPHRITIDAE)

REJANE R. DE MORAES1, JAMES L. NATION AND JAMES E. MARUNIAK Department of Entomology and Nematology Hull Road, Building 970 University of Florida Gainesville, FL 32611

1Fellow of National Research Council (CNPq). Brasilia, Brazil

ABSTRACT

Salivary glands from male Caribbean fruit flies, Anastrepha suspensa, were isolated individually by dissection and cultured in Schneider’s medium containing 10% Fetal Bovine Serum and antibiotics. A total of 103 salivary glands were successfully cultured. Two different procedures were used to evaluate viability. A trypan blue dye test revealed that no more than 10% of the cultured cells were non-viable, since they acquired blue coloration. In the second set of experiments, acid phosphatase activity of the cultured glands was measured colorimetrically from absorbance at 415 nm of liberated p-nitrophenol. This method indicated that the cultured glands stayed met- abolically active, with enzyme activity equal to that of freshly excised glands on the fifth, eighth, and fifteenth days of the experiment. Scanning electron microscopy (SEM) showed no visual differences in tissue organization and size of individual cells when freshly excised salivary glands and cultured glands were compared. SEM showed that the cells of cultured glands were neither swollen nor shrunken, and close-up views showed no evidence of cell deterioration or lysing.

Key Words: Salivary glands, organ culture, Caribbean fruit ﬂy, Tephritidae, Anas- trepha suspensa, acid phosphatase.

RESUMEN

Las glándulas salivares de los machos de la mosca caribeña de las frutas, Anas- trepha suspensa, fueron aisladas individualmente por disección y cultivadas en medio de Schneider que conteniene 10% de suero fetal bovino y antibióticos. Un total de 103 glándulas salivares fueron cultivadas con éxito. Dos procesos diferentes fueron usados para evaluar la viabilidad. Una prueba de tinción de trypan azul reveló que menos del 10% de las células cultivadas no era viable porque adquirían una coloración azul. En una segunda serie de experimentos, la actividad de fosfatasa ácida de las glándu- las cultivadas fue estudiada colorimétricamente mediante la absorbancia de p-nitro- fenol liberado a 415 nm. Este método indicó que las glándulas cultivadas permanecían metabólicamente activas, con actividad enzimática en los días quinto, octavo y déci- moquinto del experimento igual a la de las glándulas recién extirpadas. La Microsco- pía Electrónica de Barrido (MEB) no mostró diferencias visuales en la organización de los tejidos y en el tamaño de las céllulas individuales, cuando fueron comparadas las glándulas frescas y las cultivadas. La MEB mostró que las células de las glándulas cultivadas no estaban hinchadas ni contraídas, y las vistas tomadas de cerca no evi- denciaron deterioración o rompimiento celular.

Organ culture is a technique in which whole organs or representative parts are maintained as tissues and retain their intrinsic distribution, numerical and spatial

468 Florida Entomologist 78(3) September, 1995 orientation of explanted cells (Freshney 1987). Most organ cultures do not grow or, if they do, proliferation is limited to the outer cell layers. Thus, organ culture is essentially a technique for studying the behavior of integrated tissues rather than isolated cells. The in vitro culture of insect organs has made much progress in the last 30 years. This success is due mainly to the development of adequate tissue culture media by scientists, e.g., Schneider (1964). Insect organ culture has been associated primarily with insect endocrinological studies (Marks 1976). In vitro cultures of imaginal discs of Drosophila melanogaster Meigen (Raikow & Fristrom 1971), Sarcophaga peregrina (Robineau-Desvoidy) (Ohmori & Ohtaki 1972), Plodia interpunctella (Hubner) (Ober- lander 1976) and others have aided in the understanding of the role of 20-hydroxyecdysone and ecdysone in development and metabolic activity. Organ cultures of chironomid salivary glands have been extensively utilized in investigations of polytenic cell differentiation, cytogenetics, biochemistry and physiology (Firling & Kobilka 1978; Firling & Hou 1980). The salivary glands in Anastrepha suspensa (Loew) are sexually dimorphic, with males having much enlarged glands that terminate on each side of the body in a large ball of convoluted tubules in the pleural region of abdominal segments 3, 4 and 5 (Na- tion 1974, 1981). The glands in males reach maximum size at approximately the same time that males are maximally active in sexual behavior and in producing a pheromone that attracts females (Nation 1972). Pheromonal components have been detected in dissected salivary glands by gas chromatographic analysis (Nation 1989), but conclusive evidence that the glands participate in pheromone production is lack- ing. Organ cultures of the salivary glands might be used to determine whether or not they are involved in pheromone production, if a culture technique for these glands was available. For example, Srinivasan et al. (1979) successfully cultured sex pheromone glands of P. interpunctella for 10 days and were able to recover sex pheromone from the culture medium. The objectives of this study were to establish a protocol for sterile dissection of salivary glands from male Caribbean fruit ﬂies; to maintain salivary glands in artiﬁcial medium for at least seven days; and to establish protocols for monitoring the viability of cultured glands.

MATERIALS AND METHODS

Male A. suspensa were obtained from the mass rearing facility at the Florida State Department of Agriculture and Consumer Services, Gainesville, FL. Organ cultures of the male salivary glands were established in four separate experiments after dissecting the glands from 7- to 15-day-old males. All equipment and tools used to dissect the glands were sterilized. Microscopes, lamps and wax dishes were sterilized with 70% ethanol and left under a UV lamp for 20-30 min before dissections began. Scis- sors and forceps were autoclaved before each series of dissections and flamed after each fly was dissected. Glands were dissected under sterile conditions in a sterile hood. Male flies were surface-sterilized by immersion in 70% ethanol for 10 min and rinsed in sterile saline solution for 10 min. Just prior to each dissection, the head and legs of each fly were removed to diminish movement and to facilitate salivary gland removal. Each fly was pinned ventral side up in a wax dish containing sterile saline solution. A longitudinal cut was made through the cuticle and the lateral sides of the abdomen were gently pressed to release the salivary glands, which were then carefully excised. The excised glands were placed in sterile saline solution until a series of dissections were complete, then they were transferred to the culture medium.

Moraes et al.: Salivary Gland Cultures 469

Organ Culture Schneider’s medium (Schneider, 1964) containing 10% Fetal Bovine Serum and the antibiotics penicillin (100 units per ml), streptomycin (100 µg per ml), amphoteri- cin B (0.2 µg per ml), gentamycin (50 µg per ml), geocillin (100 µg per ml), kanamycin (100 µg per ml) and tetracycline (10 µg per ml) were used to culture the salivary glands. [Schneider’s medium has been used successfully to culture Drosophila cells (Schneider & Blumenthal, 1978).] A 24-cluster-well tissue culture plate (Corning) was used to culture the glands with 0.5 ml medium and two salivary glands per well. The salivary gland cultures were kept in an incubator at 27°C. Four experiments were conducted for eight to 15 days; the medium was replaced every three days in experiments 2 to 4.

Viability Tests

In the first experiment, penetration of trypan blue dye was used as an indicator of viability of the cultured glands, based on the principle that dead cells allow the dye to penetrate while live cells exclude the dye. Each day during one week of culture, 250 µl of trypan blue was added to the culture medium in one of the wells. Penetration or exclusion of the dye in the 2 glands was determined by microscopic examination. In experiments 2, 3 and 4, viability was monitored by determining the activity of acid phosphatase on the fifth, eighth and (only in experiment 4) on the fifteenth day. On each of these days, about 10 glands were removed from culture, rinsed with saline solution, and homogenized in 250 µl of 0.1 M citrate buffer, pH 5.5. Seventy µl of the glandular homogenate was added to each of two tubes containing 1 ml of 7 mM p-nitrophenol phosphate in 0.1 M citrate buffer, pH 5.5. A third tube served as a control and contained only citrate buffer. Fresh glands were dissected from flies that were the same age as the cultured glands, and these fresh glands were homogenized as described above for the cultured glands. Assay tubes identical to those for cultured glands were prepared to receive 70 µl of the fresh gland homogenate positive controls. All tubes were incubated at 30°C for 10 min. The reaction was stopped by adding 2 ml of 10% Na2CO3 to each tube. This altered the pH, stopped acid phosphatase activity and caused a shift in the electron structure of the released p-nitrophenol so that it gave a yellow color and absorbed light strongly at 415 nm. The amount of p-nitrophenol released was determined at 415 nm in a Spectronic 20D colorimeter. A standard curve was prepared from pure p-nitrophenol, and final results were expressed as pg p-nitrophenol released per gland per h. The above protocol was changed slightly for the fourth experiment as follows. The substrate was dissolved in 0.1 M citrate buffer (pH 5.5) just before use. Incubation tubes contained 2 ml of 7 mM p-nitrophenol phosphate and 0.1 ml of glandular homogenate. Tubes were incubated for 20 minutes at 30°C. The reaction was stopped by adding 3 ml of 10% Na2CO3 solution. The control contained 7mM p-nitrophenol in citrate buffer.

Scanning Electron Microscopy

Freshly dissected and cultured salivary glands were fixed for about 24 h in Bouin’s fixative, washed in 70% ethanol to remove excess picric acid, and dehydrated through an alcohol series to 100% ethanol. Glands were transferred from 100% ethanol to hex- amethyldisilasane (HMDS) (Nation 1983), transferred once to fresh HMDS and air dried. The dried tissues were mounted on stubs, gold coated with a sputter-coater, and observed in a Hitachi S 570 instrument with filament voltage at 15 or 20 KV.

470 Florida Entomologist 78(3) September, 1995

RESULTS AND DISCUSSION

A total of 103 excised salivary glands from adult male flies were successfully cultured in isolated organ cultures for up to 1 week, and, in one experiment, up to 15 days. Scanning electron microscopy (SEM) showed no visual differences in tissue organization and size of individual cells between freshly excised salivary glands and glands cultured for 8 days (Fig. 1). Cells in the cultured glands did not shrink, collapse or lyse. Each of the bilateral salivary glands has the appearance of a cluster of grapes, and each “grape” is a cell enclosing a cavity, with cells interconnected by a central ca- nal (Nation 1974). The SEM was useful in viewing the surface appearance of the gland cells and in ascertaining that the individual cells were not swelling, shrinking, or lysing. Cultured glands excluded trypan blue dye, an indication that the cells of the glands were alive and apparently functioning normally. Some blue cells were observed on the periphery of the glands, but their percentage was never greater than 10%. In a preliminary experiment to measure acid phosphatase activity, we found that 57 pg of p-nitrophenol per gland per h was released by glands cultured for 5 days post-dissection, and 61 pg was released by fresh glands of comparable age to the cultures. After 8 days of culture, 56 pg per gland per h of p-nitrophenol was released by cultured glands and 55 pg per gland per h by fresh glands of comparable age. The control was changed to contain 7mM p-nitrophenol in citrate buffer following this preliminary experiment and subsequent results are shown in Table 1. Those data show that acid phosphatase activity of freshly excised glands and cultured glands did not differ in glands that were 5, 8 or 15 days old. The substrate, p-nitrophenol phosphate, added to the standard caused a slight yellow color, lowering the values in Table 1. However, these values more accurately represent the true enzyme activity than the preliminary data above. The surface sterilization of adult flies, and the antibiotics in the culture medium were effective in maintaining sterility, since only one preparation showed evidence of contamination by becoming cloudy and discolored. Contamination also could be reduced by submerging adult flies in 70% ethanol. For example, Ohmori & Ohtaki (1972) surface-sterilized larvae of S. peregrina by submerging them in 70% ethanol for several min prior to isolating their wing discs for organ cultures, and Oberlander & Leach (1979) surfaced sterilized larvae of P. interpunctella (Hubner) by submerging them in 70% ethanol for 20 min prior to dissecting imaginal discs for organ cultures. Schneider’s medium, which was developed for Drosophila imaginal disc cultures (Schneider 1972), was adequate for culturing the glands of male Caribbean fruit flies. The exclusion of trypan blue and the activity of the enzyme, acid phosphatase, appear to be useful indicators of viability in salivary gland cultures. Enzyme activity and protein synthesis have often been used to measure the viability of cultured organs and cells. Firling & Kobilka (1978) measured the ability of cultured Chironomus tentans (Fabricius) salivary glands to incorporate 14C-leucine into trichloroacetic acid (TCA) precipitable proteins. Acid phosphatase as well as amylase and protease activity were monitored in cultured salivary glands of Calliphora larvae (Price 1974). Or- gan cultures of C. tentans salivary glands were capable of excluding a solution of 0.12% trypan blue, while injured cells did not exclude the dye (Firling & Kobilka 1978). The few blue cells observed on the surface of the male salivary glands in the present experiments may well have been injured in the dissection or handling procedures, thus causing them to leak and allow uptake of trypan blue dye. Monitoring for trypan blue dye uptake is easier than measuring acid phosphatase activity and might be the method of choice in any future experiments with salivary glands from fruit flies.

Moraes et al.: Salivary Gland Cultures 471

Figure 1. Scanning electron microscopy comparing cultured (8-days-old) and freshly excised male salivary glands (from 8-day-old males) from Caribbean fruit flies (Anastrepha suspensa). A and B, 150 × magnification; C and D, 500 × magnification.

ACKNOWLEDGMENTS

We thank Kathy Milne for technical assistance in the laboratory, Drs. Herbert Oberlander and S. M. Ferkovich for helpful criticism of earlier drafts, and the Na- tional Research Council-CNPq, Brazil, for ﬁnancial support to R. R. de Moraes. Flor- ida Agricultural Experiment Station Journal Series No. R-04079.

472 Florida Entomologist 78(3) September, 1995

TABLE 1. COMPARISON OF PHOSPHATASE ACTIVITY BETWEEN CULTURED AND FRESHLY EXCISED SALIVARY GLANDS.

P-Nitrophenol Salivary Glands pg per Gland per Hour

Mean1 ± SD

5 Days 8 Days 15 Days

Cultured 7.36 ± 0.21 7.26 ± 0.23 6.69 ± 0.02 Fresh 8.38 ± 0.11 8.49 ± 1.57 7.32 ± 0.32

1All values are the mean of two replicates except at 8 days, when four replicates were analyzed.

REFERENCES CITED

FIRLING, C. E., AND B. KOBILKA. 1979. A medium for the maintenance of Chironomus tentans salivary glands in vitro. J. Insect Physiol. 25: 93-103. FIRLING, C. E., AND A. HOU. 1980. Ecdysterone induced regression of Chironomus salivary glands maintained in vitro, p. 187-193 in E. Kurstak, K. Maramorosch, and A. Dubendorfer [eds.], Invertebrate systems in vitro. Elsevier/North Hol- land Biomedical Press, New York, NY. 598p. FRESHNEY, R. I. 1987. Culture of animal cells: A manual of basic technique. 2nd ed. Alan R. Liss, Inc, New York, NY. 397pp. MARKS, E. P. 1976. The uses of cell and organ cultures in insect endocrinology, in E. Kurstak, and K. Maramorosch, [eds.], Invertebrate tissue culture, applications in medicine, biology, and agriculture. Academic Press, New York. 398pp. NATION, J. L. 1972. Courtship behavior and evidence for a sex attractant in the male Caribbean fruit fly, Anastrepha suspensa. Ann. Entomol. Soc. America. 65: 1364-1367. NATION, J. L. 1974. The structure and development of two sex specific glands in male Caribbean fruit flies. Ann. Entomol. Soc. America. 67: 731-734. NATION, J. L. 1981. Sex-specific glands in tephritid fruit flies of the genera Anas- trepha, Ceratitis, Dacus, and Rhagoletis (Diptera: Tephritidae). Int. J. Insect Morphol. & Embryol. 10: 121-129. NATION, J. L. 1983. A new method using hexamethyldisilazane for preparation of soft insect tissues for scanning electron microscopy. Stain Technology. 58:347-351. NATION, J. L. 1989. The role of pheromones in the mating system of Anastrepha fruit flies, pp. 189-205 in A. S. Robinson and G. Hooper, [eds], Fruit Flies Their Bi- ology, Natural Enemies and Control, Vol. 3A, Elsevier Press. OBERLANDER, H. 1976. Hormonal control of growth and differentiation of insect tissues cultured in vitro. In Vitro. 12: 225-235. OBERLANDER, H., AND C. E. LEACH. 1979. The action of β-ecdysone and juvenile hormone in organ cultures of lepidopteran imaginal disks. TCA Manual. 5: 993- 995. OHMORI, K., AND T. OHTAKI. 1973. Effects of ecdysone analogues on development and metabolic activity of wing disks of the fleshfly, Sarcophaga peregrina, in vitro. J. Insect Physiol. 19: 1199-1210. PRICE, G. M. 1974. Protein Metabolism by the salivary glands and other organs of the larva of the blow fly, Calliphora erythrocephala. J. Insect Physiol. 20: 329-347. RAIKOW, R., AND J. W. FRISTROM. 1971. Effects of β-ecdysone on RNA metabolism of imaginal disks of Drosophila melanogaster. J. Insect Physiol. 17: 1599-1614. SCHNEIDER, I. 1964. Differentiation of larval Drosophila eye-antennal discs in vitro. J. Exp. Zool. 156: 91-104. Moraes et al.: Salivary Gland Cultures 473

SCHNEIDER, I. 1972. Cell lines derived form late embryonic stages of Drosophila melanogaster. J. Embryol. Exp. Morphol. 27: 353-365. SCHNEIDER, I., AND A. B. BLUMENTHAL. 1978. Drosophila cell and tissue culture, pp. 206-315 in M. Ashburner, and T. R. F. Wright, [eds], Genetics and Biology of Drosophila, Vol 2A, Academic Press, New York. SRINIVASAN, A., J. A. COFELT, AND H. OBERLANDER. 1979. In vitro maintenance of the sex pheromone gland of the female indian meal moth Plodia interpunctella (Hubner). J. Chem. Ecol. 5: 653-661.

♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦

Shelly & Villalobos: Melon Fly Behavior 473

CUE LURE AND THE MATING BEHAVIOR OF MALE MELON FLIES (DIPTERA: TEPHRITIDAE)

TODD E. SHELLY1 AND ETHEL M. VILLALOBOS2 1Hawaiian Evolutionary Biology Program, University of Hawaii, Honolulu, HI 96822

2Biology Department, Chaminade University, Honolulu, HI 96816

ABSTRACT

Laboratory tests were conducted to assess the effect of the parapheromone cue lure on the mating behavior of male Bactrocera cucurbitae (Coquillett). Exposure to cue lure resulted in a short-term mating advantage. For wild flies, treated males that fed on cue lure on the day of testing, or 1 day prior to testing, mated more frequently than control males that had no prior exposure to cue lure. However, control and treated males had similar mating success in tests performed 3 or 7 days after the treated males were exposed to the lure. Exposure to cue lure also increased the mating success of mass-reared, irradiated males relative to unexposed wild males, though this advantage was evident for only 1 day following exposure. Cue lure appeared to enhance mating performance by increasing male wing-fanning activity but not the attractiveness of the signal per se. A field study revealed that irradiated males exposed to cue lure 1 week prior to release were less likely to be captured (in Steiner traps baited with cue lure and naled) than unexposed males. These findings suggest that exposure of sterile males to cue lure might improve the effectiveness of sterile insect release as well as enable simultaneous control programs of sterile insect release and male annihilation.

Key Words: Bactrocera cucurbitae, parapheromone, sterile insect release, Hawaii.

RESUMEN

Se estudió en el laboratorio el efecto de la paraferomona “cue lure” en el comportamiento sexual de los machos de Bactrocera cucurbitae (Coquillett). La exposición a la “cue lure” tuvo como resultado ventajas en los apareamientos a corto plazo. Los machos salvajes que fueron alimentados con “cue lure” el mismo día, o el día anterior al experimento se aparearon más frecuentemente que los machos del grupo control que no habían sido expuestos a la paraferomona. Sin embargo, los machos de ambos gru- pos tuvieron éxitos de copulación similares en experimentos conducidos 3 o 7 días des- pués de haber sido expuestos a la “cue lure”. La exposición a la “cue lure” de los

474 Florida Entomologist 78(3) September, 1995 machos criados en el laboratorio e irradiados también aumentó el éxito de copulación en comparación con los machos salvajes que no habían recibido la paraferomona, pero esta ventaja sexual fue efectiva sólo un día después del tratamiento. Al paracer la “cue lure” aumenta la efectividad del comportamiento sexual porque estimula la actividad vibratoria de las alas, pero no afecta la calidad atractiva de la señal química en sí. Un estudio de campo indicó que los machos irradiados expuestos a la “cue lure” la semana anerior a ser liberados fueron capturados con menor frecuencia (en trampas de Stei- ner cebadas con “cue lure” y naled) que los machos que no fueron tratados. Los resultados sugieren que el exponer machos estériles a la “cue lure” podría aumentar la efectividad de las liberaciones de insectos estériles, asícomo permitir el uso simultá- neo de programas de control basados en la liberación de insectos estériles y aniquila- ción de machos.

The males of many tephritid species are strongly attracted to specific chemical compounds, termed male lures or parapheromones, which either occur naturally in plants or are (presumed) synthetic analogues of plant-borne substances (Chambers 1977; Sivinski & Calkins 1986; Fletcher 1987). Several well-known examples include the attraction of male Mediterranean fruit flies, Ceratitis capitata (Wiedemann), to trimedlure, male oriental fruit flies, Bactrocera dorsalis (Hendel), to methyl eugenol, and male melon flies B. curcubitae (Coquillett) to cue lure. Because they are powerful attractants, parapheromones are used in current control programs of tephritid pests for detection and monitoring of populations and for eradication via male annihilation (Chambers 1977; Economopoulos & Haniotakis 1994). Recent studies (Shelly & Dewire 1994; Shelly 1994) on B. dorsalis suggest another potential use of male lures in control efforts. Data collected in the laboratory showed that (treated) males exposed to methyl eugenol mated more frequently than (control) males that had no prior exposure to the lure. Moreover, the effect of methyl eugenol was long-lasting, and males that fed for only 30 s on methyl eugenol had a mating advantage as long as 35 days later. Enhanced mating success appeared to result from 2 factors: treated males signaled (wing-fanned) more frequently and for a given level of signaling attracted more females per min than control males. In addition, field tests showed that treated males were less likely to be captured (in traps baited with methyl eugenol and naled) than control males for as long as 35 days after exposure. These studies suggest that the effectiveness of control efforts might be enhanced by exposing sterile males to parapheromone prior to release, making it possible to combine programs of male annihilation and sterile insect release. The present study further examines the relationship between parapheromones and sexual behavior by investigating the influence of cue lure on the mating behavior of male B. cucurbitae under laboratory conditions. Using laboratory-reared wild flies, we initially tested whether exposure to cue lure enhances mating success and, if so, for how long after exposure. We then examined whether exposure to cue lure increased the mating competitiveness of irradiated (sterile) males relative to wild males and, if so, for how long after exposure. As will be described, the mating trials did, in fact, show that cue lure enhanced mating performance, and data were then gathered to assess the effect of cue lure on male signaling effort and signal attractiveness. Fi- nally, to investigate the possibility of conducting sterile insect release and male annihilation simultaneously, we tested whether exposure to cue lure reduced the probability of capturing (in traps baited with cue lure and naled) irradiated males following their release in the environment.

Shelly & Villalobos: Melon Fly Behavior 475

MATERIALS AND METHODS

Mating Behavior of B. cucurbitae Several workers have described the mating behavior of B. cucurbitae (Suzuki & Koyama 1980, 1981; Kuba & Koyama 1982, 1985; Kuba et al. 1984; Iwahashi & Ma- jima 1986; Kuba & Sokei 1988), and the following summary derives from these earlier reports. Mating occurs at dusk in male aggregations termed leks. Males perch singly on leaf under surfaces of both host and non-host plants and defend their site against intruding males. While perching, males fan their wings rapidly, producing a high- pitched buzzing sound. Wing-fanning also enhances dispersal of a pheromone, produced in the rectal gland, attractive to females (Schultz & Bousch 1971; Kobayashi et al. 1978; Baker et al. 1982). (Because acoustic, visual, and olfactory cues may be involved, the terms “signaling” and “calling”, which are used synonymously with wing- fanning, refer to the composite stimulus produced by all communicative modes.) After several seconds of wing-fanning, the male pounces on the female, and copulation en- sues. At present, the role of cue lure in male mating behavior is obscure, particularly because cue lure is a synthetic compound (unlike methyl eugenol; Metcalf et al. 1975) that has not yet been recorded from any natural source (Beroza et al. 1960). However, studies by Nishida and co-workers (Nishida et al. 1990, 1993) suggest that cue lure re- sembles a natural plant-borne substance, whose metabolites are sequestered in the rectal gland for pheromone synthesis. Initial work showed that males fed cue lure accumulated in the rectal gland a particular ketone [4-(4-hydroxyphenyl)-2-butanone; also known as Willison’s lure and raspberry ketone] known to occur in various plants and to be attractive to male B. cucurbitae. More recently, male melon ﬂies were observed to feed voraciously on ﬂowers of an orchid that contained raspberry ketone and to subsequently sequester it in the rectal gland. Though the exact role of this compound in female attraction remains unknown, its accumulation at the site of pheromone production clearly suggests a role in mate attraction and courtship.

Mating Trials Wild flies were derived from a laboratory stock started in April 1994 with 200-300 adults reared from Coccinia grandis (L.) collected in Waimanalo, Oahu, Hawaii. The present study was conducted during July-September, 1994, and consequently the flies were 3-5 generations removed from the wild. (Though this stock may have undergone genetic change as a result of colonization, the flies will be referred to as “wild” as ter- minological shorthand). The colony was housed in a large screen cage (1.2 m by 0.6 m by 0.6 m) with superabundant food (a mixture of honey and protein hydrolysate) and water. Italian squash (Cucurbita pepo L.) was provided periodically for oviposition. Room temperature was maintained at 20-22°C and relative humidity at 65-75%. In- fested squash were placed in plastic buckets containing vermiculite, and larval development and pupation proceeded in situ. Adults were separated by sex within 5 days of eclosion, well before reaching sexual maturity [at 18-20 days of age (Wong et al. 1986)]. Adults were maintained in plastic buckets (5 liters volume; 40-60 individuals per bucket) covered with screen mesh and were supplied with ample food and water. Mass-reared, irradiated flies were obtained from a colony started by the USDA/ ARS Tropical Fruit and Vegetable Laboratory, Honolulu, in 1958 (H. Chang, pers. comm.) using standard rearing procedures (Tanaka et al. 1969; males from this stock will hereafter be referred to as “sterile males”). Pupae that had been exposed to 10 krad of gamma radiation from a 60Co source were obtained 1 day before eclosion. Sexes

476 Florida Entomologist 78(3) September, 1995 were separated within 3 days of eclosion [sexual maturity in this stock is attained at about 8-10 days of age (Vargas et al. 1984; Wong et al. 1986)]. In the mating tests, 3 males from each of the 2 groups being tested were placed with 3 wild females in screen cages at least 4 h before sunset. Six to 9 cages were used on a given day. The cages were 30-cm cubes with one side open and fitted with a cloth sleeve. Before placement in the cages, males from the 2 groups were cooled for several minutes and then marked on the thorax with different colors of enamel paint. Fe- males were not marked. The cooling and painting procedures had no apparent adverse effects, and males resumed normal activities within minutes of handling. Three experiments were conducted. In the first, mating success was compared between wild males that had previous exposure to cue lure (treated males) and wild males denied access to cue lure (control males). Treated males (21-25 days old) were given unrestricted access during a 2-h period to a cotton wick (5 cm long) to which 1.5 ml of pure cue lure had been applied (cue lure obtained from the USDA/ARS Fruit and Vegetable Laboratory, Honolulu). The wicks, held upright in small plastic containers, were placed singly in the plastic buckets during midday. Treated males were used 0, 1, 3 or 7 days after exposure to cue lure. In an additional test, treated males were exposed to the wick for 2 h, but the wick was covered with screen mesh so that direct contact with it was prohibited. In these tests, control and treated males were approximately the same age. Females used in this and all subsequent experiments were 21- 28 days old. In the second experiment, the mating frequency of wild vs. sterile males was compared without previously exposing either group to cue lure. In the final experiment, cue lure was provided to either wild or sterile males during a 2-h midday period 1 day or 3 days prior to testing. In the latter two experiments, males of both groups were used during their first 2 weeks of sexual maturity (21-30 days and 12-21 days of age for wild and sterile males, respectively). In the final experiment, wild and sterile males were exposed to cue lure at 21-23 days and 12-14 days of age, respectively.

Signal Production and Attractiveness To investigate the possibility that exposure to cue lure increased signal production and attractiveness, we placed treated and control males in “minicages” within a larger flight cage and simultaneously monitored signalling by, and female attraction to, males (following the basic protocol of Poramarcom & Boake [1991]). Because wild males called only infrequently in the mini-cages, males (and females) used in this experiment were from the USDA/ARS colony mentioned above. In this case, however, we used non-irradiated, “normal” flies. Sexes were separated within 3 d of eclosion. Because single males were reluctant to call, groups of three males were placed in the minicages. Four groups—two control and two treated—were tested on a given day. All treated males fed for 1-1.5 min on a wick containing 1.5 ml of cue lure one day prior to use, and in all tests control and treated males were between 12-15 days old. The minicages were transparent, plastic cylinders (12 cm by 7 cm) whose ends were covered with nylon screening. The tests were conducted in a large screen cage (1.2 m by 0.6 m by 0.6 m) that contained three potted plants (Ficus sp.). The minicages were hung at four selected branches on these plants, and the individual cages were placed randomly at these locations for each observation day. On a given day, 45 mature virgin females (12-15 days of age) were placed in the cage 4-6 h before sunset. At the same time, the minicages (containing males) were placed at their assigned locations on the plants. Room lights were then extinguished, and the flight cage, which was near a west-facing window, received only natural light.

Shelly & Villalobos: Melon Fly Behavior 477

Starting 45 min before sunset, we checked the individual minicages at 1-min intervals and recorded the numbers of wing-fanning males and females resting on the minicages. Attractiveness of male signaling was thus based on the number of female sightings for a given minicage; because females were not marked, no data are available on the number of different females that alighted on the minicages. At the end of an observation period, individuals of both sexes were discarded, and a new set of males and females was used each day of testing. Observations were made on 7 days, and thus data were gathered for 14 minicages of control and treated males, respectively.

Capture of Sterile Males Exposed to Cue Lure A trapping experiment was conducted on the campus of the University of Hawaii at Manoa, Honolulu, Hawaii. The study area was a large, grassy lawn containing numerous ornamental trees and shrubs. Ten Steiner traps were placed singly in trees in a circle (50-m radius) around a central release point. The same plants were used in all replicates. Traps were suspended in the canopy (2 m above ground) by a 30-cm wire fastened to a branch. Each trap contained a 5-cm long cotton wick to which 2.0 ml of cue lure (5% naled) had been applied. Treated sterile flies (12-15 days old) were given access to a wick containing 1.5 ml of cue lure for a 2-h period 1 day prior to release. Similarly-aged control males (18-23 day old) that had no exposure to cue lure were also released. Prior to release, control and treated males were marked on the thorax with different color combinations of enamel paint (a given combination was used only once). Flies were released between 1000-1100 hours by removing the screen top and gently tapping the bucket to induce flight. Traps were checked 48 h after release, and captured flies were examined individually for markings in the laboratory. Seven replicates were conducted with 100 males released per group (control or treatment) per replicate. During the study period, days were generally sunny or only partly cloudy, and ambient daylight temperatures ranged between 25-33°C.

RESULTS

Mating trials Cue lure conferred a short-term mating advantage (Fig. 1). In the first experiment, treated males tested on the same day they were exposed to cue lure achieved 62% (55/ 88) of all matings recorded (t=2.3; P < 0.05; binomial test). Similarly, treated males tested on the day following exposure obtained 67% (58/86) of the matings (t=3.4; P < 0.001; binomial test). However, no difference in mating success was observed between control and treated males when treated males were tested 3 days (t=0.8) or 7 days after exposure to cue lure (P > 0.05 in both cases; binomial test). Also, treated males that were permitted only to approach a cue lure source without contacting it did not have a mating advantage over control males. Treated males obtained only 51% (38/74) of all matings (t=0.2; P > 0.05; binomial test). Cue lure also influenced the relative mating competitiveness of wild and sterile males. In the second experiment (where cue lure was not given to either wild or sterile males), sterile males had a mating advantage over wild males and obtained 86% (75/ 87) of all matings (t=9.6; P < 0.001; binomial test). In the final experiment, sterile flies exposed to cue lure 1 day prior to testing obtained 98% (93/95) of all matings, a proportion significantly greater than that observed in the preceding experiment (G=8.0; P < 0.01; G test with Yates correction). Wild flies exposed to cue lure 1 day before test-

478 Florida Entomologist 78(3) September, 1995

Figure 1. Numbers of matings obtained by control vs. treated wild males in trials conducted 0, 1, 3, or 7 days after treated males were exposed to cue lure. ing also displayed increased mating success and obtained nearly 1/4 (22/95=23%) of the matings (about twice the proportion recorded when cue lure was not provided), though this increase was not statistically signiﬁcantly (G=2.0; P < 0.20; G test with Yates correction). Despite their improved performance, wild males still accounted for a disproportionately small number of matings relative to sterile males (t=6.4; P < 0.001; binomial test). Exposure to cue lure 3 days prior to testing had no effect on the relative mating success of wild and sterile males. Sterile males obtained 89% (10/88; t=11.4; P < 0.001; binomial test) and 88% (70/80; t=10.0; P < 0.001; binomial test) of all matings when wild and sterile males, respectively, were tested 3 days after exposure to cue lure. In both cases, the results were similar to those obtained when cue lure was not given to either group of males (P > 0.05 in both comparisons; G test with Yates correction).

Signal Production and Attractiveness Males exposed to cue lure called more frequently than control males (Fig. 2). An average of 70 instances of wing-fanning (maximum value possible=135=3 males per minicage over 45 checks) was recorded for minicages with treated males compared to only 54 instances for minicages with containing control males (U=149; P < 0.001; Mann-Whitney test). In addition, more females were sighted on minicages with treated males than control males (Fig. 2). On average, 56 female sightings were recorded for minicages with treated males but only 34 were observed for minicages with control males (U=161; P < 0.001; Mann-Whitney test). This difference in female arriv-

Shelly & Villalobos: Melon Fly Behavior 479

Figure 2. Relationship between female sightings and wing-fanning for control (●) and treated (*) males. Each point represents a minicage that contained three males. Abscissa represents the total number of wing-fanning instances recorded for all 3 males/minicage over the 1-min checks. Ordinate represents the total number of female sightings on a minicage over the 1-min checks. Regression equations: control males - Y=1.05X-25.2; r2=0.75; treated males - Y=0.81X-0.6; pooled (shown) - Y=1.07X- 22.6; r2=0.64. als appeared to reﬂect the differential calling activity between control and treated males and not a difference in signal attractiveness per se between the two groups, since the rate at which female sightings increased with male signaling was similar for control and treated males (t=0.7; P > 0.05); Fig. 2).

Capture of Sterile Males Exposed to Cue Lure

Sterile males that had been exposed to cue lure were trapped less frequently than sterile males that had no exposure to the lure. On average, only 6 treated males (range: 2-12) were captured per replicate compared to 19 control males (range: 14-28; U=49; P < 0.001; Mann-Whitney test).

DISCUSSION

The present study shows that, under laboratory conditions, exposure to cue lure confers a mating advantage to male B. cucurbitae. However, whereas a positive effect of methyl eugenol was evident one month after exposure (Shelly & Dewire 1994), cue lure provided a mating advantage for only one day following feeding. In the melon ﬂy,

480 Florida Entomologist 78(3) September, 1995 ingestion of cue lure appeared to increase mating success via an increased level of sexual signaling. Though metabolites of cue lure (or its natural analogue) may be used for pheromone synthesis (Nishida et al. 1990, 1993), we found no evidence that the signals produced by males exposed to cue lure were more attractive than those generated by control males: for a given level of wing-fanning, similar numbers of females were sighted on minicages with control and treated males. This situation differs from that reported for B. dorsalis, where (independent of signaling effort) females showed a striking preference for males that fed on methyl eugenol over control males (Shelly & Dewire 1994). A separate study (Whittier & Shelly 1995) indicates that trimedlure has an effect on male medflies similar to that described here for cue lure and male melon flies. Though they do not feed on the chemical, C. capitata males exposed to trimedlure displayed higher calling levels and mated more often than control males. However, based on the number of female arrivals to calling males, treated males did not produce more attractive signals than control males. Also, as with the melon fly, the positive effect of the lure was short-lived and was, in fact, evident only on the day that males were exposed to the lure. Cue lure appears to be far less attractive to B. cucurbitae males than is methyl eugenol to B. dorsalis males. When exposed to the lure for a 30-min period in screen cages, nearly 90% of wild B. dorsalis males fed on methyl eugenol at their first opportunity (Shelly 1994), but only 33% (22/64) of wild B. cucurbitae males fed on cue lure during initial exposure (T.E.S., unpublished data). Similarly, trapping experiments (using identical procedures at the same site) with untreated, irradiated males in the two species showed a pronounced difference in capture probability: an average of 32% of B. dorsalis males were captured per replicate (Shelly, unpublished data) compared to only 19% of B. cucurbitae males. The comparatively low attractiveness of cue lure is consistent with its short-lived effect on mating performance. Alternatively, and independent of its effects on male behavior, cue lure may be an “imperfect” mimic of a natural analogue and hence a less potent attractant than methyl eugenol, a natural plant product. As in B. dorsalis, B. cucurbitae males that fed on cue lure were unlikely to re-visit a cue lure source for additional feeding. In the field experiment, an average of only 6 previously exposed males were captured per replicate compared to 19 control males per replicate. Similarly, among the 22 males that fed on cue lure during an initial 30- min trial (see above), not a single individual fed on cue lure during a second trial conducted 7 days later. These results are consistent with Chambers et al. (1972), who reported reduced trap captures of B. cucurbitae males for 3 weeks following exposure to cue lure. Given that cue lure confers only a short-term mating advantage, it remains unclear why male melon flies show a reduced tendency to re-visit a cue lure source. Regarding potential application to control efforts, the present study suggests that pre-release exposure of sterile B. cucurbitae males to cue lure may both enhance their mating competitiveness in the wild and decrease their attraction to lure-baited traps. Thus, pre-release exposure to lure may represent a relatively simple and inexpensive means to combine programs of sterile insect release and male annihilation (see also Chambers et al. 1972). However, compared to the oriental fruit fly, the effectiveness of pre-release exposure may be limited for the melon fly given the short-lived mating advantage conferred by the lure. For both species, assessing the value of this technique will ultimately require field tests in which wild populations are compared between areas receiving, for example, (1) sterile insect release with or without pre- release exposure to lure or (2) male annihilation alone or in combination with sterile insect release with pre-release exposure to lure.

Shelly & Villalobos: Melon Fly Behavior 481

ACKNOWLEDGMENTS We thank S. Fong, K. Luke, and F. Meng for their help in rearing and marking the ﬂies. This research was funded by grants from the USDA/ARS (58-91H2-6-42) and USDA/CSRS (9401860).

REFERENCES CITED

BAKER, R., R. H. HERBERT, AND R. A. LOMER. 1982. Chemical components of the rectal gland secretions of male Dacus cucurbitae, the melon fly. Experientia 38: 232- 233. BEROZA, M., B. H. ALEXANDER, L. F. STEINER, W. C. MITCHELL, AND D. H. MIYASHITA. 1960. New synthetic lure for the male melon fly. Science 131: 1044-1045. CHAMBERS, D. L., K. OHINATA, M. FUJIMOTO, AND S. KASHIWAI. 1972. Treating tephritids with attractants to enhance their effectiveness in sterile-release programs. J. Econ. Entomol. 65: 279-282. CHAMBERS, D. L. 1977. Attractants for fruit fly survey and control, pp. 327-344 in H. H. Shorey and J. J. McKelvey [eds.], Chemical control of insect behavior. Wiley, New York. ECONOMOPOULOS, A. P., AND G. E. HANIOTAKIS. 1994. Advances in attractant and trapping technologies for tephritids, pp. 57-66 in C. O. Calkins, W. Klassen, and P. Liedo [eds.], Fruit flies and the sterile insect technique. CRC Press, Boca Ra- ton. FLETCHER, B. S. 1987. The biology of dacine fruit flies. Ann. Rev. Entomol. 32: 115- 144. IWAHASHI, O., AND T. MAJIMA. 1986. Lek formation and male-male competition in the melon fly, Dacus cucurbitae Coquillett (Diptera: Tephritidae). Appl. Entomol. Zool. 21: 70-75. KOBAYASHI, R. M., K. OHINATA, D. L. CHAMBERS, AND M. S. FUJIMOTO. 1978. Sex pheromones of the oriental fruit fly and the melon fly: mating behavior, bioassay method, and attraction of females by live males and by suspected pheromone glands of males. Environ. Entomol. 7: 107-112. KUBA, H., AND J. KOYAMA. 1982. Mating behavior of the melon fly, Dacus cucurbitae Coquillett (Diptera: Tephritidae): comparative studies of one wild and two laboratory strains. Appl. Entomol. Zool. 17: 559-568. KUBA, H., AND J. KOYAMA. 1985. Mating behavior of wild melon flies, Dacus cucurbitae Coquillett (Diptera: Tephritidae) in a field cage: courtship behavior. Appl. Entomol. Zool. 20: 365-372. KUBA, H., J. KOYAMA, AND R. J. PROKOPY. 1984. Mating behavior of wild melon flies, Dacus cucurbitae Coquillett (Diptera: Tephritidae) in a field cage: distribution and behavior of flies. Appl. Entomol. Zool. 19: 367-373. KUBA, H., AND Y. SOKEI. 1988. The production of pheromone clouds by spraying in the melon fly, Dacus cucurbitae Coquillett (Diptera: Tephritidae). J. Ethol. 6: 105- 110. METCALF, R. L., W. C. MITCHELL, T. R. FUKUTO, AND E. R. METCALF. 1975. Attraction of the oriental fruit fly, Dacus dorsalis, to methyl eugenol and related olfactory stimulants. Proc. Nat. Acad. Sci. 72: 2501-2505. NISHIDA, R., O. IWAHASHI, AND K. H. TAN. 1993. Accumulation of Dendrobium super- bum (Orchidaceae) fragrance in the rectal glands by males of the melon fly, Da- cus cucurbitae. J. Chem. Ecol. 19: 713-722. NISHIDA, R., K. H. TAN, S. TAKAHASHI, AND H. FUKAMI. 1990. Volatile components of male rectal glands of the melon fly, Dacus cucurbitae Coquillett (Diptera: Te- phritidae). Appl. Entomol. Zool. 25: 105-112. PORAMARCOM, R., AND C. R. B. BOAKE. 1991. Behavioural influences on male mating success in the Oriental fruit fly, Dacus dorsalis Hendel. Anim. Behav. 42: 453- 460. 482 Florida Entomologist 78(3) September, 1995

SCHULTZ, G. A., AND G. M. BOUSCH. 1971. Suspected sex pheromone glands in three economically important species of Dacus. J. Econ. Entomol. 64: 347-349. SHELLY, T. E. 1994. Consumption of methyl eugenol by male Bactrocera dorsalis (Diptera: Tephritidae): low incidence of repeat feeding. Florida Entomol. 77: 201-208. SHELLY, T. E., AND A. DEWIRE. 1994. Chemically mediated mating success in male Oriental fruit flies, Bactrocera dorsalis (Diptera: Tephritidae). Ann. Entomol. Soc. America 87: 375-382. SIVINSKI, J. M., AND C. CALKINS. 1986. Pheromones and parapheromones in the control of tephritids. Florida Entomol. 69: 157-168. SUZUKI, Y., AND J. KOYAMA. 1980. Temporal aspects of mating behavior of the melon fly, Dacus cuccurbitae Coquillett (Diptera: Tephritidae): a comparison between laboratory and wild strains. Appl. Entomol. Zool. 15: 215-224. SUZUKI, Y., AND J. KOYAMA. 1981. Courtship behavior of the melon fly, Dacus cucurbitae Coquillett (Diptera: Tephritidae). Appl. Entomol. Zool. 16: 164-166. TANAKA, N., L. F. STEINER, K. OHINATA, AND R. OKAMOTO. 1969. Low- cost larval rearing medium for mass production of Oriental and Mediterranean fruit flies. J. Econ. Entomol. 62: 967-968. VARGAS, R. I., D. MIYASHITA, AND T. NISHIDA. 1984. Life history and demographic parameters of three laboratory-reared tephritids (Diptera: Tephritidae). Ann. En- tomol. Soc. America 77: 651-656. WHITTIER, T. S., AND T. E. SHELLY. 1995. Trimedlure affects mating success and mate attraction in male Mediterranean fruit flies. Entomol. Exp. Appl. In review. WONG, T. T. Y., D. O. MCINNIS, AND J. I. NISHIMOTO. 1986. Melon fly (Diptera: Te- phritidae): sexual maturation rates and mating responses of laboratory-reared and wild flies. Ann. Entomol. Soc. America 79: 605-609.

♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦

482 Florida Entomologist 78(3) September, 1995

CHEMICAL FACTORS INVOLVED IN SELECTION OF HOST PLANT FOR OVIPOSITION BY THE PICKLEWORM MOTH (LEPIDOPTERA: PYRALIDAE)

JOSEPH K. PETERSON AND KENT D. ELSEY USDA, Agricultural Research Service Vegetable Laboratory 2875 Savannah Highway Charleston, SC 29414-5334

ABSTRACT

Studies with caged gravid females of the pickleworm, [Diaphania nitidalis (Stoll.)], revealed that leaves of yellow squash (Curcurbita pepo L.) contain small (MW <1000), non-volatile and highly polar amphoteric compounds which stimulate oviposition on artiﬁcial sites. Several compounds, extracted from paper chromatograms, caused moderate stimulation of oviposition. When these extracts were recombined, the mixture proved highly active. Addition of whole leaf volatiles to the active water- soluble fraction increased oviposition. The whole leaf volatiles mixture could be substituted with volatiles originating from leaf glandular trichomes.

Index Words. Lepidoptera, Diaphania nitidalis, oviposition, host plant, squash, Cur- curbita pepo, trichomes, volatiles.

Peterson & Elsey: Chemical Factors for Pickleworm Oviposition 483

RESUMEN Estudios sobre el gusano de los pepinos [Diaphania nitidalis (Stoll.)] revelaron que las hojas de la calabaza (Cucurbita pepo L.) contienen pequeños (MW<1000) compuestos anfotéricos no volátiles y altamente polares que estimulan la ovoposición en sitios artiﬁciales. Varios compuestos, extraidos a partir de cromatogramas en papel, estimu- laron moderadamente la ovoposición. Cuando esos extractos fueron recombinados, la mezcla fue muy activa. La adición de volátiles de la hoja completa a la fracción activa hidrosoluble incrementó la ovoposición. La mexcla de volátiles de la hoja completa po- dría ser sustituida por volátiles originados de los tricomas glandulares de las hojas.

The pickleworm, [Diaphania nitidalis (Stoll.)], is a severe pest of cucurbit vegetables in the southeastern United States, mainly because the larvae tunnel into the fruit. Commercially, damage tolerance is low, especially in the pickling industry (El- sey et al. 1989). Infestations in the field vary with cucurbit species as well as with cul- tivar (Corley 1973, Day et al. 1961, Pulliam 1979, Quisumbing & Lower 1975). Bioassays with cucumber cultivars showed that little antibiosis to larvae exists (Weh- ner et al. 1985). However, non-preference to oviposition was shown in glabrous mutants of muskmelon and cucumber (Day et al. 1961, Elsey & Wann 1982). Since nearly isogenic lines of glabrous and pubescent cucumbers were evaluated, it appeared that resistance (avoidance) was due to the lack of plant trichomes (Elsey & Wann, 1982). Elsey (1985) showed that resistance in Cucurbita moschata was due to oviposition non-preference. A study with whole plants in large field screen cages showed that pickleworm oviposition preference decreased in the order of yellow squash, cucumber and watermelon (Elsey 1981). A similar pattern was observed when ethanol extracts of whole leaves were applied to artificial oviposition sites (Elsey & McFadden 1981). Because chemical factors were suspected to be involved in oviposition preference by pickleworms, this study was undertaken to isolate these stimulants. Leaves of straightneck yellow squash were used because this yellow squash is a preferred host (Dilbeck & Canerday 1968) and most pickleworm eggs are laid on foliage. Effects of volatile and non-volatile compounds were evaluated and attempts to isolate non-volatile stimulants are described.

MATERIALS AND METHODS

Insect Rearing Diaphania nitidalis were raised in the laboratory according to methods described by Elsey et al. (1984).

Culture of Plant Material Seedlings of yellow squash (Cucurbita pepo L. var. Early Prolific Straightneck) were planted in the field and raised using standard cultural practices. Healthy leaves, approximately one half ultimate size (10 - 12 cm longest dimension) were collected from the field early in the morning and immediately processed or used for bioassays.

Bioassays Assays involving oviposition were conducted according to Elsey & McFadden (1981) with some modiﬁcation to improve quantitative aspects. Pads of building insu-

484 Florida Entomologist 78(3) September, 1995

TABLE 1. OVIPOSITION BIOASSAYS OF FILTRATES OBTAINED BY SEQUENTIAL FILTRA- TION THROUGH MEMBRANE MOLECULAR FILTERS.

Filter Average Oviposition1,2

(MWCO) (Percent of Total)

none 11.5 b 20 K 15.8 bc 5 K 22.1 de 1 K 26.9 e 500 19.9 cd blank3 3.9 a

1Average of four independent bioassays. 2Means followed by the same letter are not significantly different. Means separation by Fisher’s LSD (P = 0.05). 3No fraction applied. lation glasswool (diam 9 cm, thickness 2-3 cm) were placed in petri dishes containing the solutions to be tested. After adsorption was completed, the solvent was evaporated. Subsequently, the pads were suspended from the top of wire mesh cages (120 × 90 × 90 cm), which contained about 100 pairs of 3-day-old moths. The cages were located in a humidified room held at 23 C, and a photoperiod of 16:8 (L:D). Testing material was placed in the cages in the late afternoon (before the start of the scoto period), removed the next morning and the eggs counted. Common amino acids were tested for stimulation of oviposition by dissolving them in 50% ethanol and applying 50 µg or 200 µg per pad. The following amino acids were used: L-alanine, γ-amino butyric acid, L-arginine, L-asparagine, aspartic acid, citrulline, cysteine, L-glutamic acid, L-glutamine, glycine, histidine, DL-homocysteine, hy- droxyproline, isoleucine, leucine, lysine, DL-methionine, phenylalanine, proline, serine, DL-threonine, tryptophan, L-tyrosine and DL-valine. A test consisted of four pads containing one (different) amino acid each, one pad with 0.25 g equivalent of an aqueous polyamide fraction (control, contains no volatiles), and one pad without application (blank). Volatiles from small whole leaves (6-9 cm longest dimension) or droplets (trichome exudates) were tested by placing them between glass fiber pads. Alternatively, the fresh material was lyophilized for 72 hrs under high vacuum while the volatiles were trapped under liquid nitrogen. The leaves, now free of volatiles, were extracted and tested.

Extractions and Separations Leaves approximately two-thirds full size [4-7 g fresh weight (FW), 14-18 cm longest dimension of leaf blade] were harvested from greenhouse grown squash plants. The leaves were lyophilized and ground to pass a 0.85 mm mesh screen. The material was subsequently extracted with hexane for approximately 12 hours on a shaker and filtered. The procedure was repeated twice and the residue was extracted with hot 95% ethanol in a homogenizer and filtered. After two repetitions, the combined ethanol extracts were condensed to a syrup, and water was added. The extract was filtered through a series of surface membrane molecular filters under nitrogen pressure in a stirring cell. The filters had a nominal molecular weight cut-off (MWCO) of 20K, 5K,

Peterson & Elsey: Chemical Factors for Pickleworm Oviposition 485

Figure 1. Extraction and partitioning scheme for the isolation of oviposition stimulants. All fractions were assayed simultaneously, and the assay was performed three times. Numbers in parentheses indicate the amount of oviposition as a percent of the total number of eggs deposited. Numbers were corrected for ‘background’, i.e. eggs deposited on pads that received no extracts.

1K and 500 respectively. Since the 500 MWCO gave rise to losses of activity (Table 1), ﬁltrations of subsequent extracts were limited to a MWCO of 1K. Molecular ﬁltration was followed by partitioning against dichloromethane and ethylacetate respectively. The extraction scheme is presented in Fig. 1.

Chromatography

The aqueous fraction obtained after the last partitioning was subjected to low- pressure column chromatography using polyamide as stationary phase (column di- mensions: 25 × 2.6 cm diam). The polyamide was prewashed with one liter each of 0.5% formic acid and water, respectively. An extract aliquot representing up to 100 g leaf fresh weight equivalent (FWE) in 10 ml water was pumped onto the column and eluted stepwise with methanol/acetic acid (995:5), 100% methanol and water respectively. Absorbance was monitored at 280 nm. The fractions were tested for stimulation of oviposition and checked with p-anisidine phthalate to detect reducing sugars (Stahl 1969, p. 857) and with ninhydrin (Stahl 1969, p. 889) or vanillin-potassium hydroxide (Stahl 1969, p. 904) to test for primary or secondary amino functions.

486 Florida Entomologist 78(3) September, 1995

Figure 2. Dose-response curve of oviposition (% of total number eggs produced per bioassay) versus the logarithm of the amount of extract used. All concentrations were represented once in a bioassay and the entire assay was performed four times. Re- gression analysis: y=1.18 + 11.86x, r2=0.91, P<0.01.

The active fraction from the polyamide column (water fraction) was condensed and used for paper chromatography on Whatman 3 MM sheets (46 × 57 cm). Development was in the descending mode using tert-butanol: acetic acid:water-3:1:1. After drying, the sheets were sprayed on the sides with ninhydrin and heated at 110°C. The area corresponding with the colored bands, as well as the areas in between, were extracted with 50% ethanol and bioassayed. The active fraction from the polyamide column was also run through a Dowex 50 × 8 cation exchange column (18 cm × 1 cm diam; sample 10 g FWE). The sample was adjusted to pH 2.5 with concentrated formic acid and eluted with 8N formic acid, water, and triethylamine (in 20% aqueous acetone, pH=10). After the column had been standing for three days, more activity could be eluted with triethylamine [Table 3, Et3 N(2)]. The fractions were checked with ninhydrin and tested for activity. Hydrolysis was performed on the active polyamide fraction. The reaction was conducted in a sealed vial at 110°C for 24 hrs, in 2N HC1. The products were condensed to dryness, redissolved and bioassayed. Seven such tests were performed.

Sources of Chemicals

L-glutamine, L-asparagine, L-glutamic acid, L-arginine, proline, L-tyrosine and tryptophane were obtained from Aldrich Chemical Co., Milwaukee, WI. L-alanine, aspartic acid, γ-amino butyric acid, citrulline, cysteine, glycine, histidine, DL-homocysteine, isoleucine, leucine, lysine, DL-methionine, phenylalanine, hydroxy-L-proline,

Peterson & Elsey: Chemical Factors for Pickleworm Oviposition 487 serine, DL-threonine, DL-valine and ninhydrin were obtained from Sigma Chemical Co., St. Louis, MO. p-Anisidine phthalate was obtained from Pfaltz and Bauer, Inc. Waterbury, CT. Polyamide was obtained from Universal Scientiﬁc Inc. Atlanta, GA. Hexane, dichloromethane and ethylacetate (HPLC grade) were obtained from Bur- dick and Jackson, Muskegon, MI. Tertiary butanol was obtained from J. T. Baker Inc., Philipsburg, NJ.

Statistical Analyses Statistical analyses, consisting of one-way ANOVA, students t-test, and regression techniques, were performed where appropriate on a personal computer with the SOLO statistical system (BMDP Statistical Software, Inc., Los Angeles, CA). Means separations were by Fisher’s LSD (P=0.05).

RESULTS AND DISCUSSION Preliminary experiments showed that increasing amounts of crude 95% ethanol extracts of fresh squash leaves applied to glasswool pads increased egg-laying. At approximately 1 g FWE, i.e., the amount obtained from 1 g fresh leaf material, no further increase was obtained, and at higher dosages very erratic dose-response data were obtained. A linear dose-response (r2 = 0.91) was obtained when the percent of the total number of eggs laid was plotted against the logarithm of the extract concentration (Fig. 2). The partial isolation scheme (Fig. 1) shows that the oviposition stimulants remained in the aqueous fractions. Oviposition increased considerably when fractions of low polarity were removed (Fig. 1) or when molecular filtration was performed (Table 1). Although this may suggest the presence of inhibitors, the fractions with low polarity showed activity above ‘background’, i.e., when applied to glasswool pads, they received more eggs than non-treated pads. In addition to the molecular filtration down to a MWCO=1000, a filter with a MWCO=500 was used; in this case a significant amount of activity was lost. When the pH of the final aqueous fraction (Fig. 1) was adjusted to either 10 or 2.5, no activity could be extracted by ethyl acetate, suggesting that oviposition stimulants were highly polar molecules. The final active aqueous fraction obtained through partitioning (Fig. 1) was subjected to low pressure column chromatography, using polyamide as the stationary phase, and the fractions were tested for activity (Fig. 3). The first fraction (aqueous) contained virtually all of the activity. The first eluate from the polyamide column may contain salts, organic acids, sugars (Winter & Hermann 1986) and some water soluble

TABLE 2. EFFECT OF HYDROLYSIS OF THE AQUEOUS POLYAMIDE (ACTIVE) FRACTION1 ON OVIPOSITION BY THE PICKLEWORM MOTH.

Treatment Oviposition (% of Total) Mean SEM

Non-hydrolyzed 50.3 25.1 37.2 35.6 49.0 49.2 51.3 42.5 a2 3.8 Hydrolyzed 35.4 61.8 46.5 46.8 37.1 34.8 40.2 43.2 a 3.6 Blank (no application) 14.3 13.2 16.4 17.6 13.9 16.0 8.5 14.3 b 1.1 Total oviposition 796 1595 807 1585 2132 762 995

10.25 g FWE applied to oviposition sites. 2Means followed by the same letter are not signiﬁcantly different. Means separation by Fisher’s LSD (P = 0.05).

488 Florida Entomologist 78(3) September, 1995

TABLE 3. CATION EXCHANGE CHROMATOGRAPHY OF ACTIVE AQUEOUS FRACTION FROM POLYAMIDE CHROMATOGRAPHY.

Mean Ninhydrin Oviposition Eluent Vol. (ml) Reaction (% of Total)1 SEM

HCOOH (8N) 106 negative 28.52 b3 2.30

H2O 86 trace 13.54 a 1.68 2 Et3N (1) 84 posit. (purple) 25.09 b 2.42

Et3N (2) 92 posit. (purple) 26.21 b 1.78 Blank 6.63 a 0.43

1Six experiments were performed (n=6). A total of 700, 1345, 1320, 1198, 1610, and 1819 eggs were layed per experiment. 2 Et3N = triethylamine. 3Means followed by the same letter are not signiﬁcantly different (Fisher’s LSD, P = 0.05). phenolic glycosides (Thieme 1964). Activity was also found in the very ﬁrst part of the aqueous eluate, before any absorption at 280 nm was observed, excluding phenolic compounds. Para-anisidine phthalate gave no reaction, indicating that no reducing sugars were present (Stahl 1969). Vanillin-potassium hydroxide or ninhydrin gave positive reactions, indicating the presence of primary or secondary amino groups

Figure 3. Polyamide column chromatography of active aqueous extract (see Fig. 1), monitored at 280 nm. Activity is expressed as percent of the activity in the pre-column matrix (active solution applied to column; activity = 100%). Numbers were corrected for the blank (no application). Four independent assays were performed and data were averaged. Peterson & Elsey: Chemical Factors for Pickleworm Oviposition 489

TABLE 4. PAPER CHROMATOGRAPHY1 OF ACTIVE FRACTION (AQUEOUS) OF POLYAMIDE COLUMN CHROMATOGRAPHY.

Oviposition (% Control)3

Band hRf2 Mean SEM

1 0-11.1 10.7 de4 3.9 2 11.1-19.5 10.1 de 3.7 3 19.5-26.5 6.4 e 4.2 4 26.5-34.7 48.6 a 32.9 5 34.7-40.5 17.6 cde 2.3 6 40.5-47.9 33.9 b 11.0 7 47.9-50.8 25.3 bc 4.4 8 50.8-57.7 22.0 bcd 6.4 9 57.7-63.8 11.6 de 3.1 10 63.8-68.3 24.6 bc 12.5 11 68.3-76.2 14.9 cde 1.6 12 76.2-100 52.8 a 18.3 Recombined5 255 65.4 Blank6 15 2.6

1 Solvent system: tert-BuOH:HOAc:H2O-3:1:1. 4.0 g FWE of the aqueous fraction of the polyamide chromatography applied to paper sheets. Control: 0.25 g FWE of the polyamide fraction. 2All bands stained with ninhydrin, except number 12. hRf: relative traveling distance of compound, 0-compound did not move, 100-compound moved with solvent front. 3Data obtained from 4 chromatograms. Oviposition for the 12 bands and the recombined sample was expressed as follows: sample - blank (no applic.) × 100% control - blank 4Means followed by the same letter are not significantly different. Means separation by Fisher’s LSD (P = 0.05). 50.5 g FWE of each chromatographic band was recombined. 6No application of sample. (Stahl 1969). Hydrolysis of the active aqueous fraction from the polyamide column did not change the activity significantly (Table 2). These results indicate that no active small peptides were present, or, if present, did not play a significant role. The active fraction from the polyamide column was further separated on a cation exchange column. Activity was present in all fractions except the aqueous (Table 3), indicating acidic and basic compounds were involved in stimulation of oviposition. Numerous paper chromatograms (PC) were developed from the active polyamide fraction. The developed papers were sprayed on the side with ninhydrin, and the entire chromatograms stained (hues of blue, purple and pink), except the front area. Twelve bands were cut, eluted and tested. Also tested were all the extracted regions recombined. An example of the PC and bioassay data, presented in Table 4, show that various regions were active, and that the recombined sample was very active. Since PC and parallel bioassays showed that active regions, with the exception of one, could be stained with ninhydrin, primary or secondary amino functions, were involved. Fur- thermore, ion-exchange data showed that acidic and basic compounds were involved. These compounds could not be extracted from an aqueous solution at high or low pH, indicating amphoteric characteristics of the active compounds. These properties strongly indicate amino acids. For this purpose common amino acids were tested. The 490 Florida Entomologist 78(3) September, 1995

TABLE 5. INFLUENCE OF COMMON AUTHENTIC AMINO ACIDS1 ON OVIPOSITION BY THE PICKLEWORM MOTH.

Amino Acid Oviposition2

50 µg 200 µg

L-alanine 4.5 -2.5 γ-amino butyric acid 1.7 1.9 L-arginine -0.3 -3.2 L-asparagine 4.8 2.3 Aspartic acid 1.1 4.4 Citrulline 4.6 1.8 Cysteine 3.7 4.0 L-glutamic acid 3.3 2.4 L-glutamine 2.0 1.3 Glycine 4.5 3.4 Histidine 2.4 4.1 DL-homocysteine 5.8 2.3 Hydroxy-L-proline -0.4 1.7 Isoleucine 0.3 -1.4 Leucine 0.4 5.2 Lysine 5.2 3.5 DL-methionine 1.7 2.1 Phenylalanine 0.4 4.9 Proline 0.8 3.4 Serine 3.9 4.6 Threonine -0.8 3.9 Tryptophane 8.3 3.0 L-tyrosine 2.6 1.0 DL-valine 3.5 -1.6

1Amino acids were tested in random order, three per test, at amounts of 50 µg and 200 µg per oviposition site. 2Oviposition was expressed as follows: sample - blank (no application) × 100% control - blank where the extract used was 0.25 g FW squash leaf equivalent, obtained by extraction with 95% ethanol. Total eggs layed per experiment ranged from 1341 to 3778. highest response was 8.3% of control (Table 5), all other responses were considerably lower, sometimes slightly negative. These responses were far below those obtained by extracts of the PC bands (Table 4), where six bands caused increases in excess of 20% of the control and the highest response was 52.8%. It therefore seems likely that some of the non-protein amino acids, which are ubiquitous in the cucurbit family (Noe & Fowden 1960, Fang et al. 1961, Dunnill & Fowden 1965) are responsible for stimulation of oviposition. Since these particular non-protein amino acids seem to be limited to the cucurbits, a particular composition with respect to identity and relative concentrations could explain pickleworm host plant speciﬁcity. Peterson & Elsey: Chemical Factors for Pickleworm Oviposition 491

TABLE 6. INFLUENCE OF ‘WHOLE LEAF’ VOLATILES1 ON OVIPOSITION ON ARTIFICIAL SITES TREATED WITH SQUASH LEAF EXTRACTS.2

Treatment Oviposition (% Total) Mean SEM

Extr. without volatiles 38.4 37.9 45.4 42.8 34.4 39.8 a3 1.9 Extr. with volatiles 59.3 54.7 53.6 46.6 49.8 52.8 b 2.2 Blank (no application) 2.4 7.4 1.0 10.6 15.8 7.4 c 2.7

Total oviposition 4687 686 3415 1985 2210

1A small whole leaf sandwiched between glass ﬁber pads; bottom of leaf facing treated side. 2Extracts free of volatiles; 0.25 FWE of aqueous polyamide fraction. 3Means followed by the same letter are not signiﬁcantly different. Means separation by Fisher’s LSD (P = 0.05).

TABLE 7. INFLUENCE OF LEAF GLANDULAR TRICHOME EXUDATES1 ON OVIPOSITION ON ARTIFICIAL SITES TREATED WITH SQUASH LEAF EXTRACTS.2

Treatment Oviposition (% Total) Mean SEM

Ext. without volatiles 40.8 40.9 45.7 42.9 43.1 42.7 a3 0.89 Ext. without volatiles, 56.4 56.0 52.3 54.8 54.3 54.8 b 0.72 With trichome droplets Blank (no application) 2.8 3.1 2.0 2.3 2.7 2.6 c 0.20

Total oviposition 5231 6840 9283 7230 4663

1Approximately 200 droplets (trichome exudates) applied between two layers of glass ﬁber pads. 2Extracts free of volatiles; 0.25 g FWE of aqueous polyamide fraction. 3Means followed by the same letter are not signiﬁcantly different. Means separation by Fisher’s LSD (P = 0.05).

Volatiles originating from whole leaves increased oviposition signiﬁcantly (Table 6). A similar increase was obtained when ‘whole leaf’ volatiles were substituted by volatiles originating from droplets produced by leaf trichome glands (Table 7). Subse- quent research (Peterson, et al. 1994) showed that these volatiles primarily cause increased attraction rather than stimulation of oviposition. Small, highly polar and non-volatile compounds serve as oviposition stimulants. Most, if not all of these compounds carry a charge, are amphoteric, and most react positively for primary or secondary amino functions. The active compounds were stable and hydrolysis did not decrease activity.

ENDNOTE Mention of a trademark or proprietary product does not constitute a guarantee or warranty of the product by the U.S. Dep. Agric. and does not imply its approval to the exclusion of other products that also may be suitable.

REFERENCES CITED

CORLEY, W. L. 1973. Response of muskmelon botanical varieties to pickleworm infestation. HortScience 13:286-287. 492 Florida Entomologist 78(3) September, 1995

DAY, A., P. E. NUGENT, AND J. F. ROBINSON. 1961. Resistance of squash varieties to the pickleworm and the value of resistance to insecticide control. J. Econ. En- tomol. 54: 1191-1197. DILBECK, J. D., AND T. D. CANERDAY. 1968. Resistance of Cucurbita to the pickleworm. J. Econ. Entomol. 61:1410-1413. DUNNILL, P. M., AND L. FOWDEN. 1965. The amino acids of seeds of the cucurbitaceae. Phytochemistry. 4:933-944. ELSEY, K. D. 1981. Pickleworm: Survival, development, and oviposition on selected hosts. Ann. Entomol. Soc. America 74:96-99. ELSEY, K. D., AND T. L. MCFADDEN. 1981. Pickleworm: oviposition on an artificial surface treated with cucurbit extract. J. Econ. Entomol. 74:473-474. ELSEY, K. D., AND E. V. WANN. 1982. Differences in infestation of pubescent and glabrous forms of cucumber by pickleworms and melonworms. HortScience 17:253-254. ELSEY, K. D., T. L. MCFADDEN, AND R. B. CUTHBERT. 1984. Improved rearing system for pickleworm and melonworm (Lepidoptera: Pyralidae). J. Econ. Entomol. 77:1070-1072. ELSEY, K. D. 1985. Resistance mechanisms in Cucurbita moschata to pickleworm and melonworm (Lepidoptera: Pyralidae). J. Econ. Entomol. 78:1048-1051. ELSEY, K. D., J. A. KLUN, AND M. SCHWARTZ. 1989. Pickleworm sex pheromone: potential for use in cucumber pest management. J. Agric. Entomol. 6:275-282. FANG, S-D., L-C LI, C-I NIU, AND K. F. TSENG. 1961. Chemical studies on Cucurbita moschata Duch. I. The isolation and structural studies of cucurbitine, a new amino acid. Scientia Sinica Vol. X, No. 7, 845-851. NOE, F. F., AND L. FOWDEN. 1960. β-Pyrazol-l-ylalanine, an amino acid from watermelon seeds (Citrullus vulgaris). Biochem. J. 77:543-546. PETERSON, J. K., R. J. HORVAT, AND K. D. ELSEY. 1994. Squash leaf glandular trichome volatiles: Identification and influence on behavior of female pickleworm moth [Diaphania nitidalis (Stoll.)] (Lepidoptera: Pyralidae). J. Chem. Ecol. 20:2099-2109. PULLIAM, T. L. 1979. The effect of plant types and morphology of the cucumber, Cucu- mis sativus L., on infestation by the pickleworm, Diaphania nitidalis (Stoll.). M.S. Thesis, North Carolina State University. QUISUMBING, A. R., AND R. L. LOWER. 1975. Screening cucumber cultivars for pickleworm resistance. HortScience 10:147 (Abstr.). STAHL, E. 1969. Thin-layer chromatography. Second Edition. Springer-Verlag, New York, Heidelberg, Berlin. THIEME, H. 1964. Die Phenolglykoside der Salicaceen. 2. Mitteilung:Isolierung und Nachweis. Pharmazie, 19:471-475. WEHNER, T. C., K. D. ELSEY, AND G. G. KENNEDY. 1985. Screening for antibiosis to pickleworm. HortScience 20:1117-1119. WINTER, M., AND K. HERMANN. 1986. Bestimmung von Hydroxyzimtsäure- Derivaten in Obst und Gemüse durch HPLC nach Polyamid- Fraktionierung. Lebensmit- telchem. Gerichtl. Chem. 40:6-22.

Orozco-Santos et al.: Floating Row Cover & Transparent Mulch 493

FLOATING ROW COVER AND TRANSPARENT MULCH TO REDUCE INSECT POPULATIONS, VIRUS DISEASES AND INCREASE YIELD IN CANTALOUPE

MARIO OROZCO-SANTOS, OCTAVIO PEREZ-ZAMORA AND OSCAR LOPEZ-ARRIAGA Instituto Nacional de Investigaciones Forestales y Agropecuarias Centro de Investigaciones del Paciﬁco Centro, Campo Experimental Tecoman, Apartado postal 88 Tecoman, Colima 28100, México

ABSTRACT

The influence of floating row cover on cantaloupe, alone or combined with transparent polyethylene mulch, was evaluated to determine its effect on population densities of aphids, sweetpotato whitefly, Bemisia tabaci (Gennadius), and the vegetable leafminer, Lyriomyza sativae Blanchard, as well as virus incidence and yield of cantaloupe, Cucumis melo L. Cv. Durango. The study was carried out in a dry tropic region of Colima, Mexico. Aphids, sweetpotato whitefly and vegetable leafminer were completely excluded by floating row cover while the plots were covered. Transparent polyethylene mulch reduced aphids and whitefly populations, but did not affect infestation by L. sativae. Floating row cover delayed the appearance of virus-diseased plants for two weeks with respect to control (bare soil). Furthermore, the transparent mulch reduced virus incidence. The yield (weight) and number of cantaloupe melons harvested were nearly four-fold higher in mulch plus row cover plots (when the cover was removed during perfect flowering) compared with the control plots, while the yield with floating row cover alone was tripled. The yield from mulched plants alone and covered plants (the cover removed during vegetative growth) was higher than that from plants grown on bare soil.

Key Words: Cucumis melo, aphids, sweetpotato whiteﬂy, vegetable leafminer.

RESUMEN Se evaluó la influencia de las cubiertas flotantes, solas o combinadas con una co- bertura de plástico transparente, sobre la densidad de población de áfidos y de mosca blanca de la batata, Bemisia tabaci (Gennadius), y sobre la infestación del minador de la hoja Lyriomyza sativae Blanchard, así como sobre la incidencia de virus y la pro- ducción dé melón “Cantaloupe”, Cucumis melo L. Cv. Durango. La investigación se realizó en la región tropical seca del estado de Colima, México. Los áfidos, la mosca blanca de la batata y el minador de la hoja fueron completamente excluídos por las cubiertas flotantes, mientras las plantas estuvieron protegidas con la tela de polipropi- leno. El acolchado con plástico transparente redujo la población de áfidos y mosca blanca, pero no tuvo efectos sobre la infestación por L. sativae. La cubierta flotante re- trasó durante dos semanas la presencia de plantas afectadas por virus con respecto al testigo (suelo desnudo). El plástico transparente también redujo la incidencia de virus. El número de frutos y el rendimiento del melón en las parcelas con cubierta de plástico + cubierta flotante (la cubierta fue retirada durante la floración perfecta) fue casi cuatro veces más alto en comparación a las parcelas con suelo desnudo, mientras que en aquellas con cubierta flotante sola, el rendimiento se triplicó. La producción de las plantas con cubierta plástica sola o de las plantas con cubierta flotante (retirada durante el crecimiento vegetativo) fue más alta que la de las plantas cutivadas sobre suelo desnudo.

494 Florida Entomologist 78(3) September, 1995

Cantaloupe melon, Cucumis melo L., is the major cucurbit crop in the dry tropic areas of Colima State, Mexico, during fall and winter, with more than 1500 ha grown annually. Recently in this area, aphid-borne virus diseases have been a problem for cucurbit crops. In Mexico, a cucumovirus (cucumber mosaic) and potyviruses (watermelon mosaic-2, papaya ringspot-W and zucchini yellow mosaic) are the major virus diseases affecting squash, watermelon, cucumber and cantaloupe (Delgadillo et al. 1989). These viruses are transmitted efficiently by aphids in a nonpersistent manner (Nameth et al. 1986). The sweetpotato whitefly, Bemisia tabaci Gennadius, is an important pest of agricultural and horticultural crops throughout the world, causing damage directly by feeding and indirectly through the excretion of honeydew (Byrne et al. 1990) and the transmission of virus diseases (Cohen 1990). In the past decade, this insect has been reported as a serious pest of various crops in Florida (Schuster & Price 1987), Arizona (Butler 1982) and California (Perring et al. 1991). Recently, it has been found in extremely high populations in the agricultural areas of Arizona, California, Texas and northwestern Mexico (Gill 1992). Also, in the last two years, sweetpotato whitefly has become the most important pest in melon-growing areas of Colima, Mexico (Farias et al. 1994). Other pests affecting melon crops are the vegetable leafminer, Lyriomyza sativae Blanchard, and some species of Diabrotica. Most of the management strategies against cucurbit virus diseases have been targeted at the aphid-borne viruses. Several measures are used to influence the behavior of the insect vectors in an attempt to interfere with virus transmission, including the use of transparent polyethylene mulch (Brown et al. 1993, Lecoq & Pitrat 1983, Orozco et al.1994), oil sprays (Simons & Zitter 1980) and floating row covers (Natwick & Durazo 1985, Natwick & Laemmlen 1993, Orozco et al. 1994, Perring et al. 1989, Webb & Linda 1992). However, in dry tropic conditions, the use of oil sprays can cause phytotoxic symptoms on the foliage and reduce the fruit yield of cantaloupe (Orozco et al. 1994). The use of plastic mulches in agriculture continues to increase in high technology systems for horticultural crop production, and more recently, floating row covers of spunbonded polyester or polyethylene have been introduced to protect cucurbits from insect-vectored viruses and their aphid and whitefly vectors (Natwick & Laemmlen 1993, Perring et al. 1989, Webb & Linda 1992) and from direct damage from pests, such as the vegetable leafminer (Natwick & Laemmlen 1993), melonworm, Diaphania hyalinata (L.), pickleworm, D. nitidalis (Stoll) (Webb & Linda 1992) and cucumber beetles (Adams et al. 1990). Plastic mulch in combination with floating row covers show promise as a method for excluding insects, reducing virus diseases and increasing yield of cucurbit crops (Orozco et al. 1994, Ramirez & Lopez 1991, Webb & Linda 1992). However, there is little information on the development of the cantaloupe crop grown under floating row cover in a semi-arid tropical region. This study was carried out to evaluate the effect of transparent polyethylene mulch and floating row covers, with or without transparent mulch, on insect populations, viral diseases and yield of melons grown in the field under semi-arid tropical conditions.

MATERIALS AND METHODS The field experiment was conducted at the Tecoman Experimental Station in Colima, Mexico during the winter-spring of 1992-1993. On February 19, 1993, melon (Cucumis melo L.) Cv. Durango was directly seeded at 23 cm intervals in the center of 1.78 m wide beds, and plots were watered with a drip irrigation system to promote germination. Each bed had an emitter tube with emission spacing of 0.5 m, and flow rate of 2.3 liter per h. Five treatments were evaluated: (1) transparent polyethylene mulch (TPM), (2) floating row cover (FRC, spunbonded polyethylene fiber: AGRIBON-

Orozco-Santos et al.: Floating Row Cover & Transparent Mulch 495

17, Bonlam S.A. de C.V. San Luis Potosi, Mexico) removed at the vegetative stage (20 March, before staminated flowers appeared in the plants) or (3) the reproductive stage (28 March, when 50% of the plants had their first perfect flower); and (4) TPM + FRC removed at the reproductive stage. Bare soil plots (5) were included as the check (control). After seeding, FRC was set on the beds with both edges and ends of covers secured with soil. Weeds became a problem under the covers, mainly in the floating row cover with bare soil, and it was necessary to control grass with the herbicide Fluazifop-butil (FusiladeTM); broad-leaf weeds were removed manually. In TPM and bare soil plots, eleven insecticide sprays were applied during all the crop cycle, mainly for sweetpotato whitefly control, while in FRC plots, alone or combined with TPM removed at reproductive stage, only five sprays were necessary after the covers were removed. The insecticides used were endosulfan, methamidophos, diazinon, bifenthrin, delta- methrin, cyfluthrin, lambdacyhalothrin, and permethrin. Each treatment consisted of five 6.9 m-long beds replicated four times in a randomized complete block design. Yellow pan plastic traps for aphids (28 × 22 × 12 cm) containing a mixture of water and detergent were placed in the middle of one row in each plot, and the number of aphids trapped were determined twice weekly to determine the effectiveness of each treatment in delaying aphid infestation. Aphids were stored in 70% alcohol and were identified to species using several aphid taxonomy keys (Basky 1993, Peña 1991, Pike et al. 1990). Counts of sweetpotato whitefly adults were made on 16 plants (two leaves per plant) in early morning (7-8 h A.M.), before flight activity occurred. Leaves were turned over gently and adults visually counted (Butler & Henneberry 1991). On 24 March and 1 April (34 and 41 days after planting, respectively), the numbers of sweetpotato whitefly nymphs on five leaflets from two lower leaves were counted from eight plants with the aid of a 10x dissecting microscope. Also, on 19, 26 March, 2 and 12 April, vegetable leafminer damage was assessed on the main guide shoot on 16 plants per plot. A leaf was considered damaged if more than three miners per leaf were found. To evaluate virus incidence, all of the plants in each plot showing disease symptoms were counted weekly. At harvest, plant tissue was analyzed by ELISA test (Clark & Adams 1977) to detect infection by cucumber mosaic virus (CMV), watermelon mosaic virus-2 (WMV-2), papaya ringspot virus-W (PRSV-W), zucchini yellow mosaic virus (ZYMV) and squash mosaic virus (SqMV). Yield was assessed by picking all the fruits in two central beds of each experimental plot and sorting them into marketable and culled categories. The harvest was done every day during two weeks, and marketable fruit (based on export regulations) was sized 9, 12, 15, 18 and 23 fruits per carton.

RESULTS AND DISCUSSION

Fourteen aphid species were caught in yellow water pan traps throughout the experimental period. These species were (% of total): Aphis spiraecola Patch (23%), Uro- leucon ambrosiae (Thomas) (22%), Myzus persicae (Sulzer) (13%), A. gossypii (Glover) (12%), Rhopalosiphum ruﬁabdominalis (Sasaki) (11%), R. maidis (Fitch) (4%), A. craccivora (Koch) (3%), Schizaphis graminum (Rondani) (3%), A. nerii (Boyer de Fonsc) (2%), Macrosiphum euphorbiae (Thomas) (2%), Lipaphis erysimi Kaltenbach (2%), R. padi (L.) (1%), Toxoptera aurantii Boyer de Fonsc. (1%) and A. fabae Scopoli (1%). No alate aphids were caught in yellow traps in the plots with FRC until the covers were removed, while on the bare soil plots the highest aphid populations were caught during the ﬁrst 50 days after sowing (Fig. 1). The use of TPM resulted in fewer aphids

496 Florida Entomologist 78(3) September, 1995

Figure 1. The number of winged aphids collected in yellow water pan traps placed in cantaloupe plots with transparent mulch, floating row cover and bare soil. VS and RS = row cover removed at vegetative and reproductive stage, respectively. per trap until 41 days after sowing. However, after 50 days, the numbers of trapped aphids were about the same for all the treatments. Sweetpotato whitefly populations were extremely high, mainly in the bare soil control plots, where 52, 35 and 32 adults per leaf were recorded 25, 32 and 38 days, respectively, after sowing (Fig. 2). Populations of whitefly nymphs and adults were excluded from the melon plants while the polypropylene row cover was on the bed. Transparent polyethylene mulch reduced the number of whitefly adults per leaf with respect to bare soil on some dates during the first 50 days after sowing. In plants with FRC, alone or combined with TPM removed at the reproductive stage, no whitefly nymphs were detected for 35 and 42 days after sowing (Table 1), while leaves from the plants with FRC removed at the vegetative stage had no nymphs at 35 days, and only a few nymphs (4.5 per cm2) at 42 days. The TPM plots significantly reduced the number of nymphs (about one half) compared to the control plots. In contrast, bare soil plots had significantly greater numbers of whitefly nymphs on both dates than the FRC and TPM treatments. The appearance of virus disease symptoms were delayed one week on plants when FRC was removed at vegetative stage, and symptoms were delayed 2-3 weeks when FRC was removed at reproductive stage (Fig. 3). At harvest (67 days after sowing), the control plots had the highest percentage (24%) of virus-infected plants followed by FRC on bare soil plots (16%). The TPM plots and the TPM + FRC plots showed the lowest (5-12%) number of virus-diseased plants. ELISA tests indicated that PRSV-W was the most prevalent virus in this study, however, ZYMV, CMV, WMV-2 and SqMV were also detected. The PRSV-W, ZYMV, CMV and WMV-2 are transmitted in a nonpersistent manner by some aphid species, while SqMV is transmitted by beetles (Nameth et al, 1986). Leafminer damage was avoided for 35 and 42 days after sowing with FRC removed at the vegetative and reproductive stage, respectively, compared with bare soil (Fig.

Orozco-Santos et al.: Floating Row Cover & Transparent Mulch 497

Figure 2. Effect of transparent mulch and ﬂoating row cover on whiteﬂy populations in cantaloupe foliage. VS and RS = Row cover removed at vegetative and reproductive stage, respectively.

4). During the ﬁrst 35 days, control and TPM plots had the highest leafminer damage; however, by 52 days the percentage of damaged leaves was similar in all the treatments. Yield of cantaloupe melons, expressed as mean number and weight of fruit was sig- niﬁcantly higher for TPM + FRC plots than TPM and bare soil with or without FRC (Table 2). The yield of TPM + FRC (removed during reproductive stage) plots was

TABLE 1. EFFECT OF TRANSPARENT MULCH AND FLOATING ROW COVER ON THE NUM- BER OF WHITEFLY NYMPHS ON CANTALOUPE LEAVES. TECOMAN VALLEY, COLIMA, MEXICO. 1993.

Nymphs/cm2

26 March 2 April Treatment (35 DAS1) (42 DAS1)

Transparent polyethylene mulch (TPM) 2.7 b4 9.1 c4 Floating row cover (VS2) 0.0 a 4.5 b Floating row cover (RS3) 0.0 a 0.0 a TPM + Floating row cover (RS) 0.0 a 0.0 a Bare soil (Control) 5.4 c 19.1 d

1DAS = Days after sowing. 2VS = Removed at vegetative stage. 3RS = Removed at reproductive stage. 4 = Mean separation within columns by Duncan’s multiple test (P < 0.05). 498 Florida Entomologist 78(3) September, 1995

Figure 3. Effect of transparent mulch and floating row cover on incidence of cantaloupe plants with virus symptoms. VS and RS = row cover removed at vegetative and reproductive stage, respectively. nearly four-fold higher than in control plots, and three-fold higher than with FRC alone. Moreover, the yield obtained in TPM plots and bare soil with FRC (removed during vegetative growth) was higher than that on bare soil plots. The results of this study demonstrate the beneficial effects of FRC alone or combined with reflective mulch for cantaloupe grown in dry tropical conditions. Floating row covers excluded beetles, leafminers, sweetpotato whitefly, and aphids. The exclusion of insect pests reduced the use of insecticides by 50%. In addition, the foliage of plants was not colonized by whiteflies during the first 42 days with FRC removed at the reproductive stage. Furthermore, TPM resulted in fewer winged aphids caught in yellow traps and fewer whitefly nymphs and adults on foliage with respect to control. These results with FRC are similar to previous research (Natwick & Durazo 1985, Natwick & Laemmlen 1993, Orozco et al. 1994, Perring et al. 1989, Webb & Linda 1992). The repellent effect of the plastic mulch on aphids (Brown et al. 1993, Jones 1991) and whitefly (Kelly et al. 1989) has been attributed to its high reflectance of light (Kring & Schuster 1992). Virus disease incidence was low in the control plots; this may be related to the low aphid populations caught in yellow traps (maximum 65 aphids per trap per week). In this experiment, the major viruses found were PRSV-W, ZYMV, CMV and WMV-2 which are transmitted by aphids (Nameth et al. 1986). Because of exclusion of aphid vectors in covered plots, and the fewer numbers of aphids alighting on the plants grown in TPM plots, TPM and FRC had a lower virus incidence compared with the control. These data support the observations of Brown et al. 1993, Kring & Schuster 1992, Natwick & Durazo 1985, Orozco et al. 1994, Perring et al. 1989, and Webb & Linda 1992. The yield in weight of cantaloupe plants grown with TPM + FRC was higher than that of TPM mulch alone and bare soil. This can be attributed to more vigorous growth of plants in TPM plots alone or combined with FRC because of moisture con- Orozco-Santos et al.: Floating Row Cover & Transparent Mulch 499

Figure 4. Effect of transparent mulch and floating row cover on the percentage of cantaloupe leaves damaged by the vegetable leafminer. VS and RS = row cover removed at vegetative and reproductive stage, respectively. servation and higher soil temperature. Preliminary studies have shown a yield increase of cantaloupe with the use of TPM (Orozco et al. 1994) and FRC (Orozco et al. 1994, Perring et al. 1989, Webb & Linda 1992). Also, the effects of transparent mulch on soil temperature (Al-Assir et al. 1992) and moisture conservation (Tindall et al. 1991) have been previously demonstrated. The beneficial effects of polyethylene transparent mulch and floating row cover might be a practical and economical management tool for producing cantaloupe under the dry tropical conditions of Colima, Mexico. However, additional on-farm trials are

TABLE 2. EFFECT OF TRANSPARENT MULCH AND FLOATING ROW COVER ON THE NUM- BER OF FRUIT AND YIELD OF CANTALOUPE. TECOMAN VALLEY, COLIMA, MEX- ICO. 1993.

Weight (kg) of No. of Fruit per Fruit per 6.9 m of Row1 6.9 m of Row1

Transparent polyethylene mulch (TPM) 33.0 c 28.37 c Floating row cover (VS2) 29.5 c 24.69 c Floating row cover (RS3) 54.2 b 47.04 b TPM + Floating row cover (RS) 68.6 a 62.51 a Bare soil (Control) 19.2 d 16.08 d

1 = Mean separation within columns by Duncan’s multiple range tests (P < 0.05). 2VS = Removed at vegetative stage. 3RS = Removed at reproductive stage. 500 Florida Entomologist 78(3) September, 1995 needed to determine whether or not these methods can be practically integrated into commercial cantaloupe production.

ACKNOWLEDGMENTS This study was supported by funds from the Patronato para la Investigación y Ex- perimentación Agrícola del Estado de Colima, México (PIEAEC) and Instituto Nacio- nal de Investigaciones Forestales y Agropecuarias (INIFAP).

REFERENCES CITED

ADAMS, R. G., R. A. ASHLEY, AND M. J. BRENNAN. 1990. Row covers for excluding insect pests from broccoli and summer squash plantings. J. Econ. Entomol. 83: 948-954. AL-ASSIR, I. A., I. G. RUBEIZ, AND R. Y. KHOURY. 1992. Yield response of greenhouse cantaloupe to clear and black plastic mulches. Biological Agriculture and Hor- ticulture 8:205-209. BASKY, Z. 1993. Identification key for alate aphids caught in yellow pan traps. Acta Phytopathologica et Entomologica Hungarica, 28: 71-121. BROWN, J. E., J. M. DANGLER, F. M. WOODS, K. M. TILT, M. D. HENSHAW, W. A. GRIFFEY, AND M. S. WEST. 1993. Delay in mosaic virus onset and aphid vector reduction in summer squash grown on reflective mulches. HortScience 28:895- 896. BUTLER, G. D., JR. 1982. Sweetpotato whitefly, Bemisia tabaci, a new pest of cotton in Arizona. Ariz. Agr. Expt. Sta. P-56: 49-50. BUTLER, G. R., AND T. J. HENNEBERRY. 1991. Effect of oil sprays on sweetpotato whitefly and phytotoxicity on watermelons, squash and cucumbers. Southwestern Entomologist 16:63-72. BYRNE, D. N., T. S. BELLOWS JR., AND M. P. PARRELLA. 1990. Whiteflies in agricultural systems, pp. 227-261 in D. Gerling [ed.] Whiteflies: their bionomics, pest status and management. Intercept. Great Britain. CLARK, M. F., AND A. N. ADAMS. 1977. Characteristics of the microplate method of enzyme-linked immunosorbant assay for the detection of plant viruses. Journal Gen. Virol. 34:475-483. COHEN, S. 1990. Epidemiology of whitefly-transmitted viruses, pp. 211-225 in D. Ger- ling. [ed.] Whiteflies: their bionomics, pest status and management Intercept. Great Britain. DELGADILLO, S. F., T. J. A. GARZON, AND P. A. VEGA. 1989. Cucurbit viruses in Mexico: A survey. Rev. Mex. de Fitopatología 7:136-139. FARIAS-L., J., M. OROZCO-S., AND S. GUZMAN-G. 1994. Effect of Imidacloprid on sweetpotato whitefly (Homoptera: Aleyrodidae) populations on cantaloupe melon. XXIX Cong. Nal. de Entomología. Annual meeting southwestern branch-ESA. Monterrey, N.L., México. p.188-189. GILL, R. J. 1992. A review of the sweetpotato whitefly in southern California. Pan-Pa- cific Entomologist 68:144-152. JONES, R. A. C. 1991. Reflective mulch decreases the spread of two non-persistently aphid transmitted viruses to lupin (Lupinus angustifolius). Ann. appl. Biol. 118:79-85. KELLY, J. W., P. H. ADLER, D. R. DECOTEAU, AND S. LAWRENCE. 1989. Colored reflective surfaces to control whitefly on poinsettia. HortScience 24:1045. KRING, J. B., AND D. J. SCHUSTER. 1992. Management of insects on pepper and tomato with UV-reflective mulches. Florida Entomol. 75:119-129. LECOQ, H., AND M. PITRAT. 1983. Field experiments on the integrated control of aphid-borne viruses in muskmelon, p. 169-176 in R. T. Plumb and J. M. Thresh [eds.] Plant virus epidemiology. Blackwell Scientific Publication. Oxford, Eng- land. Orozco-Santos et al.: Floating Row Cover & Transparent Mulch 501

NAMETH, S. T., J. A. DODDS, A. O. PAULUS, AND F. F. LAEMMLEN. 1986. Cucurbit viruses of California: An ever-changing problem. Plant Disease 70:8-12. NATWICK, E. T., AND A. DURAZO. 1985. Polyester covers protect vegetables from whiteflies and virus disease. California Agric. 39:21-22. NATWICK, E. T., AND F. F. LAEMMLEN. 1993. Protection from phytophagous insects and virus vectors in honeydew melons using row covers. Florida Entomol. 76:120-126. OROZCO-S., M., O. LOPEZ-A., O. PEREZ-Z., AND F. DELGADILLO-S. 1994. Effect of transparent mulch, floating row cover and oil sprays on insect populations, virus diseases and yield of cantaloupe. Biological Agriculture and Horticulture 10:229- 234. PEÑA, M. R. 1991. Manual de prácticas del curso: Afidos como vectores de virus. Cen- tro de Fitopatología. Colegio de Postgraduados. Montecillo, México. 148 p. PERRING, T. M., A. COOPER., D. J. KAZMER, C. SHIELDS, AND J. SHIELDS. 1991. New strain of sweetpotato whitefly invades California vegetables. California Agric. 45:10-12. PERRING, T. M., R. N. ROYALTY, AND C. A. FARRAR. 1989. Floating row covers for the exclusion of virus vectors and the effect on disease incidence and yield of cantaloupe. J. Econ. Entomol. 82:1709-1715. PIKE, K. S., L. BOYDSTON, AND D. ALLISON. 1990. Alate aphid vivipare associated with small grains in North America: A key and morphometric characterization. J. Kansas Entomol. Soc. 63:559-602. RAMIREZ, V. J., AND V. C. LOPEZ. 1991. Efecto de cubiertas flotantes y acolchado de plástico sobre el control de virosis e incremento de rendimientos en pepino y melón. Memorias del XVIII Cong. Nal. de Fitopatología. Puebla, México. p. 205. SCHUSTER, D. J., AND J. F. PRICE. 1987. The western flower thrips and the sweetpotato whitefly: New pests threatening Florida tomato production. Univ. of Fla., IFAS, Bradenton GCREC Res. Rep. BRA1987-16. SIMONS, J. N., AND T. A. ZITTER. 1980. Use of oils to control aphid-borne viruses. Plant Disease 64:542-546. TINDALL, J. A., R. B. BEVERLY, AND D. E. RADCLIFFE. 1991. Mulch effect on soil properties and tomato growth using micro-irrigation. Agron. J. 83:1028-1034. WEBB, S. E., AND S. B. LINDA. 1992. Evaluation of spunbonded polyethylene row covers as a method of excluding insects and viruses affecting fall-grown squash in Florida. J. Econ. Entomol. 85:2344-2352.

502 Florida Entomologist 78(3) September, 1995

PROBABILITY OF DETECTING CARIBBEAN FRUIT FLY (DIPTERA: TEPHRITIDAE) INFESTATIONS BY FRUIT DISSECTION

WALTER P. GOULD USDA-ARS, Subtropical Horticulture Research Station, Miami, FL

ABSTRACT Quarantine inspectors search for fruit fly infestations in incoming shipments by visual inspection and by dissecting or cutting a sample of fruit in each shipment. The reliability of the latter procedure for detecting fruit fly larvae is questionable and, therefore, a test was conducted to determine its effectiveness. Infested grapefruit, mangoes, guavas, and carambolas were cut open to determine the efficacy of cutting fruit in detecting larval infestations of Caribbean fruit flies, Anastrepha suspensa (Loew). From 1 to 36% of the larvae were detected by dissection, but 17.9 to 83.5% of the infested fruit were detected. The percentage of Caribbean fruit fly larvae present in an infested fruit that can be detected by cutting varies with the type of fruit infested. The overall percentage of infested fruit detected also varies in the same manner. Carambolas were the easiest fruit in which to find larvae and green guavas the most difficult. Overall, only 9.5% of the larvae in field-infested guavas and 16.9% of those in cage-infested fruits were detected. Between 17.9% (green guavas) and 83.5% (carambolas) of the infested fruit was detected. There was considerable variation in the number of larvae found by different inspectors.

Key Words: Caribbean fruit ﬂy, Anastrepha suspensa, larval detection.

RESUMEN

Los inspectores cuarentenarios detectan infestaciones de moscas de las frutas bus- cando frutas dañadas y cortando una muestra de ellas en un determinado carga- mento. Se disectaron toronjas, mangos, guayabas, y carambolas infestadas para determinar la eﬁciencia del corte de frutas en la deteccción de la infestación larval de Anastrepha suspensa (Loew). Del 1 al 36% de las larvas fueron detectados mediante la disección; sin embargo, este método permitió encontrar del 17.9 al 83.5% de las frutas infestadas. El porcentaje de larvas de moscas de las frutas del Caribe detectado cortando frutas infestadas varió con los diferentes tipos de frutas. El porcentaje de frutas infestadas varió de la misma manera. La fruta en la cual fue más fácil encontrar larvas fue la carambola y la más difícil la guayaba verde. En resumen, solamente fue detectado el 9.5% de las larvas en las guayabas, y el 16.9% en las frutas que fueron infestadas en jaulas. Fue detectado entre el 17.9% (guayabas verdes) y el 83.5% (carambolas) de las frutas infestadas. Hay considerable variación en el número de larvas encontradas por diferentes inspectores.

Prevention of the spread of undesirable pests is a worldwide problem and quaran- tines have been established for many pest organisms. The family Tephritidae includes many species of fruit ﬂies whose larvae attack valuable cultivated fruits and vegetables. Two dozen important species and hundreds of species of minor importance exist around the world (Robinson & Hooper 1989). Most fruit ﬂies lay their eggs under the skins of fruits and vegetables. The larvae burrow and feed inside the fruit or vegetable

Gould: Fruit Fly Detection by Fruit Cutting 503 making it unpalatable and accelerating decay. Many countries have plant protection agencies whose mission is to exclude these unwelcome pests (Shannon 1994). The main line of defense is the vigilance of inspectors who check incoming fruit and vegetable shipments. The main method of detecting fruit flies in a shipment of a commodity is to sample a number of fruits or vegetables and examine them for possible signs of infestation, a procedure that often includes cutting open some of the fruits and ex- amining them for eggs and/or larvae (Anon. 1993). Cutting fruits and vegetables to detect insect infestation is time consuming and only a fraction of a shipment can be inspected in this way, but there are no feasible alternatives. Holding the commodity until any larvae emerge is not practical because it would hold up the shipment for too long and destroy its value. Other methods, such as detecting sound of feeding larvae, have not yet proven practical (Calkins & Webb 1988, Hansen et al. 1988, Hansen et al. 1992, Harrison et al. 1993). Many factors can affect the probability of finding a larva, such as the ability of the inspector, the stage of the insect present, and the type of fruit. In this study, I determined the probability of detecting Caribbean fruit fly, Anastrepha suspensa (Loew), larvae given three variables. I looked at four types of fruit, different stages of ripeness, and variations in the efficiency of the inspectors.

MATERIALS AND METHODS Four different types of fruit were used. Guavas were obtained from J. R. Brooks and Son, a local packing house, and ripe guavas were harvested from trees on the USDA station in Miami. Mangoes and carambolas were also obtained from J. R. Brooks and Son. Grapefruits were obtained from Bernard Egan & Co., a packing house in central Florida. Guavas received from the packing house were separated into ripe (yellow) and immature (green). The guavas had a significant field infestation. Grapefruits, mangoes, and carambolas did not have a detectable infestation in the field so they were exposed to approximately 100,000 Caribbean fruit flies in an out- door cage for 1 to 3 days (carambolas and mangoes), or 7 days (grapefruit). Field-infested guavas were cut open 1 day after being received from the packing house or being picked and could contain all stages of eggs and larvae. Mangoes were held 5 to 13 days after infestation before being cut open, and grapefruit and carambolas were held for 7 days after infestation before cutting. These holding periods allowed development of larvae to the third instar and precluded presence of eggs and early instars. The fruits were randomly divided into a control and a treatment group. The number of fruits used per replicate depended on the number available, between 15 and 150 per replicate, and the experiment was replicated three to five times for each type of fruit. To compare effect of fruit type and ripeness, cutting was done by one inspector using the following guidelines: Each fruit was cut with a 15-cm utility knife into 1-cm thick slices (from 5 to 10 depending upon fruit size). Mangoes were sliced from each side until the seed was reached, then turned 90° and sliced again until the seed was reached. Each slice was examined for the presence of Caribbean fruit fly larvae, and the total number observed was recorded. Sliced fruits with larvae and unsliced control fruits were placed in 1-liter plastic buckets containing vermiculite and held for several weeks. The vermiculite was then sifted for larvae and pupae. Another experiment was conducted to examine the effect of the inspector on the probability of detection. Five additional inspectors were recruited for the test. All were either active employees of the Florida Department of Agriculture & Consumer Services, Division of Plant Industry (DPI) or ex-DPI members who had routinely cut fruits looking for larvae. The five DPI or ex-DPI inspectors and the original inspector

504 Florida Entomologist 78(3) September, 1995 each cut 10 or 20 infested grapefruit (fly cage infested). The five DPI inspectors were allowed to cut fruit any way they chose. The original inspector followed the previous guidelines. Percentage of larvae detected by cutting and percentage of infested fruits detected by cutting were transformed using the angular transformation (arcsin) and both transformed and untransformed data analyzed using unbalanced ANOVA by GLM (SAS Institute 1988). Untransformed percentages are presented in the tables for comparison. If the ANOVA F-test was significant at the 0.05 level, the Waller-Duncan K-ratio t-test (SAS Institute 1988) was used to separate means. T-test and Proc Reg were also used (SAS Institute 1988).

RESULTS AND DISCUSSION

There was no signiﬁcant difference between the number of larvae emerging from control and the cut fruits (paired t-test, P=.998, α=0.05). Therefore, very few larvae were physically damaged by the cutting, and tearing the fruit open did not affect the larvae’s ability to mature and leave the fruit. The infestation levels in the fruits varied (Table 1). Green guavas from Brooks had the lowest infestation level, ripe guavas from Brooks slightly higher, and ripe guavas from the USDA station, much higher levels. The infestations from the cage varied as well. In most of the replications, the infestations were low (one or fewer larvae per fruit); however, there were several with moderate infestations and several with extremely high infestations. There was considerable variation between replications in the percentage of larvae detected by cutting for all fruit types (Table 2). Green guavas overall had the lowest percentage of larvae detected (1.1%). Carambolas had the highest (about 36%). Only 9.5% of the larvae in ﬁeld-infested guavas were detected by cutting. Only 16.9% of the larvae in all of the cage-infested fruits (carambolas, grapefruits, mangoes) together were detected by cutting. From 64 to 98% of larvae escaped detection. However, since many larvae were often found in one fruit, there was a higher probability of an infested fruit being de-

TABLE 1. FRUIT INFESTATION LEVELS OF A. SUSPENSA (LARVAE PER FRUIT).

Field-Infested Guavas Replication Mean

12345

Green-Brooks 6.000 0.167 0 0.200 0 1.273 Ripe-Brooks 8.662 0.967 0.240 0.200 — 2.517 Ripe-USDA station 39.120 38.920 13.320 — — 30.453

Cage-Infested Fruits Replication Mean

1234

Grapefruits 0.633 3.840 0.680 58.580 15.933 Mangoes 225.750 0.100 0.750 1.600 57.050 Carambolas 20.220 11.000 0.320 — 10.513

Gould: Fruit Fly Detection by Fruit Cutting 505

TABLE 2. PERCENTAGE OF A. SUSPENSA LARVAE DETECTED BY CUTTING.

Fruit type Replication Mean1

1234

Carambolas 9.0 35.8 62.5 — 35.8a Ripe Guavas-Brooks 29.6 13.5 25.6 — 22.9ab Grapefruits 15.8 7.8 11.8 2.8 9.5bc Mangoes 1.9 0.0 20.0 0.0 5.5bc Ripe Guavas-USDA station 4.7 0.0 9.0 — 4.6bc Green Guavas-Brooks 1.1 2.2 0.0 — 1.1c

1Means within a column followed by the same letter are not significantly different by ANOVA followed by Waller-Duncan K-ratio t-test. tected than Table 2 would indicate (Table 3). As many as 83.5% of carambolas with larvae were detected. This is probably because carambolas are a fairly translucent fruit and larvae are more easily observed. Green guavas were the most difficult in which to detect larvae; only 17.9% of the infested fruit were detected. This may be because they were inspected one day after we received them, thus the fruit flies could be in the egg stage or very early instar and very difficult to see. This fruit is also tough and difficult for the inspector to cut up. Analysis of variance found that in the percentage of infested fruit detected, there was a significant difference only between the carambolas and all of the other fruit, except ripe guavas from Brooks. None of the other pairwise comparisons was significant. The effect of size of infestation and the percentage of larvae found did not show a high correlation, but generally the more larvae that were present, the lower the percentage found by the inspector. This may be due to a fatigue factor in those extreme infestations where thousands of larvae were present. Taking log ten of the number of larvae present makes the distribution of the percentage of larvae detected look like the normal distribution. The best regression had an r2 of 0.11; with an outlier deleted, it improved only to 0.29.

TABLE 3. PERCENTAGE OF A. SUSPENSA INFESTED FRUITS DETECTED BY CUTTING.

Fruit Type Replication Mean1

12345

Carambolas 60.5 90.0 1 — — 83.5a Ripe Guavas-Brooks 77.8 66.7 60.0 0.0 — 51.1ab Ripe Guavas-USDA station 66.7 36.0 48.0 36.0 — 46.7b Grapefruits 23.3 22.2 58.0 — — 34.5b Mangoes 85.0 0.0 28.6 0.0 — 28.4b Green Guavas-Brooks 7.2 22.3 0.0 60.0 0.0 17.9b

1Means within a column followed by the same letter are not signiﬁcantly different by ANOVA followed by Waller-Duncan K-ratio t-test on arcsin transformed data. Untransformed means are presented.

506 Florida Entomologist 78(3) September, 1995

TABLE 4. PERCENTAGE OF A. SUSPENSA LARVAE DETECTED BY CUTTING BY DIFFERENT INSPECTORS.

Inspector 1 08.4 Inspector 2 49.1 Inspector 3 29.5 Inspector 4 17.8 Inspector 5 17.7 Inspector 6 16.3

Mean 23.1 St. Dev. 14.4

In the inspector comparison, the percentage of larvae found varied from 8% to 49% (Table 4). Most of the difference could be attributed to the methods each inspector used. Inspector 1 was obviously hurrying and found the fewest larvae. Inspector 2 shredded the fruit into tiny pieces, took much longer than the others, and found more larvae. The other three inspectors’ results were very similar. Because the inspectors knew their results were being compared and studied, they probably found more larvae than they would have otherwise. The main reason for performing this test was to determine the relationship between the method and the inspector so we could assess the results of the first part of this study compared to ‘real world’ inspectors. We con- cluded that the data obtained with the inspector in the first part of this study (using a consistent cutting method) were reasonably comparable to those of the DPI inspectors. In summary, the percentage of Caribbean fruit fly larvae present in an infested fruit that can be detected by cutting varied with the type of fruit infested. The overall percentage of infested fruit detected also varied in the same manner. Carambolas were the easiest fruit in which to find larvae and green guavas the most difficult. Overall, only 9.5% of the larvae in guavas and 16.9% of those in cage-infested fruits were detected. Between 17.9 and 83.5% of the infested fruit was detected. There was considerable variation among inspectors based on the method used to inspect the fruit.

ACKNOWLEDGMENTS I thank Craig Campbell, J. R. Brooks and Son, Inc., and Bernard Egan and Co. for contributing fruit used in this study. I also thank D. Chalot, W. Francillon and L. Lodyga, of the Florida Department of Agriculture and Consumer Services, Division of Plant Industry for their assistance. I thank W. Montgomery, R. Pantaleon, and D. Storch of USDA-ARS for their assistance. I thank G. Hallman and M. Hennessey of USDA-ARS for their critical review of this manuscript, and also G. Hallman and E. Schnell of USDA-ARS for translation of the abstract to Spanish.

REFERENCES CITED

ANONYMOUS. 1993. Animal Plant Health Inspection Service. Plant Import Manual: Nonpropagative. U.S. Govt. Printing Ofﬁce. Washington, D.C. CALKINS, C. O., AND J. C. WEBB. 1988. Temporal and seasonal differences in movement of the Caribbean fruit ﬂy larvae in grapefruit and the relationship to detection by acoustics. Florida Entomol. 71: 409-416.

Gould: Fruit Fly Detection by Fruit Cutting 507

HANSEN, J. D., C. L. EMERSON, AND D. A. SIGNOROTTI. 1992. Visual detection of sweetpotato weevil by non-invasive methods. Florida Entomol. 75: 369-375. HANSEN, J. D., J. C. WEBB, J. W. ARMSTRONG, AND S. A. BROWN. 1988. Acoustical detection of oriental fruit ﬂy (Diptera: Tephritidae) larvae in papaya. J. Econ. En- tomol. 81: 963-965. HARRISON, R. D., W. A. GARDNER, W. E. TOLLNER, AND D. J. KINARD. 1993. X-ray com- puted tomography studies of the burrowing behavior of fourth-instar pecan weevil (Coleoptera: Curculionidae). J. Econ. Entomol. 86: 1714-1719. ROBINSON, A. S., AND G. HOOPER, (eds.) 1989. Fruit Flies, Vol. 3A World Crop Pests. Elsevier, Amsterdam. 372 pp. SAS INSTITUTE. 1988. SAS/STAT User’s Guide. Release 6.03 Ed. SAS Institute, Cary, NC. SHANNON, M. J. 1994. APHIS, pp. 1-10 in J. L. Sharp and G. J. Hallman [eds.] Quar- antine treatments for pests of food plants. Westview Press, Boulder, Colorado. 290 pp.

♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦

Smith & Zungoli: Termites and Rigid Board Insulation 507

RIGID BOARD INSULATION IN SOUTH CAROLINA: ITS IMPACT ON DAMAGE, INSPECTION AND CONTROL OF TERMITES (ISOPTERA: RHINOTERMITIDAE)

B. C. SMITH AND P. A. ZUNGOLI Urban Entomology Research Laboratory Clemson University Clemson, SC 29634 0365

ABSTRACT A total of 225 pest control companies was randomly selected from the 315 members of the South Carolina Pest Control Association to determine their opinion on the impact of rigid board construction on termite control. Of those surveyed, 77% conducted inspections and/or treatments for subterranean termites (Isoptera: Rhinoter- mitidae, Reticulitermes and Coptotermes spp.) in South Carolina. The focus of survey questions was on demographics, occurrence of rigid board insulation, prevalence of rigid board insulation damage and structural damage, and treatment of structures when rigid board insulation was present. Results indicated termite infestations associated with rigid board insulation are not uncommon; 34% of the companies reported the presence of rigid board insulation on structures that have been treated or inspected for termites. Companies chose not to treat structures with rigid board insulation 43% of the time. Twenty-ﬁve percent (25%) of the companies reported damage to rigid board insulation due to insects other than termites. In addition, 12% of the respondents have been sued because of termite damage hidden by rigid board insulation. The primary recommendation to solve these problems is to remove a small gap of insulation, a vision strip, just above the soil level along the perimeter of a structure.

Key Words: Rigid board insulation, polystyrene, polyisocyanurate, Reticulitermes spp., Coptotermes spp.

RESUMEN Una muestra de 225 compañías de control de plagas fue aleatoriamente seleccio- nada de entre los 315 miembos de la Asociación de Control de Plagas de Carolina del

508 Florida Entomologist 78(3) September, 1995

Sur, para una encuesta en Febrero de 1994. De las compañías encuestadas que respondieron, el 77% se dedica a inspecciones o tratamientos para termites subterráneos (Isoptera: Rhinotermitidae, Reticulitermes y Coptotermes spp.) en Cariolina del Sur. El foco de las preguntas de la encuesta fue la demografía, aparición y prevalencia de daños en estructuras rígidas de aislamiento térmico, el daño estructural, y el tratamiento de las estructuras con aislamiento térmico rígido. Los resultados indicaron que las infestaciones de termites asociadas con estructuras rígidas de aislamiento tér- mico son comunes; el 34% de las compañías reportó la presencia de aislamiento tér- mico rígido en las estructuras que habían sido tratadas o inspeccionadas. Las compañías decidieron no tratar estructuras con aislamiento térmico rígido en el 43% de los casos. El 25% de las compañías reportó daños a estructuras rígidas de aislamiento térmico debido a otros insectos. En adición, el 12% de los que respondieron la encuesta había sido demandado debido a daños de termites ocultos por las estructuras rígidas de aislamiento térmico. La recomendación primaria para resolver estos problemas es quitar un pequeño pedazo del aislamiento, una banda de visión, justamente por encima del nivel del suelo a lo largo del perímetro de la estructura.

In the last few years, there has been an increasing awareness of the use of rigid board insulation to increase the energy efficiency of structures. Great concern has been expressed by Pest Control Operators (PCOs) with regard to their ability to prop- erly inspect and treat these structures for subterranean termites (Reticulitermes spp. and Coptotermes spp.), and other insects to the extent that the National Pest Control Association has issued guidelines for its members (Kramer 1993). Problems with termite treatments in the presence of rigid board insulation have also recently gained attention in the literature (Guyette 1994). Rigid board insulation, primarily foam board insulation, includes materials such as polystyrene (styrofoam) and polyisocyanurate. Rigid board insulation is used as a perimeter insulation around structures; the insulation is usually placed against the foundation walls on the exterior of the masonry structure hidden from view by skin coats such as wood, aluminum or vinyl siding, or stucco. When rigid board insulation is in direct contact with the ground, it is essentially the same as wood-to-ground contact. The presence of these materials creates a situation which is conducive to termite infestations because of continuous hidden pathways directly into the wooden members of a structure. This makes it extremely difficult for the PCO to visually inspect and treat for termites. In addition, these materials are associated with an increase in moisture levels, which allows for highly favorable conditions to harbor termites. Rigid board is being used as insulation due to low costs and good energy saving qualities. In 1985, approximately 5% of the building stock contained rigid board insulation (Christian 1990). By 1992, 50% of new construction contained some rigid board insulation (Kramer 1993). Good thermal characteristics, as well as high strength-to-weight ratios, have increased the use of these materials (Strzepek 1990). However, the strength and hardness of the rigid board may also be adversely affected by termite excretions (Building Research Station 1966). Hickin (1971) reported that plastic materials do not serve as a source of nutrition for termites, and damage of these materials is from termites physically removing the rigid board from their path. Plastics and other materials are not passed through the digestive system, rather the bitten-off fragments are removed or incorporated into tunnels. Polystyrene is subject to severe termite damage (Gay & Wetherly 1969). Tests showed that exposed ends and edges are more prone to attack than flat or curved surfaces. Cracks, rough edges, holes, and other deviations may promote termite activity on plastic materials (Becker

Smith & Zungoli: Termites and Rigid Board Insulation 509

1972). The Formosan subterranean termite, Coptotermes formosanus Shiraki, damaged soft styrene resins in building materials (Shimizu et al. 1970). The purpose of this survey was to demonstrate problems associated with the presence of rigid board insulation in South Carolina, primarily with termites, but also other insects. Surveys were sent to owners, managers, PCOs, and other persons associated with pest control companies to gain their perspective on the extent to which the problem exists. Recommendations to solve these problems, compiled from survey responses, will be discussed.

MATERIALS AND METHODS A sample of 225 individuals was chosen from the 315 pest control companies be- longing to the South Carolina Pest Control Association (SCPCA); a total of 775 companies are operating in South Carolina. Survey forms were mailed in February, 1994, in a timed and organized mailing strategy and numbered to facilitate identiﬁcation of returned surveys. The survey requested information for 1993 only. One week following the initial mailing, a follow-up letter was sent as a reminder to complete and to return survey forms. Three weeks following the initial mailing, a repeat survey form was mailed to all non-respondents. At eight weeks, another survey form was mailed to all non-respondents. All analyses of the data were performed using SAS (SAS In- stitute 1985). Results are primarily descriptive statistics. The content of questions consisted of four types: demographics, occurrence of rigid board insulation, prevalence of rigid board insulation damage and structural damage, and treatment of structures when rigid board insulation was present.

RESULTS AND DISCUSSION

Demographics Of 225 survey forms, 187 were returned (83%); of those, 173 (77%) represented respondents who conduct termite inspections and/or treatments. Respondents were stratiﬁed by three regions within the state; Piedmont (the upper northwest portion), Central (middle one-third portion), and Coastal (lower southeastern portion) (Fig. 1). These regions were selected based on climatic differences within the state. Of those respondents who performed termite inspections and/or treatments, 17% were in the Piedmont, 38% in the Central, and 40% in the Coastal regions, with 5% in combinations of the three regions. Most respondents (73%) were owners of the pest control companies surveyed, suggesting that information obtained was reliable. The number of termite treatments and inspections performed, in addition to the number of technicians employed, can be found in Table 1. Most companies were relatively small with less than 5 termite technicians employed. However, this does not imply the operations were small, as evidenced from the number of inspections and treatments performed in a given year: over three-quarters (82%) of all the termite treatments were performed by companies with 5 or fewer technicians.

Occurrence of Rigid Board Insulation The documented presence of rigid board insulation on structures treated or inspected for termites was reported by 34% of the companies. Of these companies, 17% reported rigid board insulation on structures in the Piedmont, 39% in the Central, and 39% in the Coastal regions, with 5% in combinations of the three regions. These

510 Florida Entomologist 78(3) September, 1995

Figure 1. Regions of South Carolina. values might not necessarily indicate the amount of new housing construction in each region, but certainly might reﬂect the increased termite pressures farther south. As many as 81% of companies reported up to 10% of structures they treated and/ or inspected for termites had rigid board insulation, and 2% of the companies reported greater than 75% of the structures had rigid board insulation. Rigid board insulation was reported to be installed above grade on structures by 41% of the companies, and below grade on structures by 59% of the companies.

Prevalence of Rigid Board Insulation Damage and Structural Damage

Eighty-one percent of the companies reported termite damage to rigid board insulation installed above grade in up to 10% of the structures they inspected. Seventy-two percent of the companies reported termite damage to rigid board insulation installed below grade in up to 10% of the structures they inspected, and 9% of the companies reported termite damage to rigid board insulation installed below grade in greater than 75% of the structures (Table 3). Termite damage to the wooden structural members in the presence of rigid board insulation installed both above and below grade was reported by the occurrence on structures and percent time seen by companies (Table 4). Excessive termite damage to rigid board insulation was seen by 10% of the companies, whereas moderate and slight damage to insulation was seen by 40% and 50% of the companies, respectively. Excessive termite damage to the wooden members of a

Smith & Zungoli: Termites and Rigid Board Insulation 511

TABLE 1. PERCENTAGE OF TREATMENTS, INSPECTIONS, AND TECHNICIANS REPORTED BY RESPONDENTS AS ASSESSED BY A MAIL SURVEY OF MEMBER COMPANIES OF THE SOUTH CAROLINA PEST CONTROL ASSOCIATION, APRIL 1994, N=225.

Treatments Inspections Technicians

Number % Number % Number %

1-50 35 1-50 5 1-5 82 51-100 15 51-100 10 6-10 10 101-150 10 101-150 7 11-15 3 151-200 5 151-200 10 16-20 0 201-250 5 201-250 7 21-25 0 251-300 5 251-300 8 >25 5 >300 25 >300 53 structure was seen by 30% of the companies; moderate and slight damage was seen by 40% and 30% of the companies, respectively. The amounts of termite damage to rigid board insulation and wooden structural members, at various locations on a structure, was reported by mean percent damage (Table 5). The amount of damage in wooden members is positively correlated to the amount of damage seen in rigid board insulation in most locations throughout a structure (Table 6). When the following foundation types were encountered, termite damage to rigid board insulation was present in the following amounts: monolithic slab (33%), supported slab (9%), floating slab (19%), basement (7%), crawlspace (23%), and any combination of one or more of the previous types (9%) (Table 7). Twenty-five percent of the companies reported damage to rigid board insulation caused by insects other than termites. Carpenter ants were responsible for 87% of the damage, while other insects such as wasps (Sphecidae) and dermestids (Dermestidae) were to blame for the remaining 13% damage. These figures indicate that rigid board insulation is subject to damage on a wider scope than previously thought.

TABLE 2. PERCENTAGE OF TIME RIGID BOARD INSULATION DAMAGE DUE TO TERMITES WAS SEEN (ABOVE AND BELOW GRADE) BY RESPONDENTS, BY FREQUENCY OF OCCURRENCE ON STRUCTURES AS ASSESSED BY A MAIL SURVEY OF MEMBER COMPANIES OF THE SOUTH CAROLINA PEST CONTROL ASSOCIATION, APRIL 1994, N=225.

Above grade Below grade

Occurrence (%) % Time Seen Occurrence (%) % Time Seen

1-10 81 1-10 72 11-25 15 11-25 10.5 26-50 4 26-50 3.5 51-75 0 51-75 5 >75 0 >75 9

512 Florida Entomologist 78(3) September, 1995

TABLE 3. PERCENTAGE OF TIME STRUCTURAL DAMAGE DUE TO TERMITES WAS SEEN (ABOVE AND BELOW GRADE) BY RESPONDENTS, BY FREQUENCY OF OCCUR- RENCE ON STRUCTURES AS ASSESSED BY A MAIL SURVEY OF MEMBER COM- PANIES OF THE SOUTH CAROLINA PEST CONTROL ASSOCIATION, APRIL 1994, N=225.

Above grade Below grade

Occurrence (%) % Time Seen Occurrence (%) % Time Seen

1-10 76 1-10 74 11-25 12 11-25 15 26-50 8 26-50 4 51-75 0 51-75 0 >75 4 >75 7

Treatment When Rigid Board Insulation Was Present

Pest control companies chose to treat structures with rigid board insulation 57% of the time, compared to 43% that did not. Of those respondents that chose to treat termite-infested structures, 24% treated as they would any other structure and provided a guarantee, 33% treated but did not provide a guarantee, 31% removed insulation below grade (Table 7), and another 13% did something other than those options mentioned. Responses for the latter category were: trenched out below the insulation, treated, and backfilled; cut insulation off above grade and then sealed bottom of insulation; recommended that a qualified builder replace insulation and check structure for soundness; pretreated but did not know rigid board insulation was later installed; only treated if insulation was not below grade; modified guarantee such that company was not responsible for damage as a result of hidden entry through insulation; treated as result of having won a bid. Of those respondents that chose not to treat structures, 34% believed the insulation created a situation conducive to termite infestation that was not treatable or correctable, 45% would not guarantee the treatment (Table 7), another 21% did something other than those options mentioned. Responses to the latter category in-

TABLE 4. MEAN PERCENT INSULATION AND STRUCTURAL DAMAGE DUE TO TERMITES AS ASSESSED BY A MAIL SURVEY OF MEMBER COMPANIES OF THE SOUTH CAROLINA PEST CONTROL ASSOCIATION, APRIL 1994, N=225.

Insulation Structural

Location % Std. Dev. Location % Std. Dev.

Between grade level 42.25 26.21 Between grade level 9.20 4.82 and footings and footings Just below grade 9.50 5.23 Just below grade 12.75 5.99 At grade level 14.85 7.08 At grade level 20.13 18.11 Between grade level 22.15 20.16 Between grade level 28.61 23.65 and sill plate and sill plate Above the sill 11.25 6.35 Above the sill 29.31 14.21 Smith & Zungoli: Termites and Rigid Board Insulation 513

TABLE 5. CORRELATION BETWEEN PERCENT OCCURRENCE OF INSULATION AND STRUC- TURAL DAMAGE DUE TO TERMITES, USING PEARSON CORRELATION COEFFI- CIENTS (α=0.05) AS ASSESSED BY A MAIL SURVEY OF MEMBER COMPANIES OF THE SOUTH CAROLINA PEST CONTROL ASSOCIATION, APRIL 1994, N=225.

Location Correlation Coefﬁcient p-Value

Between grade level and footings 0.2443 0.5598 Just below grade 0.7657 0.0037 At grade level 0.4725 0.1030 Between grade level and sill plate 0.6832 0.0006 Above the sill plate 0.5244 0.0658

cluded: customer did not want insulation removed; gave job to another company; rec- ognized situation due to previous problems experienced with this type of insulation. Trends show that PCOs generally do one of two things once rigid board is detected on a structure to be treated or inspected. They remove perimeter insulation just above the soil level in order to visually inspect for termites, or refuse to guarantee the treatment or treat the structure at all. Since only 25% of companies treated structures as they would any other, this suggests that the PCO is aware of the problem, but education is still necessary to inform those who do not know. This survey will be a vital part of the process to inform PCOs. Although the problem is evident in South Carolina, it is not unique to this state alone. With the ever increasing government regulations on new housing, the problem will continue to grow. The real problem lies in the fact that rigid board is placed in a continuous fashion which does not allow for a proper inspection to take place. Insulation may be, and usually is, placed in contact with the soil and is often installed below grade, even extended down to the footings of a structure. The Model Energy Code (MEC) states the following for slab-on-grade ﬂoors, i.e. monolithic slab construction (abundantly present in South Carolina): “In climates below 6,000 annual Fahrenheit heating degree days (HDD), the insulation shall extend downward from the top of the slab for a minimum distance of 24 inches . . . greater than 6,000 annual Fahrenheit heating degree days (HDD), the insulation shall extend downward from the top of the slab for a minimum of 48 inches . . .” (Council of American Building Ofﬁcials 1992). The MEC is essentially the standard regulation set and used for most building codes throughout

TABLE 6. PERCENT INSULATION DAMAGE DUE TO TERMITES, BY FOUNDATION TYPE AS ASSESSED BY A MAIL SURVEY OF MEMBER COMPANIES OF THE SOUTH CARO- LINA PEST CONTROL ASSOCIATION, APRIL 1994, N=225.

Foundation % Std. Dev.

Monolithic slab 33.33 40.81 Supported slab 9.17 1.82 Floating slab 18.45 19.20 Basement 7.05 1.14 Crawlspace 22.62 19.49 Combination of above types 9.38 2.18 514 Florida Entomologist 78(3) September, 1995

TABLE 7. TREATMENT OPTIONS AND THE PERCENTAGE OF COMPANIES THAT CHOSE TO TREAT STRUCTURES VERSUS NOT TREAT STRUCTURES AS ASSESSED BY A MAIL SURVEY OF MEMBER COMPANIES OF THE SOUTH CAROLINA PEST CON- TROL ASSOCIATION, APRIL 1994, N=225.

Chose to treat % Chose not to treat %

Treated as they would any 24 Insulation created a situation 34 other structure and provided a conducive to termite infestation guarantee that was not treatable or correctable Treated but did not provide a Could not guarantee the treat- guarantee 33 ment 45 Removed insulation below Did something other than grade 31 above options 21 Did something other than 13 above options the country. Unfortunately, enforcement of codes in South Carolina is done on a vol- untary basis only, within each county. In either scenario, the government mandates that rigid board insulation be placed, not only in contact with the soil, but as deep as over 1 m below grade level. We believe that the physical removal of the rigid board just above grade level (a vision strip of 15 cm, for example) around the structure’s perimeter would allow the PCO to detect termite activity, including the usual first signs - shelter tubes. Admittedly, there will be a thermal break (thermal short-circuit) caused by removing rigid board insulation, but we feel that the amount of material the termites are removing (the thermal loss as a result of termites) as they damage rigid board and structural members would warrant measures deemed necessary to correct the problem. In addition to the monetary loss of rigid board, the monetary loss of structural wooden members certainly is great, but precise figures are unavailable. The benefits of removing a small gap of insulation above grade level along the perimeter of structures would certainly outweigh the costs in the long run; additional insulation can be placed elsewhere in the structure to recoup the thermal loss from this adjustment. Some rigid board manufacturers are investigating rigid board with the addition of inert chemical fillers such as disodium octaborate tetrahydrate (Kramer 1993). Al- though this could potentially exclude termites from damaging the materials, it would not necessarily deter termites from tunneling behind the materials where gaps in construction are present. Hidden termite entry once again would ensue. We believe one final note is rather disturbing. Twelve percent of the respondents have been involved in litigation due to termite damage hidden by rigid board insulation. The majority of these companies, 57%, were from the Coastal region. The companies that belong to the SCPCA are among the best in South Carolina. Past experience with many of these companies demonstrates they are dedicated to their work and, therefore, less likely to be sued. This last number alone serves to illustrate the point that there is, indeed, a problem requiring attention at many levels, not just from the PCO.

ACKNOWLEDGMENTS We wish to express thanks to DowElanco and the South Carolina Pest Control As- sociation for funding this project. The help of all those Pest Control Operators in Smith & Zungoli: Termites and Rigid Board Insulation 515

South Carolina who participated in this survey is greatly appreciated. Without their assistance this work would not have been possible. Gratitude goes out to Kevin Hathorne and Terry Pizzuto for helping prepare the survey forms to be mailed. In addition, thanks to Dr. Hoke Hill for statistical guidance. This is technical contribution No. 4006 of the South Carolina Agricultural Experiment Station.

REFERENCES CITED

BECKER, G. 1972. Problems of Testing Materials with Termites, pp. 249-255 in A. H. Walters and E. H. Hueck-Van der Plas [eds]. International Biodeterioration Symposium. Vol. 2. Applied Science Publishers Ltd., London. BUILDING RESEARCH STATION (Watford, UK), Overseas Division. 1966. Plastics in Building. Overseas Building Notes. 108: 1-9. CHRISTIAN, J. E. 1990. The Most Needed Building Foundations Research Products, pp. 655-662 in Insulation Materials, Testing, and Applications. American Soci- ety for Testing and Materials, Philadelphia. 759 pp. COUNCIL OF AMERICAN BUILDING OFFICIALS. 1992. Model Energy Code. Council of American Building Ofﬁcials, Leesburg, VA. 87 pp. GAY, F. J., AND A. H. WETHERLY. 1969. Termite Resistance of Plastics V. CSIRO, Can- berra. GUYETTE, J. E. 1994. Termites Targeting Foam Insulation. Pest Control. 62: 49, 50, 52. HICKIN, N. E. 1971. Termites: a world problem. Hutchinson & Co. Ltd., London. 233 pp. KRAMER, R. D. 1993. Foam Insulation and Termites. Service Letter ESPC 055075, Number 1284. National Pest Control Association, Inc., Dunn Loring, VA. 3 pp. SAS INSTITUTE. 1985. SAS User’s Guide: Statistics. SAS Institute, Cary, NC. 956 pp. SHIMIZU, K., Y. NAKASHIMA, AND A. SAKANOSHITA. 1970. Studies on termite damage to a synthetic high polymer. Bulletin of the Faculty of Agriculture, Miyazaki University. 17: 281-285. Author has only read English summary. STRZEPEK, W. R. 1990. Overview of Physical Properties of Cellular Thermal Insula- tions, pp. 121-140 in D. L. McElroy and J. F. Kimpﬂen [eds]. Insulation Mate- rials, Testing, and Applications. American Society for Testing and Materials, Philadelphia. 759 pp.

516 Florida Entomologist 78(3) September, 1995

DORMANT SEASON APPLICATION OF STEINERNEMA CARPOCAPSAE (RHABDITIDA: STEINERNEMATIDAE) AND HETERORHABDITIS SP. (RHABDITIDA: HETERORHABDITIDAE) ON ALMOND FOR CONTROL OF OVERWINTERING AMYELOIS TRANSITELLA AND ANARSIA LINEATELLA (LEPIDOPTERA: GELECHIIDAE)

FERNANDO AGUDELO-SILVA1, FRANK G. ZALOM2, ALBERT HOM1, AND LONNIE HENDRICKS3 Biosys, 1057 E. Meadow Circle Palo Alto, CA 94303

1Current address: FAS Technologies. 1550 Walnut St. #5. Berkeley, CA 94709

2Department of Entomology, University of California, Davis, CA 95616

3Cooperative Extension, 2145 W. Wardrobe, Merced, CA 95340

ABSTRACT Overwintering larval populations of Amyelois transitella in mummy almonds and Anarsia lineatella in hibernaculae on almond trees in California orchards were reduced by infectives of Steinernema carpocapsae strain All and a cold tolerant Heter- orhabditis species (HL81) applied in the dormant season. The effect of this reduction in terms of preventing economic damage the following year is not known. Although the population reduction was statistically signiﬁcant, it is unlikely the increase in mortality achieved would prove to be an economically viable alternative to current practices.

Key Words: Almond, Steinernema carpocapsae, Heterorhabditis sp., Amyelois transitella, Anarsia lineatella

RESUMEN Poblaciones hibernanates de larvas de Amyelois transitella en “almendras momi- ﬁcadas”, y poblaciones hibernanates de Anarsia lineatella en hibernáculos de almen- dros en California, fueron reducidas por estados infectivos de dos nemátodos entomopatógenos, Steinernema carpocapsae cepa A11 y una especie de Heterorhabdi- tis (HL81) tolerante a baja temperatura, aplicados durante la estación de dormancia. El efecto de esta reducción en términos de impedir el daño económico al año siguiente no es bien conocido. Aunque la reducción en la población fue estadísticamente signiﬁ- cativa, se duda que el incremento de la mortalidad logrado sea una alternativa econó- micamente viable para las prácticas presentes de control de estos insectos.

For many years entomopathogenic nematodes have been considered to be potential biological mortality agents of many economically important insect pests (Glaser et al. 1940, Forschler & Gardner 1991). Because their nutritional requirements were poorly known for some time, it was difﬁcult to produce adequate quantities of the nematodes for large scale ﬁeld experiments. Advances in the understanding of the nutritional requirements (Buecher & Hansen 1977) and the development of mass cultur-

Agudelo et al.: Nematodes Reduce Overwintering Larvae 517 ing of these nematodes (Bedding 1984, Friedman 1990) has made it possible to produce them in sufficient quantities for such research. Increased emphasis on field evaluation will result in a better assessment of these organisms as practical insect control agents. Entomopathogenic nematodes appear to have the greatest potential as control agents when directed against immature insect stages occupying secluded niches, such as split almonds (Lindegren et al. 1978, Lindegren et al. 1987), tree galleries (Deseo & Miller 1985), leaf sheaths and cabbage heads (Welch & Briand 1961), artichoke crowns (Bari & Kaya 1984) and soil crevices (Wright et al. 1987, Shanks & Agudelo- Silva 1990). The use of entomopathogenic nematodes against two important insect pests of Prunus spp. that occupy secluded niches as larvae during the dormant season has not been fully investigated. These insects are the navel orangeworm, Amyelois transitella (Walker) (Lepidoptera: Pyralidae) and the peach twig borer Anarsia lineatella (Zeller), Lepidoptera: Gelechiidae). Together, these insects are estimated to cause direct damage of more than $10 million to California almond growers alone, not including treatment costs. A. transitella is the most important insect pest of almonds. Larvae of the two summer generations invade the nuts following almond hullsplit and feed on the meats. Almond handlers accept up to 4% damaged meats, but they award bonuses to growers for lower damage (Klonsky et al. 1990). After harvest, some nuts remain on the trees where they continue to attract A. transitella. The nuts dry and become “mummies” which often contain A. transitella larvae. The overwintering larvae become adults the following spring and are the source of infestation in the subsequent crop. Cultural controls for A. transitella include removing these “mummies” and harvesting nuts as early as possible in the season to avoid peak moth flights. Almost half of the orchards (~ 100,000 hectares in a given year) receive in-season organophosphate sprays for A. transitella (Klonsky et al. 1990). While there is evidence that applications of beneficial nematodes can infect A. transitella larvae infesting newly split almonds in the summer (Agudelo-Silva et al. 1987, Lindegren et al. 1978), control of overwintering larvae has not been attempted. A. lineatella is an important pest of both almonds and stone fruits in California, and the summer generation causes direct damage to nut meats and fruit. The larvae overwinter in hibernaculae built in crotches of twigs or branches. The larvae emerge during February and early March (Zalom et al. 1992) and migrate to shoot tips where they burrow into them to complete larval development. Organophosphate insecticides are routinely applied during the dormant season for A. lineatella control, and these have been the recommended alternative to disruptive in-season sprays (Rice & Jones 1988, Zalom et al. 1991). Recently, the application of dormant sprays has been criti- cized for environmental reasons, and suitable alternatives or mitigating measures are being pursued (e.g. Barnett et al. 1993). Here we report results of field and laboratory trials employing two entomopathogenic nematode species applied during the dormant season for control of overwintering A. transitella and A. lineatella larvae in almonds. In the dormant season, pest populations are relatively low and concentrated on the mummy nuts.

MATERIALS AND METHODS

Nematodes

Steinernema carpocapsae strain All and a cold tolerant Heterorhabditis species (currently undescribed but hereafter referred to as HL81) were used in the tests. Het-

518 Florida Entomologist 78(3) September, 1995 erorhabditis sp. strain HL81 (isolated in Holland) has been shown to parasitize insects at 9°C and 10°C(Simmons & Van der Schaf 1986). Infective juveniles were artificially mass reared (Biosys unpubl.) and stored in moist foam at 5°C(S. carpocapsae strain All) or at 10°C (Heterorhabditis sp. strain HL81) until needed. Before each test application, the nematodes were examined under magnification to confirm that the survival rate was greater than 90 percent.

A. transitella

Trials were conducted in Le Grand, Merced Co., and Durham, Butte Co., Califor- nia, during February and December, 1987, respectively. The Le Grand test was conducted to determine if mortality of overwintering A. transitella could be increased by a dormant season S. carpocapsae application. Treatments were applied on 23 Febru- ary in a commercial orchard (var. “Le Grand”) which was heavily infested with overwintering A. transitella (33% mummies infested in a pretreatment survey). Six applications rates were evaluated: 0, 0.62, 1.25, 2.5, 5.0, 10.0 and 20.0 million nematodes per tree. Each experimental unit was a tree and treatments were replicated ﬁve times. A totally randomized design was utilized. Nematodes were applied suspended in water (10.2 liters per tree) using a handgun sprayer (Farmtec, Oakland, CA). Im- mediately after the application, ten mummies were collected from each treated tree, individually placed in plastic bags in ice chests containing ice and brought to the laboratory to be examined for presence of nematodes. Nematodes were extracted from the mummies following the method developed by Lindegren et al (1987). One week after the application, 50 mummies were randomly collected from each of the trees, cracked by hand and examined for A. transitella. Larvae were tallied as alive or dead. Dead larvae were dissected and examined microscopically to ascertain the presence of nematodes. The Durham test was conducted to compare the ability of both S. carpocapsae and Heterorhabditis sp. strain HL81 to infect A. transitella larvae at low temperature. Treatments were applied on 15 December in a commercial orchard (var. “Nonpareil”, “Ne-plus Ultra,” and “Mission”; 2:1:1 ratio). Treatments evaluated were S. carpocapsae, Heterorhabditis sp. strain HL81 and an untreated control. Mummies were randomly collected from trees and divided into 15 groups each of which contained 130 mummies. The mummies were treated such that there was certainty that all the mummies received nematodes. Each group of mummies was placed on a piece of ﬂat plastic laid out on the ground in the orchard and then arranged on the plastic touching each other, in a single layer. Twenty mummies were taken from each group and examined for A. transitella larvae. Infestation of these mummies was determined to be 59.5%. Each group was randomly assigned to one of the treatments, and the treatments were applied using a previously cali- brated spray bottle. Nematodes were suspended in water and applied at a rate of 51 nematodes per mummy (5,610 nematodes suspended in water were applied to each group of 110 mummies). Immediately after application (less than 15 minutes), 10 mummies were randomly chosen from each group, placed in individual bags and stored in ice chests to be brought to the laboratory for nematode extraction (Lindegren et al.1987). The remaining mummies (100 per group) were placed in individual wire mesh cages which were suspended from branches in the almond trees. Two weeks after the nematodes were applied, the mummies from each of the cages were placed in ice chests containing ice and brought to the laboratory to be examined for the presence of larvae. Dead larvae were dissected and microscopically examined to ascertain the presence of nematodes.

Agudelo et al.: Nematodes Reduce Overwintering Larvae 519

A. lineatella This trial was conducted in a 2-year old non-bearing almond orchard near Cortez, Merced Co., CA, to determine the efficacy of a dormant application of S. carpocapsae on overwintering A. lineatella larvae. The treated trees were infested with A. lineatella overwintering in hibernaculae on the branches. Treatments were applied on 20 January, 1987, with a handgun sprayer and consisted of S. carpocapsae at rates of 1.5 and 0.5 million per tree, a conventional dormant spray of diazinon (16.0 liter AI per ha) plus Volck supreme oil (48 liter per ha) and water as a check. The treatments were arranged in a completely randomized design with twenty single tree replicates, each treatment tree separated by an untreated buffer tree. No attempt was made to find dead larvae after the application of the treatments because the hibernacula are difficult to locate in significant quantity. It was decided that a more appropriate and practical evaluation of the efficacy of the nematodes on mortality of overwintering larvae would be to count shoot strikes in the spring. Shoot strikes can be used to evaluate larval control (Summers et al. 1959) since the overwintering larvae which emerge feed on new shoots and bore into them. The shoot tips which are killed as a result of this feeding are relatively easy to identify. Efficacy was evaluated on 27 March, 1987, by counting the total number of shoot tips damaged by A. lineatella larvae following emergence of the overwintering generation.

RESULTS AND DISCUSSION

A. transitella

In the Le Grand test, the nematode applications did not statistically increase the mortality of overwintering larvae. Although the mortality rates for the 0.62 and 5.00 million S. carpocapsae per tree application seemed to increase larval mortality, there was no signiﬁcant difference (P >0.05) between rates of application when data were subjected to one way ANOVA. Similarly, no signiﬁcant relationship was found (r2= 0.014; P >0.05) when dose/mortality data was subjected to regression analysis (Table 1). No nematodes were found inside the dead larvae. There was a positive relationship between the number of nematodes applied per tree and the number of nematodes delivered to the mummies (F= 66.3; P <0.001) (Ta- ble 1). The relationship between dose rate and number of mummies receiving nematodes was also positive, indicating that better coverage is obtained as the number of nematodes applied per tree increases. These are important results as they demonstrate the potential of delivering nematodes to the site of the overwintering larvae using commercially available spraying equipment. Larval mortality in this trial was much lower than that observed by Lindegren et al. (1978) and Agudelo-Silva et al. (1987) for applications at hullsplit (summer). Since there were no differences in larval mortality between treatments and no nematodes were found in the dead larvae, most likely the nematodes did not infect them. This may have been due, at least partially, to temperatures during the time of this study. The low temperature probably im- paired the ability of nematodes to move on the mummies to reach the larvae inside them, and even if a nematode ultimately reached a larva, the nematode and its symbiont bacteria may not have lethally infected it. Although the temperature reached 18°C for a few hours during the seven days after the nematodes were applied, most of the time the temperature was too low to allow nematode infection and development. During the seven days following application, temperatures did not exceed 10°C for 60% of the time (101 hours). Development and reproduction of S. carpocapsae occurs between 15°C and 27°C (Kaya 1977).

520 Florida Entomologist 78(3) September, 1995

TABLE 1. STEINERNEMA CARPOCAPSAE STRAIN ALL JUVENILES APPLIED TO ALMOND TREES FOR CONTROL OF AMYELOIS TRANSITELLA LARVAE OVERWINTERING IN MUMMY ALMONDS, 23 FEBRUARY, 1987 (N=5). LARVAL MORTALITY WAS EVALUATED 7 DAYS AFTER TREATMENT.

2 Mean Nematodes per Mummy % Mummies Number Nematodes per Percent Dead1 Mean %2 with Tree (× 106) Larvae ± SD ± SD Range Nematodes

0.00 2.5 (3.6) 0.0 (0) — 0 0.62 4.0 (6.9) 27.6 (35.6) 12-108 60 1.25 2.7 (6.0) 21.2 (10.0) 12-84 60 2.50 2.6 (2.4) 13.2 (9.8) 12-204 48 5.00 5.3 (11.0) 55.6 (34.6) 12-204 88 10.00 2.0 (2.6) 68.6 (61.1) 12-240 88 20.00 2.1 (3.3) 232.0 (138.0) 36-588 92

1Fifty mummy nuts per replicate. 2Ten mummy nuts per replicate.

In the Durham test, it was hoped that Heterorhabditis sp. strain HL81, a cold-tolerant nematode strain, would prove more effective controlling A. transitella during its overwintering period. Acceptable delivery of nematodes to the mummies was achieved with 90 and 82 percent of the mummies for S. carpocapsae and Heterorhabditis sp. respectively releasing nematodes in the extraction procedure (Lindegren et al 1987) (Table 2). Larval mortality of mummies treated with either nematode species was 2.2 times greater than mortality observed from mummies that were treated with water alone. This difference was statistically signiﬁcant (P < 0.05) and strongly suggests that applications of entomopathogenic nematodes can increase mortality of A. transitella in mummy nuts during the dormant period. No nematodes were found inside the dead larvae. Unfortunately, A. transitella survival was over 88% for both species. The fact that no nematodes were found inside the dead larvae is not surprising. Most likely, nematodes invaded larvae and killed them, but died before completing development and reproduction. The dead nematodes would have decayed before the dead A. transitella larvae were examined. The periods of low temperatures (< 10°C) that occurred between the time of nematode application and the time when the dead larvae were examined (when infection by the nematodes must have taken place) were not conducive for the successful growth of the nematode and its symbiont bacteria (Moly- neux, 1986). Successful growth of the symbiont is necessary for nematode development (Poinar & Thomas 1966). The signiﬁcant (P < 0.05) difference between the number of dead larvae found in the mummies treated with nematodes and those treated with water only suggests that the nematodes were responsible for the differences in mortality. It is possible that more cold-tolerant nematode strains than those applied in this trial could increase mortality. However, the increase in larval mortality that could be attributed to nematode treatment in this trial, although important biologically, would not be acceptable relative to conventional controls currently available.

A. lineatella The difference in the number of A. lineatella shoot strikes between the almond trees treated with S. carpocapsae during the dormant season (Table 3) at 1.5 million Agudelo et al.: Nematodes Reduce Overwintering Larvae 521

TABLE 2. STEINERNEMA CARPOCAPSAE STRAIN ALL AND HETERORHABDITIS SP. STRAIN (HL81) APPLIED AT A RATE OF 51 NEMATODES PER MUMMY FOR CONTROL OF AMYELOIS TRANSITELLA LARVAE, 15 DECEMBER 1987.

Mean Number Mean Percent Nematodes2 Range of Percent Dead Larvae1 per Mummy Nematodes per Mummies with Treatment ± SD ± SD Mummy Nematodes

S. carpocapsae 11.8 (4.1)a 26.8 (25.02) 0-126 90 Heterorhabditis sp. 11.6 (2.3)a 19.4 (26.51) 0-162 82 Water check 3.6 (2.8)b 0.0 (0.0) — —

1Means followed by the same letter are not significantly different (P > 0.05) when compared by Duncan’s (1995) new multiple range test. 2Ten mummies nuts per replicate. and 0.5 million per tree and the water check was statistically significant (P < 0.05), but the level of control (about 24%) was significantly (P < 0.05) less than the level of control (~ 93%) achieved with the conventional dormant treatment of diazinon plus oil (F=34.27; DF=3). These data suggest that S. carpocapsae have the ability to penetrate overwintering hibernaculae, and kill the larvae inside them, reducing their overwintering populations and subsequent damage to trees. The hourly temperatures during some of the days when the larvae were exposed to the nematodes was high enough (up to 23°C) to allow nematode movement and larval infection. However, the temperature was low (less than 10°C) for several days during the time that the larvae were exposed to the nematodes, and this probably diminished the growth, reproduction and infectivity of the nematodes (Kaya 1977). The ability of entomopathogenic nematodes to reduce overwintering A. transitella and A. lineatella larval populations is not resolved. It was promising that larval mortality was achieved with existing nematode strains under field conditions, and that nematodes could be reliably delivered to mummy almonds in commercial orchards (Le Grand test). However, larval mortality of both A. transitella and A. lineatella, while significant, was below commercially acceptable levels. It can be hypothesized that if a nematode were available that would remain active in larvae for extended periods of time below 10°C, the level of larval mortality achieved could be greater. It is also pos-

TABLE 3. STEINERNEMA CARPOCAPSAE STRAIN ALL APPLIED TO ALMOND TREES FOR CONTROL OF ANARSIA LINEATELLA LARVAE OVERWINTERING IN HIBERNACU- LAE, 20 JANUARY, 1987 (N=20). EFFECT OF TREATMENTS MEASURED AS NUMBER OF LARVAL SHOOT STRIKES PER TREE ON 27 MARCH, 1987.

Mean Shoot Strikes1 Treatment Rate per Tree ± SD

Water check — 28.4 (10.6)c S. carpocapsae 1.5 × 107 per tree 22.3 (9.8)b S. carpocapsae 0.5 × 107 per tree 21.4 (9.3)b Diazinon + oil 16 l AI per Ha. + 48 L per Ha. 2.1 (2.7)a

1Means followed by the same letter are not signiﬁcantly different (P > 0.05) when compared by Duncan’s (1955) new multiple range test. 522 Florida Entomologist 78(3) September, 1995 sible that treatments for both pest species could be delayed until early bloom when warmer conditions would be present, yet overwintering larvae would be present as the treatment target. In order to be commercially feasible at present, the level of larval mortality achieved must reach 70 to 90 percent, and the cost of applying nematodes must be competitive with conventional in-season sprays for A. transitella, ($24 to $40 per ha) and dormant sprays for A. lineatella ($20 to $56 per ha).

ACKNOWLEDGMENTS

This study was supported in part by the Almond Board of California. We especially thank B. Wilk and C. Kitayama, Scientiﬁc Methods, Chico, CA, for their valuable assistance.

REFERENCES CITED

AGUDELO-SILVA, F., J. E. LINDEGREN, AND K. A. VALERO. 1987. Persistence of Neoa- plectana carpocapsae (Kapow selection) infectives in almonds under field conditions. Florida Entomol. 70: 288-291. BARI, M. A., AND H. H. KAYA. 1984. Evaluation of the entomogenous nematode Neoa- plectana carpocapsae (Steinernema feltiae) Weiser (Rhabditida: Steinernema- tidae) and the bacterium Bacillus thuringiensis for suppression of the artichoke plume moth (Lepidoptera: Pterophoridae). J. Econ. Entomol. 1: 225- 229. BARNETT, W. W., J. EDSTROM, R. COVIELLO, AND F. G. ZALOM. 1993. Insect pathogen “Bt” controls peach twig borer on fruits and almonds. Calif. Agric, 47: 4-6. BEDDING, R. A. 1984. Large scale production, storage and transport of insect-parasitic nematodes Neoaplectana spp, and Heterorhabditis spp. Ann. Appl. Biol. 101: 117-120. BUECHER, E. J., AND E. L. HANSEN. 1977. Mass culture of axenic nematodes using continuous aeration. J. Nematol. 3: 199-200. DESEO, K. V., AND L. A. MILLER. 1985 Efficacy of entomogeneous nematodes Stein- ernema spp., against clearwing moths, Synanthedon spp., in north Italian apple orchards. Nematologica 31: 100-108. DUNCAN, D. B. 1955. Multiple range and multiple F tests. Biometrics 11: 4-6. FORSCHLER, B. T., AND W. A. GARDNER. 1991. Field efficacy and persistence of entomopathogenic nematodes in the management of white grubs (Coleoptera: Scar- abeidae) in turf and pasture. J. Econ. Entomol. 84: 1454-1459. FRIEDMAN, M. J. 1990. Commercial production and development, pp. 153-172 in R. Gaugler and H. K. Kaya, [eds.] Entomopathogenic Nematodes in Biological Control. CRC Press. Boca Raton. GLASER, R. W., E. E. MCCOY, AND H. B. GIRTH. 1940. The biology and economic importance of a nematode parasitic in insects. J. Parasitol. 26: 479-495. KAYA, H. K. 1977. Development of the DD-136 strain of Neoaplectana carpocapsae at constant temperatures. J. Nematol. 9: 346-349. KLONSKY, K., F. G. ZALOM, AND W. W. BARNET. 1990. California’s almond IPM program. California Agric. 44: 21-24. LINDEGREN, J. E., C. E. CURTIS, AND G. O. POINAR. 1978. Parasitic nematodes seeks out navel orangeworm in almond orchards. California Agric. 32: 10-11. LINDEGREN, J. E., F. AGUDELO-SILVA, K. A. VALERO, AND C. E. CURTIS. 1987. Compar- ative small-scale field application of Steinernema feltiae for navel orangeworm control. J. Nematol. 19: 503-504. MOLYNEAUX, A. S. 1986. Heterorhabditis spp. and Steinernema (=Neoaplectana) spp: temperature, and aspects of behavior and infectivity. Exp. Parasitol. 62: 169- 180. POINAR, G. O. JR., AND G. M. THOMAS. 1966. Significance of Achromobacter nematophilus Poinar and Thomas (Achromobacter nematophilus: Eubacteriales) in the Agudelo et al.: Nematodes Reduce Overwintering Larvae 523

development of the nematode DD-136 (Neoaplectana: Steinernematidae). Par- asitol. 56: 385-390. RICE, R. E., AND R. A. JONES. 1988. Timing post-bloom sprays for peach twig borer (Lepidoptera; Gelechiidae) and San Jose Scale (Homoptera: Diaspididae). J. Econ. Entomol. 81: 293-299. SHANKS, C. H., AND F. AGUDELO-SILVA. 1990. Field pathogenicity of heterorhabiditid and steinernematid nematodes (Nematoda) infecting black vine weevil larvae (Coleoptera: Curculionidae) in cranberry bogs. J. Econ. Entomol. 83: 107-110. SIMMONS, W. R., AND D. A. VAN DER SCHAFF. 1986. Infectivity of three Heterorhabdi- tis isolates for Otiorhynchus sulcatus at different temperatures, pp. 285-289 in R. A. Samson, J. M. Vlak, and D. Peters., [eds.]. Fundamental and Applied As- pects of Invertebrate Pathology Foundation of the Fourth International Collo- quium of Invertebrate Pathology. Wageningen. The Netherlands. SUMMERS, F. M., W. W. BARNETT, R. E. RICE, AND C. V. WEEKLY. 1995. Control of pear twig borer on almonds and peaches. California J. Econ. Entomol. 52(4): 637- 639. WELCH, H. E., AND L. J. BRIAND. 1961. Field experiment on the use of a nematode for the control of vegetable crop insects. Proc. Entomol. Soc. Ontario. 91: 197-202. WRIGHT, R. J., F. AGUDELO-SILVA, AND R. GEORGIS. 1987. Soil applications of Stein- ermatid and Heterorhabditid nematodes for control of Colorado Potato Beetle, Leptinotarsa decemlineata (Say). J. Nematol. 19: 201-206. ZALOM, F. G., R. A. VAN STEENWYK, W. J. BENTLEY, R. COVIELLO, R. E. RICE, W. W. BARNETT, M. A. HOY, M. M. BARNES, B. A. TEVIOTDALE, W. D. GUBLER, AND M. V. MCKENRY. 1991. Almond pest management guidelines. Univ. of California PMG Publ. 1. 22pp. ZALOM, F. G., W. W. BARNETT, R. E. RICE, AND C. V. WEAKLEY. 1992. Factors associated with ﬂight patterns of the peach twig borer (Lepidoptera: Geleichiidae) observed using pheromone traps. J. Econ. Entomol. 85: 1904-1909.

♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦

Landolt: Sweet Baits for Moths 523

ATTRACTION OF MOCIS LATIPES (LEPIDOPTERA: NOCTUIDAE) TO SWEET BAITS IN TRAPS

PETER J. LANDOLT U.S. Department of Agriculture, Agricultural Research Service P. O. Box 14565, Gainesville, FL 32604

ABSTRACT Traps baited with solutions of molasses or jaggery (unrefined palm sugar) captured significant numbers of the moth Mocis latipes Guenee, indicating their attraction to these baits. Numbers of moths captured were affected by bait concentration and by bait age. Greatest moth captures were obtained with 20% molasses in water or 5, 10 or 20% jaggery in water. Molasses and jaggery baits aged in the laboratory for up to three days before field testing were more attractive than freshly-made baits to M. latipes moths. This is the first demonstration of M. latipes attraction to sugar- based baits. Key Words: Insecta, moth, sugar, fermentation, microbial.

RESUMEN Las trampas cebadas con soluciones de melaza o azúcar de palma sin reﬁnar cap- turaron cantidades signiﬁcativas de la polilla Mocis latipes Guenee, indicando su

524 Florida Entomologist 78(3) September, 1995 atracción a esos cebos. El número de polillas capturadas fue afectado por la concen- tración y edad de los cebos. Las mayores capturas de polillas fueron obtenidas con 20% de melaza en agua o 5, 10 o 20% de miel de palma en agua. La atraccción de los cebos de melaza y azúcar de plama envejecidos hasta tres días en el laboratorio antes de probarse en el campo fue superior a la de los cebos frescos. Esta es la primera demos- tración de la atracción de M. latipes hacia cebos azucarados.

Sweet baits (sugar baits) are widely used by collectors of Lepidoptera to attract moths and butterflies to a site where they may be captured, as described by Holland (1903) and Sargent (1976). The ingredients used, although varied, include sugar and sugar-rich materials, as well as fruits and alcoholic beverages. It is thought that the attraction and feeding of moths on artificial sugar baits is an indication of their tendency to feed on natural sources of sugars, such as rotting fruit, tree sap, insect honeydew, and flower nectars (Norris 1935). There are no direct comparisons of formulations of sweet baits or quantitative efforts to optimize bait effectiveness in attracting Lepidoptera, although they appear to lure a wide variety of moths and butterflies (Sargent 1976, Norris 1935 and references therein). Although such baits are commonly used to collect non-pest species for taxonomic or life history studies, there are indications that they may be attractive to a number of important pest species of moths as well. Norris (1935) supposed that nearly all tem- perate Noctuidae (which would necessarily include many pests) are attracted to sugar baits, but did not provide documentation. Frost (1928) captured over 48,000 noctuid moths in 2 years of trapping tests in peach orchards using pails of molasses or syrup solutions, but did not report the identity of the species captured. Molasses and sugar syrups have been evaluated as a bait placed in pails for attracting and killing Gra- pholita molesta (Busck) moths (Frost 1926, 1928, 1929) and Cydia pomonella (L.) moths (Eyer 1931) in fruit orchards. Similarly, pans of poisoned vinegar and molasses were reported to attract and kill corn earworm moths, Helicoverpa zea (Boddie) (Glover 1855, cited by Ditman & Cory 1933). Similar materials may provide sources of useful attractants for monitoring pest populations or developing attracticidal approaches to suppress pest populations. This paper reports the attractiveness of selected sweet baits to the moth, Mocis latipes Guenee, as evidenced by captures in baited traps. This species is a pest of pasture grasses, sugar cane and sorghum in the southeastern United States and throughout much of the neotropics (Ogunwolo & Habeck 1975; Strayer 1975). A sex pheromone of this species is known to be attractive to males (Landolt & Heath 1989), but there are no methods available for attracting and trapping females. Their attraction to any type of food bait would provide the basis for development of new lures for monitoring and, possibly, controlling populations. The capture of M. latipes in traps containing sweetened food baits was previously observed by the author in tests to evaluate food-baited traps for tephritid fruit flies. The objectives of this study were to determine if these moths could be trapped with sugar-based baits and to evaluate effects of concentration and age of selected baits on numbers of moths attracted to, and captured in, a trap. Information on the optimum concentration and age of baits would be of use in future attempts to isolate and identify attractive volatile chemicals ema- nating from these baits. Also, the determination of an aging effect on bait effectiveness would support previous suggestions that microbial fermentation is a crucial component of a good sugar bait for Lepidoptera (Norris 1935).

Landolt: Sweet Baits for Moths 525

MATERIALS AND METHODS

Four trapping tests were conducted to determine the attractiveness of baits to the moths. These were: 1) an initial comparison of moths captured with five different baits and a control, 2) an assessment of the effect of bait age on moths captured, 3) an assessment of effects of bait concentration on moths captured, and 4) a comparison of the effectiveness of two optimized baits. Invaginated glass traps, McPhail traps (Newell 1936), were used for all tests. These traps, used for monitoring tephritid fruit flies, hold about 0.5 liters of fluid and have a 6.5-cm diam opening at the trap bottom for insects to enter. All baits were prepared and tested in 200-ml amounts per trap. Traps were hung on wire loops mounted on wooden stakes, with the trap opening above the tallest vegetation (about 50 cm above ground). This height was selected because M. latipes were observed in early morning flight just above the top of the grass and weed canopy. A randomized complete block design was used for the first three tests, with blocks at four different sites. Traps within a block were 10 m apart. All trap sites were in grassy areas along roads bordering crops (soybean, peanut, sugar cane). All captured moths were sorted and counted by species and sex. However, numbers suitable for statistical analyses were obtained only for M. latipes, which was abundant throughout the study period (late August to early November).

Bait Comparison

The first experiment was a preliminary comparison of six different baits. The treatment selection was based partly on unreported observations of moths in traps baited for fruit flies. Bait treatments were 1) de-ionized water; 2) fruit pectin (Sure- Jell, Kraft General Foods, White Plains, NY); 3) brewer’s yeast (Fleischmann’s Yeast, Specialty Brands, San Francisco, CA) and sucrose; 4) honey; 5) molasses (Grandma’s Molasses, Mott’s, Stanford, CT); and 6) jaggery or unrefined palm sugar (Indian Kol- hapur Jaggery, House of Spices Inc., Jackson Heights, N.Y.). For treatments 2, 4, 5, and 6, 20 g of material was mixed in 200 ml of de-ionized water for each trap. For treatment three, 20 g of purified cane sugar and one packet of yeast (7 g) were added to 200 ml of water. All baits were prepared on the day they were used in traps. The experiment was set up as three blocks of traps; each block was a set of 6 traps, each trap containing a different bait. Traps were removed, emptied, cleaned, and rebaited after 48 hours. A total of 11 block replicates was accumulated over four different baiting times (three block sites by four baiting times, with one block missing due to trap breakage). Treatments were randomized within each block.

Effects of Bait Age

The second experiment compared numbers of moths trapped in either 10% molasses or 10% jaggery solutions that were aged before placement in the field. Multiple bait samples (200 ml) were prepared 1, 2, 3, 5, and 7 days before the beginning of a trapping experiment and held in the laboratory at 23 ± 0.5°C in 500-ml glass jars covered with paper toweling. Bait solutions were placed in traps in the field in the afternoon, and traps were removed and emptied the following morning. Trap blocks containing one of each of the five treatment ages were set up at four different sites. For each material this test was conducted on three different days providing 12 block replicates. Treatments were randomized within all blocks.

526 Florida Entomologist 78(3) September, 1995

Effects of Dose

The third experiment compared the effectiveness of a range of concentrations of molasses and jaggery on numbers of moths captured. Concentrations tested were 2.5, 5, 10, 20, and 40 g of material per 200 ml water per trap. Baits were prepared two days before traps were baited and stored in the laboratory in 500-ml glass jars covered with paper toweling. Traps were baited in the afternoon and removed and emptied the following morning. Trap blocks (ﬁve traps) were set up at four sites, with all treatments randomized within each block. For each material, treatment comparisons were conducted on three different days, providing 12 block replicates (four sites, three days).

Comparison of Molasses and Jaggery

Solutions of 10% jaggery and 20% molasses were compared directly in McPhail traps for their effectiveness as trap baits for M. latipes. Baits were prepared two days before placement in traps in the field and stored in the laboratory in 500-ml glass jars covered with paper toweling. Traps containing either molasses or jaggery were paired in the field; 10 pairs of traps were placed in fields in the afternoon, then removed and emptied the following morning. Data for captures of M. latipes in traps from multiple comparison tests (tests 1, 2, and 3) were transformed to percentages of block totals, followed by an arcsin transformation before using an analysis of variance (Steel & Torrie 1960). Significance of differences between treatment mean percents was determined using Duncan’s multiple range test (Duncan 1955) when ANOVA indicated a significant F value. Means for

Figure 1. Mean (± SE) percentages of block totals for numbers of M. latipes moths captured in traps baited with water, or aqueous solutions of 10% pectin, 10% honey, 10% sugar and 3.5% brewer’s yeast, 10% molasses, or 10% jaggery, with a total of 177 trapped. Percentages sharing the same letter are not signiﬁcantly different by Dun- can’s multiple range test at P > 0.05.

Landolt: Sweet Baits for Moths 527 data comparing the attractiveness of molasses and jaggery were tested for signiﬁ- cance, without transformation, using Student’s t-test (Steel & Torrie 1960). For all experiments, numbers of males and females of M. latipes were pooled for analyses of data.

RESULTS

In the first experiment comparing different bait materials, significantly more M. latipes were captured in traps baited with yeast and sugar, molasses, and jaggery than in traps baited with water, pectin, or honey (F = 12.5, df = 5,60, P < 10-6) (Fig. 1). Greatest numbers of M. latipes were captured in traps containing 10% jaggery, compared to molasses, yeast and sugar, honey, pectin, or water. (Fig. 1). Significantly more moths were caught in traps baited with 10% molasses than were captured in traps baited with honey (Fig. 1). None were captured in traps baited with pectin or with water. For the second experiment, bait age significantly affected captures of M. latipes in traps baited with either molasses (F = 3.3, df = 4,45, P=0.02) or jaggery (F = 2.6, df = 4,49, P = 0.05) (Fig. 2). Captures increased significantly from the first to the third day with both materials, followed by an apparent reduction in effectiveness, although this was not significant at the 5% level. In the third experiment, captures of M. latipes in traps baited with either molasses or jaggery were significantly affected by concentration (F = 15.1, df = 4,50, P < 10-6)

Figure 2. Mean (± SE) percentages of block totals for numbers of M. latipes moths captured in traps baited with either 10% molasses (open bars) or 10% jaggery (cross- hatched bars) baits of different ages, with a total of 653 trapped. Within a bait category (molasses or jaggery) mean percentages sharing a letter are not signiﬁcantly different by Duncan’s multiple range test at P > 0.05.

528 Florida Entomologist 78(3) September, 1995

Figure 3. Mean (± SE) percentages of block totals for numbers of M. latipes moths captured in traps baited with either three-day-old dosages of molasses or three-day- old dosages of jaggery, with a total of 1,099 trapped. Within a bait category, mean percentages that share a letter are not signiﬁcantly different by Duncan’s multiple range test at P > 0.05.

(Fig. 3). Numbers of moths in traps baited with molasses increased with each increase in concentration and were signiﬁcantly greater at 20% compared to all other tested concentrations. Numbers of moths captured in traps baited with jaggery increased with concentration increases up to 5%, with no differences in trap catches among 5, 10, and 20% solutions. For the fourth experiment, traps baited with three-day-old 10% jaggery captured signiﬁcantly more M. latipes than traps baited with three-day-old 20% molasses (t = 2.97, df = 9, P = 0.016, Mean ± SE = 8.8 ± 1.9 moths per trap per night for jaggery; Mean ± SE = 4.9 ± 1.1 moths per trap per night for molasses). Although counts of male and female M. latipes were combined for statistical comparisons of bait effectiveness, total numbers of males and females trapped indicate an overall sex ratio of near 1 to 1, with 1,027 males and 1,039 females captured in all of the traps baited with molasses and jaggery for experiments 1 through 4 (Table 1). Other pest species of noctuids were also captured in these tests (Table 1), including Anticarsia gemmatalis Hubner, Mocis disseverans (Walker), and Spodoptera frugiperda (J. E. Smith). Three species of Leucania were captured but were not determined to species.

DISCUSSION The results of the experiments reported here clearly indicate that M. latipes are attracted to sweet baits, such as molasses or jaggery. Because the sex ratio for M. latipes trapped with either of these two materials was near one to one, it may be possible

Landolt: Sweet Baits for Moths 529

TABLE 1. SPECIES OF NOCTUID MOTHS CAPTURED IN TRAPS BAITED WITH UNREFINED PALM SUGAR (JAGGERY) OR MOLASSES IN SEPT./OCT. 1993. GAINESVILLE, FLORIDA.

Species Jaggery Molasses

Females Males Females Males

Mocis latipes (Guenee) 554 528 485 499 Anticarsia gemmatalis Hubner 60 106 90 177 Mocis disseverans (Walker) 26 36 36 43 Anomis erosa Hubner 30 7 15 4 Pseudaletia unipuncta (Haworth) 8753 Spodoptera frugiperda (J. E. Smith) 16 8 14 12 Spodoptera exigua (Hubner) 3020 Spodoptera ornithogalli (Guenee) 1652 Spodoptera latifascia Walker 0020 Leucania sp. 15 17 16 17 Agrotis subterranea F. 3130 Agrotis ipsilon (Hufnagel) 1010 Anicla infecta Ochsenheimer 12 8 5 8 Helicoverpa zea (Boddie) 4011

to develop attractants based on these baits for monitoring the presence of females in affected crops. Numbers of other noctuid moths trapped indicate there is some potential to develop attractants and traps based on sugar baits for other pest species of moths. It is assumed that the relatively lower numbers of these other species trapped is due to their lower populations at sites during the trapping period. The location and season for these tests were selected to maximize the presence of M. latipes. It will be necessary to target populations of other pest species of interest to more adequately determine their attraction to these baits. It is unclear what ingredients are critical to the attractiveness of the tested baits for this and other species of moths. Sargent (1976) indicated that sugar (presumably sucrose, as refined cane or beet sugar) is the only ingredient required, but also added beer to attract Catocala species. Holland (1903) described using a solution of sugar, beer and rum. Experiments to control species of Grapholita and H. zea were conducted with molasses or sugar syrups (Frost 1926, 1928, 1929; Eyer 1931; Ditman & Cory 1933). The results reported here for M. latipes indicate that honey or a mixture of sugar and brewer’s yeast were not effective and that additional factors present in the molasses and jaggery are important to moth attraction. Attractiveness of sugar baits to moths is assumed to be due in part to fermentation processes necessary for the production of attractants (Norris 1935; Sargent 1976). This assumption is supported here by the significant increase in numbers of moths captured in three-day-old traps. If this effect is due to the proliferation of microbes in the baits, the differences in response among treatments observed in the first experiment could, to some extent, reflect their suitability as a broth or culture medium for a particular type of microbial agent. 530 Florida Entomologist 78(3) September, 1995

Additional research is needed to isolate and identify attractive odorants emanat- ing from these baits when placed in traps. The effectiveness of this system is limited by the aging effects on bait attractiveness, by the small capacity of the trap for captured moths, and by the fragile nature of the trap design. Ideally, a formulated lure of identiﬁed food attractants could be used to bait a dry trap design.

ACKNOWLEDGMENTS

Technical assistance was provided by K. M. Davis. Traps were generously provided by T. Howler (USDA, APHIS, Gainesville, FL) and J. Sivinski (USDA, ARS, Gaines- ville, FL.). Jaggery was recommended as a sweet bait for insects by S. Kulkarny, Gainesville, FL, and fruit pectin was suggested by R. Martinez, USDA, APHIS, Gua- temala City, Guatemala. This work was supported by grant US-2217-92 from the United States-Israel Binational Agriculture Research and Development Fund and by the Cooperative State Research Service, USDA, Agreement No. 90-37250-5356.

REFERENCES CITED

DITMAN, L. P., AND E. N. CORY. 1933. The response of corn earworm moths to various sugar solutions. J. Econ. Entomol. 26: 109-115. DUNCAN, D. B. 1955. Multiple range and multiple F tests. Biometrics 11: 1-41. EYER, J. R. 1931. A four year study of codling moth baits in New Mexico. J. Econ. En- tomol. 24: 998-1001. FROST, S. W. 1926. Bait pails as a possible control for the oriental fruit moth. J. Econ. Entomol. 19: 441-450. FROST, S. W. 1928. Continued studies of baits for oriental fruit moth. J. Econ. Ento- mol. 21: 339-348. FROST, S. W. 1929. Fourth contribution to a study of baits, with special reference to the oriental fruit moth. J. Econ. Entomol. 22: 101-108. HOLLAND, W. J. 1903. The moth book. A guide to the moths of North America. Dover Publications Inc., New York. 479 pp. LANDOLT, P. J., AND R. R. HEATH. 1989. Lure composition, component ratio, and dose for trapping male Mocis latipes (Lepidoptera: Noctuidae) with synthetic sex pheromone. J. Econ. Entomol. 82: 307-309. NEWELL, W. 1936. Progress report on the Key West (Florida) fruit ﬂy eradication project. J. Econ. Entomol. 29: 116-120. NORRIS, M. J. 1935. The feeding habits of the adult Lepidoptera Heteroneura. Trans. Roy. Entomol. Soc. London 85: 61-90. OGUNWOLU, E. O., AND D. H. HABECK. 1975. Comparative life histories of three Mocis spp. in Florida (Lepidoptera: Noctuidae). Florida Entomol. 58: 97-103. SARGENT, T. D. 1976. Legion of night. The underwing moths. Univ. Massachusetts Press, Amherst, MA, 222 pp. STEEL, R. G. D., AND J. H. TORRIE. 1960. Principles and procedures of statistics. McGraw Hill Book Co., New York. 481 pp. STRAYER, J. 1975. Sugarcane insect control. Florida Coop. Ext. Serv., Plant Protection Pointers, Ext. Entomol. Rep No. 40. 6 pp.

Briano et al.: Fire Ant Diseases in Argentina 531

PROTOZOAN AND FUNGAL DISEASES IN SOLENOPSIS RICHTERI AND S. QUINQUECUSPIS (HYMENOPTERA: FORMICIDAE) IN BUENOS AIRES PROVINCE, ARGENTINA

J. BRIANO,1 D. JOUVENAZ,2 D. WOJCIK,2 H. CORDO1 AND R. PATTERSON2 1USDA-ARS South American Biological Control Laboratory, Bolivar 1559 (1686) Hurlingham, Buenos Aires province, Argentina

2USDA-ARS Medical and Veterinary Entomology Research Laboratory, P.O. Box 14565, Gainesville, FL, 32604, USA

ABSTRACT

The diversity and abundance of protozoa and fungi infecting colonies of the ﬁre ants Solenopsis richteri Forel and S. quinquecuspis Forel were surveyed in Buenos Aires province, Argentina. A total of 185 roadside sites was selected, and 1,836 colonies were sampled and examined under phase-contrast microscopy. Pathogens were found at 32% of the sites and in 10% of the colonies. The microsporidium Thelohania solenopsae Knell, Allen & Hazard was the most common microorganism; it was present at 25% of the sites and 8% of the colonies. In some sites within the surveyed region, the percentage of infected colonies with T. solenopsae ranged from 40 to 80%. Other pathogens present were the microsporidium Vairimorpha invictae Jouvenaz & Ellis and the fungus Myrmecomyces annellisae Jouvenaz & Kimbrough. A ﬁeld site was selected for future ecological studies.

Key Words: Microsporidia, fungi, natural enemies, biological control, imported ﬁre ants, myrmecology.

RESUMEN

La diversidad y abundancia de protozoarios y hongos infectando colonias de las “hormigas coloradas” Solenopsis richteri Forel y S. quinquecuspis Forel fueron rele- vadas en la Provincia de Buenos Aires, Argentina. Un total de 185 sitios en banquina fueron seleccionados y 1.836 colonias fueron muestreadas y examinadas con micros- copía de contraste de fase. Se encontraron patógenos en 32% de los sitios y 10% de las colonias. El microsporidio Thelohania solenopsae Knell, Allen y Hazard fue el más co- mun de los microorganismos encontrados, estuvo presente en el 25% de los sitios y en 8% de las colonias. En algunos lugares de la región relevada, el porcentage de colonias infectadas con T. solenopsae fue del 40 al 80%. Otros patógenos presentes fueron el microsporidio Vairimorpha invictae Jouvenaz y Ellis y el hongo Myrmecomyces annellisae Jouvenaz y Kimbrough. Fue seleccionado un sitio de campo para futuros estudios ecológicos.

The black imported ﬁre ant, Solenopsis richteri Forel, and the red imported ﬁre ant, S. invicta Buren, were accidentally introduced into the United States from South America (Lofgren 1986b). S. richteri has been displaced by S. invicta and now occurs in a relatively small area of northeastern Mississippi and northwestern Alabama. S. invicta has spread dramatically and has become one of the most serious medical and economic pests in 11 southeastern States and Puerto Rico (Adams 1986, Lofgren 1986a & 1986b, Mackay et al. 1992, Vinson & Mackay 1990, Vinson & Sorensen 1986).

532 Florida Entomologist 78(3) September, 1995

However, their present status as different species should be investigated because, at least in the United States, these two species cross and produce viable hybrids. The black and red imported fire ants were introduced into the United States free of their major natural enemies present in their native land (Jouvenaz et al. 1977). Consequently, a classical biological control approach could be possible with a complex of pathogens, parasites, and/or predators being introduced into the North American fire ant populations. In South America, several pathogens are known from S. invicta, S. richteri and other members of the S. saevissima Smith complex; some of them were surveyed for in Brazil, Uruguay and Argentina (Allen & Buren 1974, Allen & Silveira Guido 1974, Jouvenaz 1983, 1986, Jouvenaz et al. 1980 & 1981, Wojcik et al. 1987). Since 1988, part of the USDA research on biological control of fire ants has been established at the USDA-ARS South American Biological Control Laboratory in Hurl- ingham, Buenos Aires Province, Argentina. Because the target species, S. invicta, does not occur within that area, the primary goal has been to search, select, and evaluate the potential for natural enemies of native Solenopsis spp. as biological agents for control of S. invicta in the United States. The objectives of this work were to survey the diversity and abundance of pathogens in the area of the USDA laboratory and to select a convenient field study site to initiate ecological studies.

MATERIALS AND METHODS

The northeastern region of Buenos Aires province was surveyed from March to September 1988. The area covered approximately 40,000 km2. A total of 185 collecting sites was selected systematically every 10 to 50 km (depending on time available) along the roadsides of the major highways of the region. At each site, usually the first 10 colonies found on the roadside were sampled, which resulted in a total of 1,836 fire ant colonies being sampled. According to Trager (1991), the only fire ant species in that area are S. richteri and S. quinquecuspis. Although not all the samples were identified, we estimate that at least 80% of them were S. richteri. The colonies were sampled by inserting a 7-ml vial into the mounds. The inner walls of the vials were dusted with talc to prevent escape of the ants. When several dozen ants had fallen into the vials, they were removed, capped and placed on ice in a cooler for transportation to the laboratory. Once in the laboratory, the ants were killed by freezing. Then each sample was placed in a glass tissue grinder with about 2-4 ml of water and ground up for about 30 s. One drop of the aqueous extract was examined under phase-contrast microscopy (400×) for the presence of spores of protozoa and fungi. The sensitivity of this procedure for detecting spores in low numbers has been successfully tested in previous research (Jouvenaz et al. 1977). The main criteria in the selection of the field site for future ecological studies were: the incidence of diseases in the local fire ant populations, the type of habitat, and the proximity to the USDA laboratory. Arbitrarily, the limits of the region surveyed were established at 300 km from the USDA laboratory.

RESULTS AND DISCUSSION

The region surveyed and the approximate location of most of the pathogen-positive sampling sites are shown in Fig. 1. Of the total of 185 sites, protozoa and fungi were found at 59 (32%) of the sites (Table 1). The most common pathogen was the protozoan Thelohania solenopsae Knell, Allen, & Hazard (1977) (Microsporida: Thelohaniidae),

Briano et al.: Fire Ant Diseases in Argentina 533

Figure 1. Region surveyed in Buenos Aires province and location of most of the pathogen-positive sites. Shaded areas show the main sampling areas. it was present at 46 (25%) of the sites. The fungus Myrmecomyces annellisae Jouvenaz & Kimbrough (1991) (Deuteromycotina: Hyphomycetes) was found at 14 (8%) of the sites and the protozoan Vairimorpha invictae Jouvenaz & Ellis (1986) (Microsporida:

534 Florida Entomologist 78(3) September, 1995

TABLE 1. PATHOGEN-POSITIVE SAMPLING SITES WITHIN AREA SURVEYED IN BUENOS AIRES PROVINCE.

Number of Number (%) of Sampling Sites with Pathogens Sampling Area Sites T. solenopsae V. invicta M. annellisae Total1

Saladillo 49 23 (47) 6 (12) 7 (14) 32 (65) Isla Talavera 6 2 (33) 1 (20) 0 3 (50) Bragado 27 5 (18) 0 3 (11) 6 (22) Pergamino 24 5 (21) 0 0 5 (21) Others 79 11 (14) 2 (<3) 0 13 (16) Total 185 46 (25) 9 (5) 14 (8) 59 (32)

1Sites had >1 pathogen, so totals are lower than the sum of 3 columns.

Burenellidae) was found at 9 (5%) of the sites. At some sites, more than one pathogen was found infecting different colonies, but never the same colony. However, dual infections with T. solenopsae and V. invictae were observed within extremely stressed laboratory colonies (J. B., unpublished data). According to the incidence of diseases, the region surveyed was arbitrarily divided into 4 main areas (Fig. 1): Saladillo (livestock grazing area), Isla Talavera (swampy area), Pergamino (annual crops area), and Bragado (mixture of annual crops and livestock grazing area). The area of Saladillo (180 km SW of Buenos Aires) had the highest incidence of pathogens with 65% of the sites infected (Table 1). The most common pathogen was T. solenopsae; it was present at 47% of the sites. The fungus M. annellisae was found at 14% of the sites, and V. invictae at 12%. The area of Isla Talavera (100 km NW of Buenos Aires) showed the second highest incidence of diseases with 50% of the sites infected; T. solenopsae was found at 33% and V. invictae at 20%. In the area of Bragado (224 km W of Buenos Aires) 22% of the sites were infected, 18% with T. solenopsae and 11% with M. annellisae. The area of Pergamino (250 km NW of Bue- nos Aires) had 21% of the sites infected with T. solenopsae. No other disease was found in this area. In other areas of the province, only 16% of the sites had diseases; T. solenopsae was present in 14% and V. invictae in less than 3%. (Table 1). Of the 1,836 ﬁre ant colonies sampled, 182 (10%) were infected, 8% with T. solenopsae, 1% with V. invictae and 1% with M. annellisae (Table 2). The highest percentage of infected colonies (33%) was found in the area of Isla Talavera, then in Pergamino (17%), Saladillo (12%), and Bragado (9%). In all these areas, T. solenopsae was the most common pathogen (Table 2). In the other areas of the region, only 4% of the colonies were infected. At several collecting sites, T. solenopsae was found infecting a high proportion of colonies. For example: 80% at Rt. 8, km 254 (Pergamino); 66% at Rt. 8, km 247 (Per- gamino); 50% at Rt. 205, km 180 (Saladillo); 47% at Rt. 12, km 104 (Isla Talavera); and 40% at Rt. 5, km 224 (Bragado). At one collecting site, V. invictae infected 60% of the colonies (Rt. 12, km 91, Isla Talavera). The prevalence of T. solenopsae and V. invictae in those sites represents the highest infection rates with these pathogens in ﬁre ants ever reported for South America (See references below). However, the total percentages of colonies infected with these two microsporidia in the region surveyed (8 and 1%, respectively) were similar to those found in S. invicta in Mato Grosso and Mato Grosso do Sul, Brazil (2.2 to 11.4% and 2 to 4% of the colonies respectively) (Jou-

Briano et al.: Fire Ant Diseases in Argentina 535

TABLE 2. PATHOGEN-POSITIVE COLONIES WITHIN AREA SURVEYED IN BUENOS AIRES PROVINCE.

Number of Number (%) of Colonies with Pathogens Sampled Area Colonies T. solenopsae V. invicta M. annellisae Total

Isla Talavera 90 24 (27) 6 (7) 0 30 (33) Pergamino 175 29 (17) 0 0 29 (17) Saladillo 613 49 (8) 12 (2) 14 (2) 75 (12) Bragado 205 16 (8) 0 3 (<2) 19 (9) Others 753 27 (<4) 2 (<1) 0 29 (4) Total 1,836 145 (8) 20 (1) 17 (1) 182 (10) venaz et al. 1980, Jouvenaz 1986, Jouvenaz & Ellis 1986, Wojcik 1986, Wojcik et al. 1987). In two surveys conducted in 1987 in the provinces of Buenos Aires, Santa Fe, and Entre Rios, Argentina, 10.5% of the fire ant colonies were infected with T. solenopsae (D. J. & D. W., unpublished data). The relatively high prevalence of fire ant pathogens in Argentina and Brazil con- trasts with the lack of specific pathogens in the North American populations of the black and red imported fire ants (Jouvenaz et al. 1977). Surveys in areas infested with both species of imported fire ants in the United States revealed few colonies infected with pathogens (Lofgren et al. 1975). Jouvenaz et al. (1977) reported considerable variation in the rate of infection of S. invicta with M. annellisae in Florida, Alabama and Georgia. This is probably the only (specific?) microorganism imported into the United States in association with the imported fire ants. The prevalence of pathogens reported in this paper was probably underestimated because the sampling of the fire ant colonies included only adult workers. Possibly early infective stages were present in immature fire ants and missed. In addition, other pathogens known to occur in fire ants in South America, such as bacteria and viruses (Avery et al. 1977, Jouvenaz 1983) were not detected with the light microscope. The prevalence of these other pathogens deserves further investigation. We conclude that, within the region surveyed, the microsporidium T. solenopsae was the most common pathogen of indigenous fire ant populations. In some sites, it infected a high proportion of the colonies. Therefore, it should be evaluated as a potential biological control agent for the imported fire ants in the United States. Although the areas of Isla Talavera and Pergamino showed higher number of colonies infected with T. solenopsae, the area of Saladillo was selected as the field study site because: (1) it showed the highest number of sites infected with T. solenopsae, (2) it was the only area where the three pathogens were present, (3) it represented the most appropriate habitat for future long-term ecological studies, and (4) it was near the USDA laboratory and frequent field work would be simplified.

ACKNOWLEDGMENT We would like to thank David Williams (USDA-ARS Medical and Veterinary En- tomology Research Laboratory, Gainesville, FL, USA), Philip Koehler (University of Florida, Gainesville, FL, USA) and Daniel Gandolfo (USDA-ARS South American Bi- ological Control Laboratory, Hurlingham, Buenos Aires, Argentina) for reviewing this manuscript. 536 Florida Entomologist 78(3) September, 1995

REFERENCES CITED

ADAMS, C. T. 1986. Agricultural and medical impact of the imported fire ants, pp 48- 57 in C. S. Lofgren and R. K. Vander Meer [eds.], Fire ants and leaf-cutting ants: biology and management. Westview Press, Boulder, CO. 434 pp. ALLEN, G. E., AND W. F. BUREN. 1974. Microsporidian and fungal diseases of Solenop- sis invicta Buren in Brazil. J. New York Entomol. Soc. 82: 125-30. ALLEN, G. E., AND A. SILVEIRA-GUIDO. 1974. Occurrence of microsporidia in Solenop- sis richteri and Solenopsis sp. in Uruguay and Argentina. Florida Entomol. 57: 327-9. AVERY, S. W., D. P. JOUVENAZ, W. A. BANKS, AND D. W. ANTHONY. 1977. Virus-like particles in a fire ant, Solenopsis sp., (Hymenoptera: Formicidae) from Brazil. Florida Entomol. 60: 17-20. JOUVENAZ, D. P. 1983. Natural enemies of fire ants. Florida Entomol. 66: 111-121. JOUVENAZ, D. P. 1986. Diseases of Fire Ants: Problems and Opportunities, pp. 327- 338 in C. S. Lofgren and R. K. Vander Meer [eds.], Fire ants and leaf-cutting ants: biology and management. Westview Press, Boulder, CO. 434 pp. JOUVENAZ, D. P., AND E. A. ELLIS. 1986. Vairimorpha invictae n. sp. (Microspora: Mi- crosporida), a parasite of the red imported fire ant, Solenopsis invicta Buren (Hymenoptera: Formicidae). J. Protozool. 33: 457-461. JOUVENAZ, D. P., AND J. W. KIMBROUGH. 1991. Myrmecomyces annellisae gen. n., sp. n. (Deuteromycotina: Hyphomycetes), an endoparasitic fungus of fire ants, So- lenopsis spp. (Hymenoptera: Formicidae). Mycolog. Res. 95: 1395-1401. JOUVENAZ, D. P., G. E. ALLEN, W. A. BANKS, AND D. P. WOJCIK. 1977. A survey for pathogens of fire ants, Solenopsis spp., in the southern United States. Florida Entomol. 60: 275-279. JOUVENAZ, D. P., W. A. BANKS, AND J. D. ATWOOD. 1980. Incidence of pathogens in fire ants, Solenopsis spp. in Brazil. Florida Entomol. 63: 345-346. JOUVENAZ, D. P., C. S. LOFGREN, AND W. A. BANKS. 1981. Biological control of imported fire ants: a review of current knowledge. Bull. Entomol. Soc. America 27: 203-208. KNELL, J. D., G. E. ALLEN, AND E. I. HAZARD. 1977. Light and electron microscope study of Thelohania solenopsae n. sp. (Microsporida: Protozoa) in the red imported fire ant, Solenopsis invicta. J. Invertebr. Pathol. 29: 192-200. LOFGREN, C. S. 1986a. The economic importance and control of imported fire ants in the United States, pp. 227-256 in S. B. Vinson [ed.], Economic impact and control of social insects. Praeger, N. Y. 421 pp. LOFGREN, C. S. 1986b. History of imported fire ants in the United States, pp. 36-49 in C. S. Lofgren and R. K. Vander Meer [eds.], Fire ants and leaf-cutting ants: biology and management. Westview Press, Boulder, CO. 434 pp. LOFGREN, C. S., W. A. BANKS, AND B. M. GLANCEY. 1975. Biology and control of imported fire ants. Annu. Rev. Entomol. 20: 1-30. MACKAY, W. P., S. B. VINSON, J. IRVING, S. MAJDI, AND C. MESSER. 1992. Effect of electrical fields on the red imported fire ant (Hymenoptera: Formicidae). Environ. Entomol. 21: 866-870. TRAGER, J. C. 1991. A revision of the fire ants, Solenopsis geminata group (Hy- menoptera: Formicidae: Myrmicinae). J. New York Entomol. Soc. 99: 141-198. VINSON, S. B., AND W. P. MACKAY. 1990. Effects of the fire ant Solenopsis invicta on electrical circuits and equipment, pp. 496-503 in R. K. Vander Meer, K. Jaffe and A. Cedeño [eds.], Applied myrmecology: a world perspective. Westview Press, Boulder, CO. 741 pp. VINSON, S. B., AND A. A. SORENSEN. 1986. Imported fire ants: Life history and impact. Texas A & M University, College Station, Texas and Texas Department of Agri- culture, Austin, Texas. 28 pp. WOJCIK, D. P. 1986. Observations on the biology and ecology of fire ants in Brazil, pp. 88-103 in C. S. Lofgren and R. K. Vander Meer [eds.], Fire ants and leaf-cutting ants: biology and management. Westview Press, Boulder, CO. 434 pp. Briano et al.: Fire Ant Diseases in Argentina 537

WOJCIK, D. P., D. P. JOUVENAZ, W. A. BANKS, AND A. C. PEREIRA. 1987. Biological control agents of ﬁre ants in Brazil, pp. 627-628 in J. Eder and H. Rembold [eds.], Chemistry and biology of social insects. Verlag J. Perperny, Munchen. 757 pp.

538 Florida Entomologist 78(3) September, 1995

NATURAL ENEMIES OF BEMISIA TABACI (HOMOPTERA: ALEYRODIDAE) IN CUBA

ANTONIO CASTINEIRAS Tropical Research and Education Center University of Florida, IFAS 18905 SW 280th St. Homestead, Florida 33031

The sweet potato whitefly, Bemisia tabaci (Guennadius), reached pest status in Cuba during the growing season of 1989-90, seriously damaging tomato, eggplant, squash, cucumber and other vegetable crops. Taxonomic status of the genus Bemisia in Cuba is currently under study. As the correct identity of the species damaging crops is yet to be determined, the name B. tabaci will be used in this note. Several parasitoids (Aphelinidae, Platygasteridae, Encyrtidae), predators (Chrysopidae, Anthocoridae, Coccinellidae, Phytoseiidae, Stigmaeidae) and the entomopathogenic fungus Paecilomyces farinosus (Dickson & Fries) Brown & Smith have been reported as natural enemies of B. tabaci from different countries (Cock 1993). Polaszek et al. (1992) noted parasitoids collected from the Caribbean region, but there are no records of biological control agents of B. tabaci from Cuba. Surveys for natural enemies were conducted in the most important agricultural areas located at the Western, Central and Eastern regions of Cuba, and on the Isle of Pines, from January to April 1990, 1991 and 1992. The climate in Cuba during this period is characterized by temperatures of 23-25°C, 80-85% relative humidity and less than 5 mm rainfall. Sampled plots (1 ha) were selected from eggplant, tomato, pepper, cabbage, squash, cucumber, and sweet potato fields not heavily sprayed with chemical insecticides, or only treated with Bacillus thuringiensis (Berl.). Samples consisted of one mature leaf, taken at random from each 50 plants on an X-shaped design, once in each growing season. A total of 5,000 leaves were collected from 100 fields every year. More than 10,000 whitefly nymphs per year were collected. All leaf parts with whiteflies were cleaned of other insects, cut and kept in vials until adult parasitoids emerged. Whiteflies were considered parasitized only when a parasitoid emerged from a nymph. When fungal parasites were found, only dead nymphs surrounded by mycelium were considered infected. The fungus was isolated, identified, and inocu- lated on healthy nymphs in the laboratory to corroborate its pathogenic effect. Imma- ture predators (except spiders) were fed with whitefly larvae until development was completed. Material from the surveys was identified at the Institute of Plant Protec- tion (Ministry of Agriculture) and the National Museum of Natural History (Academy of Sciences), Havana, Cuba. Four parasitoids, an entomopathogenic fungus, one hyperparasitoid, and five predators were found (Table 1). Encarsia luteola Howard, E. nigricephala Dozier and E. quaintancei Howard.— E. luteola and E. quaintancei were the parasitoids most frequently found in Cuba during the growing seasons of 1989-92. They were present on all seven crop species on which whiteflies were collected. Less abundant than the former two species, E. nigricephala was only found in the Western region of the country. According to Polas- zek et al. (1992), E. nigricephala, E. luteola and E. quaintancei are well represented in the Caribbean Basin, but the three species had not been documented from Cuba before. Eretmocerus sp. (Aphelinidae), and Signiphora sp. (Signiphoridae).—Parasitoids of the genus Eretmocerus are frequently associated with sweet potato whitefly (Polas-

Scientiﬁc Notes 539

TABLE 1. NATURAL ENEMIES OF BEMISIA TABACI IN CUBA, EFFECT ON THE HOST, AND DISTRIBUTION.

Distribution Parasitism or Natural Enemies Predation Western Central Eastern Isle of Pines

Encarsia luteola 20-30% X X X X Howard E. nigricephala Dozier 15-20% X E. quaintancei Howard 25-35% X X X X Eretmocerus sp. Not evaluated X X Signiphora sp. Not evaluated X X Paecilomyces 70-80% X fumosorosus (Wize) Theridula 5-10 adults/web X X gonygaster Simon Theridula sp. 5-10 adults/web X Delphastus Not evaluated X X X X pallidus LeC. Chrysopa exterior Not evaluated X X X X Navas Cyrtopeltis Not evaluated X X X varians (Dist.)

zek et al. 1992). Some species, such as E. haldemani Howard (Gerling 1967) or E. mundus Mercet (Kapadia & Puri 1990), are major constituents of the parasitoid complex of B. tabaci in different countries. In Cuba, Eretmocerus sp. was only found once, on peppers, in the Central and Eastern regions in 1991. Signiphora sp. is a hyperparasitoid and it was collected from samples taken from tomato, cabbage, squash, cucumber and sweet potato from the Western and Central agricultural areas of Cuba. Paecilomyces fumosoroseus (Wize).—P. fumosoroseus is considered a potential biological control agent of the sweet potato whitefly (Osborne et al. 1990). First epizootics of P. fumosoroseus were found in April 1990 in the South Havana area after a period of heavy rain. The same situation was observed in 1991 and 1992 when the weather was very humid. Osborne et al. (1990) found over 90% mortality of fourth instar nymphs treated with the fungus in the laboratory. In Cuba, natural epizootics killed 70-80% of nymphs in the field. The fungus also appeared spontaneously on B. tabaci in some of the greenhouses in the Institute of Plant Protection in Havana. Theridula gonygaster Simon and Theridula sp.—These spiders of the family The- ridiidae build small webs on the lower surface of the leaves where they catch adults of the sweetpotato whitefly. They were observed very frequently in heavily infested cucumber, squash, sweet potato and tomato fields in the Western and Central regions of the country.

540 Florida Entomologist 78(3) September, 1995

Delphalstus pallidus LeC. (Coccinellidae), Chrysopa exterior Navas (Chrysopidae) and Cyrtopeltis varians (Dist.) (Miridae).—Despite their wide distribution in Cuba these predators were not frequently associated with B. tabaci. Immatures found on infested sweet potato and cucurbit leaves in the field completed their development in the laboratory when they were fed on sweetpotato whitefly larvae. In California, D. pusillus is under study for use in management programs for B. tabaci (Heinz et al. 1994). In general, parasitoids and predators did not seem to have a strong effect as lim- iting factors of the sweet potato whitefly populations. Apparently crop species did not affect parasitism, but a greater diversity of natural enemies was found in cucurbits, tomato and sweet potato. The fungus P. fumosoroseus showed a high percentage of parasitism, but only during very humid periods, making it a good candidate for biocontrol in greenhouses. This is Journal Experimental Series No R-04117.

SUMMARY

Surveys for natural enemies of the sweet potato whiteﬂy, Bemisia tabaci (Guenn.), were conducted in Cuba from January to April 1990, 1991 and 1992. Four parasitoids (Encarsia luteola Howard, E. nigricephala Dozier, E. quaintancei Howard and Eret- mocerus sp.), one hyperparasitoid (Signiphora sp.), an entomopathogenic fungus (Paecilomyces fumosoroseus (Wize)) and ﬁve predators (Theridula gonygaster Simon, Theridula sp. Delphastus pallidus LeC., Chrysopa exterior Navas, and Cyrtopeltis varians (Dist.)) were detected.

REFERENCES CITED

COCK, J. M. 1993. (Ed.) Bemisia tabaci-an update 1986-1992 on the cotton whitefly with an annotated bibliography. International Institute of Biological Control., Ascot, Berks, UK. 78 pp. GERLING, D. 1967. Bionomics of the whitefly-parasite complex associated with cotton in Southern California (Homoptera:Aleurodidae; Hymenoptera:Aphelinidae). Ann. Entomol. Soc. America 60:1306-1321. HEINZ, K. M., J. R. BRAZZLE, C. H. PICKETT, E. T. NATWICK, J. M. NELSON, AND M. P. PARELLA. 1994. Predatory beetle may suppress silverleaf whitefly. California. Agric. 48:35-40. KAPADIA, M. N., AND S. N. PURI. 1990. Development, relative proportions and emergence of Encarsia transvena (Timberlake) and Eretmocerus mundus Mercet, important parasitoids of Bemisia tabaci (Guennadius). Entomol. 15:235-239. OSBORNE, L. S., G. K. STOREY, C. W. MCCOY, AND J F. WALTEW. 1990. Potential for controlling the sweet potato whitefly, Bemisia tabaci, with the fungus, Paecilo- myces fumosoroseus. Proceedings and abstracts, Vth International Colloquium on Invertebrate Pathology and Microbial Control, Adelaide, Australia, 20-24 August 1990. Adelaide, Australia, pp. 386-390. POLASZEK, A., G. A. EVANS, AND F. D. BENNETT. 1992. Encarsia parasitoids of Bemisia tabaci (Hymenoptera:Aphelinidae, Homoptera: Aleyrodidae): a preliminary guide to identification. Bull. Entomol. Res. 82:375-392.

Scientiﬁc Notes 541

EXTRACTION OF HOUSE FLY (DIPTERA: MUSCIDAE) LARVAE FROM POULTRY MANURE USING BERLESE/TULLGREN FUNNELS

DONALD R. BARNARD Medical and Veterinary Entomology Research Laboratory Agriculture Research Service U.S. Department of Agriculture Gainesville, FL 32604 U.S.A. Accurate determination of the numbers of late instar house fly (Musca domestica L.) larvae in samples of poultry manure is the first step in developing density estimators for populations of pupae and emerging adult flies. These density estimators can be used to anticipate fly abatement needs on agricultural facilities and to improve the efficiency of releases of pupal parasitoids in augmentation and inundation-based biological control programs for M. domestica. Presently, there is no practical method for determining the density of house fly larvae in a sample of poultry manure. Estimates of this factor (Geden & Stoffolano 1987; Stafford & Bay 1987) have been deduced mainly from extraction data using Berlese/Tullgren funnels (Brydon & Fuller 1966). However, the accuracy of these devices for separating house fly larvae from samples of poultry manure is not well understood. This study was made to assess four parameters whose influence on the mean and variance of numbers of third instar house fly larvae extracted from samples of poultry manure placed in Berlese/Tullgren funnels is unknown. These parameters are: (1) the size of the manure sample, (2) the number of larvae in the sample, (3) the cumulative extraction rate of larvae from the manure sample over time, including the relation between extraction rate and temperature change within the manure sample, and (4) the effect of enclosure of larvae and manure in sample containers (for ≤4 h) on the accuracy of larval extraction. Berlese/Tullgren funnels were constructed according to the design of Brydon & Fuller (1966). Poultry manure for the experiments was collected for 24 h in trays suspended beneath commercial laying hens. The manure was stored at -20°C then allowed to thaw to ambient temperature 8 h before use. A 4 × 4 factorial design was used to determine the effect of larval density and manure volume on the accuracy of larval extraction. Each treatment combination was replicated four times (n = 64) and comprised one of four volumes of manure (100, 200, 300, and 400 cc) and one of four levels of larval density (50, 100, 300, 500). Two hours before a test began, known numbers of 72-hour-old early third instar larvae reared at 26°C on the Gainesville house fly diet (Hogsette 1992) were place on, and allowed to penetrate, each manure sample. Individual samples were placed onto a single layer of cheesecloth on top of the 5 mm mesh screen (Brydon & Fuller 1966) in each funnel and shaped to a thickness of ≈ 40 mm. Larvae exiting the samples, after funnel covers were positioned and the lights (100 watts) turned on, dropped into 70% ethanol in water in 0.25-liter glass jars at the bottom of each funnel. Jars were replaced at 12 h intervals until no larvae were collected for 2 consecutive intervals. To determine the cumulative number of larvae extracted from a manure sample over time, nine samples of 100 larvae in 400 cc of manure were prepared as described above and placed in funnels at 0 h. Collection jars were removed from funnels at 6, 24, 30, and 48 h. Temperature change at the top, middle, and bottom positions in three manure samples was monitored continuously, and reported hourly, using an Omni- data EL-820 data logger (Omnidata, P.O. Box 3489, Logan, UT 84321) and ES-060 temperature probes. The experiment was repeated four times (n = 36).

542 Florida Entomologist 78(3) September, 1995

A 2 × 4 factorial design was used to determine the effect of ≤4 h of enclosure of larvae and manure in sample containers on the accuracy of larval extraction. Treatment factors were open/closed sample container and the length of time the container was open or closed (1, 2, 3, or 4 h). At 1 h intervals, beginning 4 h before all samples were placed in the funnels (0 h) and ending at 1 h, six samples of 100 larvae in 400 cc of manure were placed into sealable plastic containers (10 × 10 × 15 cm) and held at ambient temperature (22-26°C). Air-tight lids were placed on three of the containers and these containers labelled as closed. The remaining three unsealed containers were labelled as open. All samples were placed in funnels at 0 h and the funnels operated, as before, for 48 h. In all tests, the percent extraction of larvae for each funnel was calculated as a ratio of the numbers of larvae in the collection jar to the numbers of larvae in the manure sample, multiplied × 100. Data for percent extraction (transformed to arcsin) were analyzed using analysis of variance procedures (SAS Institute, Inc. 1988). Means separation was performed using Tukey’s studentized range test at P = 0.05. The relationship between cumulative extraction rate (%CR) and temperature change in the manure sample at top (t), middle (m), and bottom (b) positions was determined using the regression model: %CR = t + t2 + m + m2 + b + b2 (SAS Institute, Inc. 1988). Percent extraction of third instar house fly larvae from samples of poultry manure in Berlese/Tullgren funnels was not influenced by the size of the manure sample or by the numbers of larvae in the sample (Table 1). The overall percent extraction of larvae in this study (n = 118) was 98.5 ± 1.9. The length of time Berlese/Tullgren funnels were operated influenced the percent extraction of fly larvae (F1,35 = 39.69, P = 0.01). Significantly fewer (57.7%) larvae were recovered after 6 h of operation than after 24 (90.2%), 30 (96.6%) or 48 h (96.6%). Cu- mulative percent extraction of larvae was related to the linear and quadratic effects 2 of manure temperature change (F2,143 = 54.68; R = 0.437) at the top position. There was no significant change in manure temperature at any position after 30 h of funnel operation. The enclosure of fly larvae and manure in sample containers for up to 3 h did not influence overall percent extraction of larvae compared with the data for larvae in ± ± open containers for the same length of time (xopen = 98.8 1.8%; xclosed = 99.2 0.7%). ± ± The results at 4 h (xopen = 99.3 0.05%; xclosed = 64.0 55.4%), however, were equivocal. Of the three containers closed for 4 h, one yielded 0 larvae while percent extraction for the remaining two was 96%. The results at 4 h, for closed containers, while not statistically different from the earlier times, indicate that enclosing larvae and manure in a sample container for more than 3 h at 22-26°C (as might be necessary during transport of samples) will result in variable extraction responses.

TABLE 1. MEAN PERCENT EXTRACTION (±SEM) OF THIRD INSTAR HOUSE FLY LARVAE FROM POULTRY MANURE IN BERLESE/TULLGREN FUNNELS.

Number of Larvae Manure Volume 50 100 300 500

100 cc 99.5 (1.0) 88.1 (18.7) 95.8 (3.9) 99.3 (0.9) 200 cc 95.5 (4.1) 100.0 (0.0) 99.3 (0.5) 98.8 (0.5) 300 cc 100.0 (0.0) 99.8 (0.5) 98.3 (2.2) 99.5 (1.0) 400 cc 99.3 (2.0) 98.3 (1.7) 98.5 (1.3) 98.1 (0.8)

Scientiﬁc Notes 543

SUMMARY

Berlese/Tullgren funnels provide an accurate estimate of the density of third instar house ﬂy larvae in samples of poultry manure. Percent extraction of larvae did not vary with sample size, when the range of sample sizes was between 100 and 400 cc of poultry manure, and was unaffected by larval densities from 50 to 500 per sample. Maximum extraction of house ﬂy larvae requires ≥30 h of funnel operation. The accuracy of extraction decreases when manure and larvae are enclosed together in a sample container for more than 3 h prior to placement of samples in funnels.

REFERENCES CITED

BRYDON, H. W., AND R. G. FULLER. 1966. A portable apparatus for separating fly larvae from poultry droppings. J. Econ. Entomol. 59: 448-452. GEDEN, C. J., AND J. G. STOFFOLANO. 1987. Dispersion patterns of arthropods associated with poultry manure in enclosed houses in Massachusetts: spatial distribution and effects of manure moisture and accumulation time. J. Entomol. Sci. 23: 136-148. HOGSETTE, J. A. 1992. New diets for production of house flies and stable flies (Diptera: Muscidae) in the laboratory. J. Econ. Entomol. 85: 2291-2294. SAS INSTITUTE, INC. 1988. SAS/STAT user’s guide, release 6.03. SAS Institute, Cary, NC. STAFFORD III, K. C., AND D. E. BAY. 1987. Dispersion pattern and association of house fly Musca domestica (Diptera: Muscidae) larvae and both sexes of Macrocheles muscaedomesticae (Acari: Macrochelidae) in response to poultry manure moisture, temperature, and accumulation. Environ. Entomol. 16: 159-164.

♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦

Scientiﬁc Notes 543

PARASITOIDS AND PREDATORS ASSOCIATED WITH SYNTOMEIDA EPILAIS (LEPIDOPTERA: ARCTIIDAE) ON OLEANDER

HEATHER J. MCAUSLANE1 AND FRED D. BENNETT1,2 1Department of Entomology & Nematology, University of Florida, P.O. Box 110620, Gainesville, FL 32611-0620

2Current Address: Crofton, Baldhoon Road, Laxey, IM4 7NA, Isle of Man, United Kingdom

Oleander, Nerium oleander L., is an apocynaceous ornamental shrub native to the Mediterranean region. It is planted widely in the southern United States, from Flor- ida to California, throughout the Caribbean and into Mexico (Reinert 1980). In Flor- ida, oleander is ubiquitous in central and southern regions, often planted in municipal parks, recreational areas, highway right-of-ways, around public buildings and in homeowner’s yards. One of the major insect pests of oleander in Florida and the Caribbean is the larva of the arctiid moth, Syntomeida epilais (Walker) (the “polka-dot wasp moth”). Olean- der caterpillars are predictable and perennial pests, often defoliating oleander bushes during multiple generations per year. This species appears to be a native of the Neo- tropics. It is indigenous throughout Florida, the Keys and the Dry Tortugas (Kimball

544 Florida Entomologist 78(3) September, 1995

1965), and it is present in Mexico, the Caribbean region and northern countries of South America (Bratley 1932). Larvae and adults of S. epilais contain cardiac glycosides obtained from their host plant (Rothschild et al. 1973) and are distasteful to avian predators (Jones 1934), as are other cardiac glycoside-containing insects such as the monarch butterfly, Danaus plexippus (L.) (Brower et al. 1967). Parasitoids recorded from S. epilais include the tachinid flies Chetogena (=Euphorocera) floridensis (Townsend), Lespesia aletiae (Riley) (Patton 1958) and Lespesia sp. (Fernald 1934). No published records of predation have been found. Collections of immature stages of S. epilais were made in several areas of Florida as part of an effort to understand the contribution of natural enemies to population regulation of the oleander caterpillar. In addition to several small collections, large numbers of immatures were collected on four occasions (Table 1). Larvae were fed cut oleander foliage while prepupae and pupae were placed in cylindrical plastic containers (15-cm diam × 25-cm height) for emergence of adults. On Davis Island on 4 August 1991, numerous larvae were collected on oleander and from the leaf litter below oleander. Pupae were collected from aggregations in de- pressions in the trunk of an oak tree. Individuals of an unidentified Brachymeria species were observed flying around the plants and pupation sites. These wasps were also observed in August 1994, in Clearwater, although no collections were made. Wasps flew quickly and buzzed loudly, similar to houseflies. Brachymeria incerta (Cresson) and two species of tachinid (C. floridensis and L. aletiae) emerged from S. epilais collected on Davis Island. This is a new host record for B. incerta which has been listed both as a primary parasitoid attacking pupae of many Lepidoptera and as a secondary parasitoid attacking Carcelia lagoae (Townsend) and other Lepidoptera-parasitizing tachinids (Burks 1960). Brachymeria incerta was confirmed as a primary parasitoid of S. epilais by presenting mated females with pupae of S. epilais that had been reared from second instars in the laboratory and, therefore, were unlikely to contain larvae of larval/pupal parasitizing tachinids. Adult B. incerta emerged from these pupae approximately 1 month later.

TABLE 1. DATES AND LOCATIONS OF COLLECTIONS FOR S. EPILAIS NATURAL ENEMIES FROM OLEANDER IN FLORIDA.

Collection Location S. epilais Immature Natural Enemies Collection Date (Collected by) States Collected Encountered

4-Aug-91 Davis Island, Larvae, prepupae, Brachymeria Hillsborough Co. pupae incerta, tachinids (F. D. Bennett) 11-Aug-93 Gainesville, Largae, prepupae, Brachymeria Alachua Co. pupae incerta, ﬁre ants (H. J. McAuslane) 20-Jun-94 Homestead, Larvae Pathogens, Dade Co. tachinids (H. J. McAuslane) 6-Jan-95 Gainesville, Larvae, prepupae, Tachinids, Alachua Co. pupae pteromalid (H. J. McAuslane) hyperparasite of tachinid, pathogens

Scientiﬁc Notes 545

In Gainesville, on 11 July 1993, predation by an immature pentatomid Podisus maculiventris (Say) was observed on a third-instar S. epilais feeding on oleander. The nymph was reared to adulthood on third-instar oleander caterpillars. This pentatomid is a generalist feeder but has not been recorded from S. epilais according to McPherson (1982). Many B. incerta were reared from the pupae collected from an ag- gregated pupation site in Gainesville on 11 August 1993. The geographic range of B. incerta, listed as central and southern Florida, West Indies, Guyana and Brazil (Burks 1960), can be extended to north Florida. The pupal mass was also heavily infested by red imported fire ants, Solenopsis invicta Buren, which were observed feeding actively on S. epilais pupae. Two tachinids emerged from a collection of larvae in Homestead on 20 June 1994. However, of the 300+ larvae collected, more than 80% of them succumbed to a pathogen. The symptomology proceeded from the excretion of large, liquid feces, to the des- iccation of the abdomen, to death. However, the specimens were not examined microscopically to determine the nature of the pathogen. Of the approximately 100 larvae collected in Gainesville on 8 January 1995, 13 went on to pupate and produce moths while 57 produced L. aletiae. In addition, three adults of the hyperparasitic perilampid Perilampus hyalinus Say were reared from puparia of L. aletiae. While P. hyalinus has been collected from Lespesia frenchii (Will- iston), L. euchaetiae (Webber) and L. melalophae (Allen) (Krombein et al. 1979), it has not been recorded from this species. The remainder of the larvae suffered mortality due to a pathogen that produced symptoms characteristic of a viral infection (degen- eration of the larval body to a black liquid). Despite the casual nature of these observations, they indicate that a number of potentially significant natural enemies and pathogens of S. epilais exist in Florida. However, despite these sources of mortality, oleander caterpillar continues to defoli- ate oleander regularly in southern portions of the state. Despite the toxic nature of S. epilais larvae and pupae, the predators and parasitoids associated with this species are generalist feeders. In a similar situation, Euploea core corinna (W. S. Macleay), a native nymphalid in Australia that has exploited oleander as a novel host, is attacked by two species of tachinid (Paradrino laevicula Mesnil and Winthemia neowinthemi- oides [Townsend]) and Brachymeria lasus (Walker) which also are generalist feeders (Rahman & Zalucki 1986). Thanks to F. Mead, G. Steck and L. Stange (Florida Dept. of Agriculture and Con- sumer Services, Division of Plant Industry, Gainesville, FL) for identification of the pentatomid, and tachinid and hymenopteran parasitoids, respectively. Thanks also to D. Habeck and H. Frank (Dept. of Entomology & Nematology, University of Florida) for critical reviews of an earlier version of this manuscript. This is Florida Agricul- tural Experiment Station Journal Series No. R-04448.

SUMMARY The oleander caterpillar, Syntomeida epilais, is attacked in Florida by various generalist parasitoids including the tachinids Lespesia aletia and Chetogena (=Euphoro- cera) ﬂoridensis and the chalcidid Brachymeria incerta (new host record). Fire ants, Solenopsis invicta, a predatory pentatomid, Podisus maculiventris, and pathogens also attack S. epilais.

REFERENCES CITED

BRATLEY, H. E. 1932. The oleander caterpillar, Syntomeida epilais, Walker. Florida Entomol. 15: 55-64.

546 Florida Entomologist 78(3) September, 1995

BROWER, L. P., J. V. Z. BROWER, AND J. M. CORVINO. 1967. Plant poisons in a terres- trial food chain. Proc. Natl. Acad. Sci. 57: 893-898. BURKS, B. D. 1960. A revision of the genus Brachymeria Westwood in America north of Mexico. Trans. American Entomol. Soc. 86: 225-273. FERNALD, H. T. 1934. [Oleander - Polka dot wasp moth (Syntomeida epilais Walk.)]. Bull. Insect Pest Survey 14: 23-24. JONES, F. M. 1934. Further experiments on the coloration and relative acceptability of insects to birds. Trans. Roy. Entomol. Soc. Lond. 82: 443-453. KIMBALL, C. P. 1965. Lepidoptera of Florida. Div. of Plant Ind., Florida Dept. of Agr., Gainesville, Florida. KROMBEIN, K. V., P. D. HURD, JR., D. R. SMITH, AND B. D. BURKS. 1979. Catalog of Hy- menoptera in America north of Mexico. Vol. 1. Symphyta and Apocrita (Para- sitica). Smithsonian Institution Press, Washington, D.C. MCPHERSON, J. E. 1982. The Pentatomoidea (Hemiptera) of northeastern North America with emphasis on the fauna of Illinois. Southern Illinois University Press, Carbondale, IL. PATTON, C. N. 1958. A catalogue of the Larvaevoridae of Florida. Florida Entomol. 41: 29-39, 77-89. RAHMAN, H. U., AND M. P. ZALUCKI. 1986. Parasitoid records for Euploea core corinna (W. S. Macleay) (Lepidoptera: Nymphalidae) in south-eastern Queensland. Aust. Entomol. Mag. 12: 109-111. REINERT, J. A. 1980. Control of the oleander caterpillar on oleander. Proc. Florida State Hort. Soc. 93: 168-169. ROTHSCHILD, M., J. VON EUW, AND T. REICHSTEIN. 1973. Cardiac glycosides (heart poisons) in the polka-dot moth Syntomeida epilais Walk. (Ctenuchidae: Lep.) with some observations on the toxic qualities of Amata (=Syntomis) phegea (L.). Proc. Roy. Soc. London Ser. B 183: 227-247.

♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦

546 Florida Entomologist 78(3) September, 1995

PARASITES OF THE PEPPER FLOWER-BUD MOTH (LEPIDOPTERA: GELECHIIDAE) IN FLORIDA

FRED D. BENNETT Entomology and Nematology Department, University of Florida, Gainesville, FL 32611-0620, USA Current address: Crofton, Baldhoon Road, Laxey, Isle of Man IM4 7NA The pepper flower-bud moth, Symmetrischema capsica (Bradley & Polovny,´ (Lep- idoptera: Gelechiidae) described from specimens reared from Capsicum annuum L. in Montserrat, West Indies, occurs widely in the Caribbean. It has been an important pest of Capsicum spp. in this region for at least 65 years (Wilson 1923). It occurs in states in the USA bordering the Gulf of Mexico and also has been collected in Mexico (Bradley & Polovny´ 1965). In 1988 this species was reared from an uncultivated Cap- sicum sp. in Honduras (unpubl. data). Initially placed in the genus Gnorimoschema by Bradley & Polovny´ (1965), the species was transferred to the genus Symme- trischema by Polovny´ (1967). Earlier it had been confused with, and identified as, Gnorimoschema gudmanella (Walsingham); many of the earlier references to its status and records of its natural enemies appear in the literature under that name. In the Caribbean, Parasram (1973) and des Vignes (1978) reported flower-bud drop of 40 to 100% as the result of attack by S. capsicum. This insect was first reported from Florida in 1944 and from Texas in 1945 (An- nand 1945). Schuster (1960) reported 70-100% bud damage, seasonally, in the Rio

Scientiﬁc Notes 547

TABLE 1. PARASITES RECORDED FROM SYMMETRISCHEMA CAPSICA.

Family Genus and Species Stage Distr. Ref.1

Encyrtidae Apsilophrys capsicum egg/lar. Trinidad 2 (Burks) P. Rico Braconidae Agathis sp. larval Trinidad 2 Bracon sp. larval Trinidad 2 Chelonus phthorimaeae egg/lar. Texas 3 Gahan Chelonus sp. egg/lar. P. Rico 2 Orgilis capsicolae Mues. larval Texas 3 Pseudapanteles dignus larval Trinidad 1 (Mues.) P. Rico 2 Florida 4 Bethylidae Perisierola sp. larval Trinidad 2 Eulophidae Euderus sp. larval Trinidad 1 P. Rico 2 Euderus ?purpureus Yosh. larval Florida 4 Pteromalidae Genus & sp. indet. larval Montserat 1

1Ref. = Reference: 1 = Parasram 1973; 2 = des Vignes 1978 & 1981; 3 = Schuster 1960; 4 = current study.

Grande Valley, Texas. Currently, in Florida, pepper bud-moth is not a pest of commercial peppers; possibly it is controlled by the pesticides applied against pepper weevil and beet armyworm (D. J. Schuster pers. comm.). Des Vignes (1978 & 1981) reported on studies of parasites of S. capsica which he undertook in Trinidad, whereas, the only report of parasites in the USA appears to be that given by Schuster (1960) (see Table 1). In May 1985, while surveying for natural enemies of Parlatoria ziziphi (Lucas) in the “Little Haiti” section of Miami, I collected flower buds of Capsicum minima L. infested by S. capsica (identity confirmed by John Heppner) and reared two species of parasites. I made further collections on subsequent surveys from June 1986 to July 1987 and assessed the level of attack and the occurrence of parasites. The data obtained from these collections are given in Table 2. Only the braconid Pseudapanteles dignus (Mues.) (formerly Apanteles dignus) and the eulophid Euderus sp. (possibly E. purpureus Yoshimoto) were reared. P. dignus has been reared from this host in the Caribbean and from several other gelechiid moths in the USA; among the latter are records from Florida (Marsh 1979). Because of the uncertainty of its specific identity, it is not possible to give the distribution of E. ?purpureus. Specimens identified as Euderus sp. are known to attack pepper bud-moth in Puerto Rico and Trinidad (des Vignes 1978, 1981). Records of parasites of S. capsica, including those from the current survey, are given in Table 1. In Florida, only two parasites have been reared from S. capsica, in contrast to Trinidad where des Vignes (1981), during a two-year study, recorded 6 species. It is probable that a more extensive survey in Florida would contribute new parasite records. I greatly appreciate the determinations of the parasitoid species by P. M. Marsh (Braconidae) and M. E. Schauff (Eulophidae), and of the pepper flower-bud moth by J. Heppner.

548 Florida Entomologist 78(3) September, 1995 % Parasitized 1987. ULY 1986-J UNE , J Parasitized by IAMI ”, M Pseudapanteles Euderus AITI H ITTLE 113 7 36 27.6 “L Larvae AT

Unparasitized OLLECTED C UDS Larvae Buds with -B LOWER F EPPER P IN

CAPSICA S.

ARASITISM 2. P ABLE 86.06.2886.07.2186.09.2487.01.2787.04.27 4687.05.28 5487.06.22 5387.07.28 20Total 63 45 57 32 48 18 45 08 53 33 55 18 45 07 28 03 23 21 44 18 23 07 05 03 13 — 35 — 11 — 05 — 2 12 1 — 4 — — — 36.4 8 8 — 8 — — — 43.5 20.5 52.1 — Date Buds CollectedAttacked Buds T

Scientiﬁc Notes 549

University of Florida, Florida Agricultural Experimental Station Journal Series No. R-04104.

SUMMARY

Parasites of the pepper ﬂower-bud moth Symmetrischema capsica (Bradley & Polovny),´ (Lepidoptera: Gelechiidae) are reported from Florida for the ﬁrst time. Combined parasitism by the braconid Pseudapanteles dignus (Mues.), and the eulophid Euderus ?purpureus Yoshimoto as high as 52.1% (average 28.6%) was recorded.

REFERENCES CITED

ANNAND, P. N. 1945. Report of the chief of the bureau of entomology and plant quarantine USDA. 1944-1945: 30 BRADLEY, J. D., AND D. POLOVNY´. 1965. The identity of the pepper flower-bud moth. Bull. Entomol. Res. 56: 57-65. DES VIGNES, W. G. 1978. Notes on the taxonomy, biology and economic importance of Symmetrischema capsica (Bradley & Povolny).´ Jour. Agric. Soc. Trinidad & To- bago. 78: 311-318. DES VIGNES, W. G. 1981. Notes on parasites and predators of Symmetrischema capsica (Bradley & Povolny).´ Jour. Agric. Soc. Trinidad & Tobago. 81: 50-58. MARSH, P. M. 1979. Family Braconidae, pp. 144-313 in K. V. Krombein et al. [eds.]. Catalog of Hymenoptera in America north of Mexico. Smithsonian Institution Press, Washington, DC, Vol. 1, 1198 p. PARASRAM, S. 1973. The pepper flower bud moth in the Caribbean - an evaluation. Proc. 11th Ann. Meeting Caribbean Food Crops Soc., Barbados, pp. 466-470. POLOVNY´ , D. 1967. Genitalia of some Neartic and Neotropical members of the tribe Gnorimoschemini, Lepidoptera, Gelechiidae. Acta Entomol. Mus. Prague. 97: 51-67. SCHUSTER, M. F. 1960. A pepper flower-bud moth Gnorimoschema gudmanella. Jour. Econ. Entomol. 52: 117-118. WILSON, C. E. 1923. Truck-crop pests in the Virgin Islands and methods of combatting them. Bull. Virgin Island (U.S.) Agric. Exp. Sta. 4: 24-25.

♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦

Scientiﬁc Notes 549

EFFECT OF TIME OF BAIT EXPOSURE ON NUMBER OF WIREWORMS (COLEOPTERA: ELATERIDAE) FOUND AT BAITS

RONALD H. CHERRY AND JOSE ALVAREZ Everglades Research and Education Center P. O. Box 8003 Belle Glade, Florida 33430

In Florida, the Everglades Agricultural Area (EAA) encompasses 260,000 hectares (650,000 acres) of the upper Everglades extending from Lake Okeechobee south about 50 km. Most of the soils in the EAA are organic (Histosols) generally containing 85% or more organic matter by weight (Snyder et al. 1978). The area is intensively farmed, with sugarcane and winter vegetables being the principal crops. Wireworms are ubiquitous soil insect pests of most of these crops and soil insecticides are frequently applied at planting for wireworm control. Cherry (1993) showed

550 Florida Entomologist 78(3) September, 1995 that rolled oats was a simple, attractive bait that could be used for sampling Melan- otus communis (Gyllenhal) and Conoderus sp. wireworms in the organic soils of the EAA. In that study, baits were left in fallow fields for 14 d. However, the effect of time of bait exposure on wireworm numbers found at baits was not determined. Moreover, the effect of time of bait exposure on wireworm numbers found at baits has received little attention in general. Toba & Turner (1983) reported on tests in Washington State in which wireworm baits were left in fields for one, two, or three weeks. Results were variable with significant differences in wireworm numbers, primarily Ctenicera pruinina (Horn), at baits with different time exposures in some tests and not in other tests. Seal et al. (1992) reported on the effect of the postemergence period of bait exposure on wireworm numbers in Georgia. Baits removed one week after emergence contained significantly more wireworms than similar baits removed 2 and 3 weeks after emergence in fields dominated by Conoderus scissus Schaeffer and C. amplicollis (Gyllenhal). In a field dominated by C. rudis (Brown), more wireworms were found following increased exposure of the bait. The objective of our study was to determine the effect of time of bait exposure on numbers of wireworms found at baits under the field conditions of the EAA. All tests were conducted on Histosols at the Everglades Research and Education Center at Belle Glade, Florida. Tests were conducted in 10 fallow fields which had been disked following sugarcane. Fields were surveyed by digging to determine the presence of wireworms before tests were conducted. Each bait consisted of 200 g of rolled oats placed in a plastic bag. Four baits plus two controls (= no bait) were in each replication in a 3 × 2 pattern with baits and controls 5 m apart. Ten replications were used in each field with replications being 10 m apart. A randomized complete block design was used for placement of the baits and controls. Holes were dug 15 cm deep, each bait poured from its plastic bag into the hole, a flag placed through the bait to mark for recovery, and the bait covered with soil. At the time of bait placement in the field, one unbaited control sample from each replication was dug-up to represent 0 days after bait placement. Thereafter, one bait in each replication was dug-up at 7- day intervals up to 28 days after bait placement in the field. The second unbaited control sample was also dug-up at 28 days. By comparing the two control samples (0 vs. 28 d), it was possible to determine if wireworm population density changed in the plots during the tests. If the wireworm population density did not change in the plots, this would indicate that wireworm populations at baits were due to wireworm attraction to baits over time and not changes in population levels in the plots over time. Baits were recovered by digging-up the bait and adjacent soil in a 25 × 25 × 20 cm deep sample and placing the sample in a bucket. Samples were stored in a laboratory at about 23°C. Each sample was visually examined for wireworms for 30 min in the laboratory. Wireworms were then stored in alcohol and later identified by microscopic examination. M. communis (Gyllenhal) was identified using the key of Riley & Keaster (1979). Conoderus sp. was identified by J. B. Heppner at the Florida Division of Plant Industry, Gainesville. Other wireworm species were less than 10% of total wireworms found at baits and were not identified. Ten tests were conducted in 10 different fields from October, 1992 to September, 1993. Since greater than 90% of the wireworms found at food baits were either Con- oderus sp. or M. communis, statistical analysis was restricted to these two groups. A paired t-test analysis was conducted to determine if wireworm numbers were significantly different in the controls in any of the 10 fields at 0 d versus 28 d. The main objective of our tests was to determine the overall response of wireworms to baits over time. Hence, data from the 10 fields were pooled for regression analysis. Initially, the total number of wireworms at baits was plotted against time (0, 7, 14, 21, 28 d). Visual

Scientiﬁc Notes 551

TABLE 1. REGRESSION RESULTS FROM THE ESTIMATED EQUATIONS.a

Equation

Item M. communis Conoderus sp.

Intercept -0.11 16.21 X 6.65b (0.46) -4.24 (9.70) X2 -0.15c (0.02) 1.08 (0.88) X3 — -0.03 (0.02) R2 0.99 0.92 C.V. 5.96 38.97

aFigures in parentheses are the standard errors of the estimated coefficients. bStatistically significant at P<0.005. cStatistically significant at P<0.01. observation indicated a quadratic function for M. communis and a cubic function for Conoderus sp. Thereafter, these models were used for the two wireworm groups using the General Linear Models Procedure (SAS 1990). The quadratic equation for M. communis was:

2 Y = ao + a1 Xi + a2 X i; where Y = Total number of wireworms at baits

Xi = Days baits left in ﬁeld i = 0, 7, 14, 21, and 28 days

ao, a1, and a2 = the intercept, linear term, and the quadratic term, respectively. The cubic equation for Conoderus sp. was: 2 3 Y = ao + a1 Xi + a2 X i + a3 X i; where a3 = the cubic term; all other terms were previously explained. A t-test analysis showed no significant difference (P=0.05) in mean numbers of Conoderus sp. or M. communis at 0 vs 28 d in unbaited samples in any of the 10 fields. These data indicate that wireworms at baits over time were due to bait attraction and not changes in population densities during the tests. Regression results from the estimated equations are shown in Table 1. The equation for M. communis is highly predictive of the actual observed wireworm distribution over time (Fig. 1). Both linear (X) and quadratic (X2) terms are highly significant statistically; furthermore, the coefficient of determination is a high 0.9954 and the coefficient of variation is a low 5.96. None of the coefficients from the equation for Con- oderus sp. are statistically significant. However, the equation is very close to the actual distribution of wireworms in the fields (Fig. 2), as shown by the high coefficient of determination (0.9196), but not by the coefficient of variation (39). Both equations show that time of bait exposure was a strong predictor of wireworm numbers found at rolled oats baits. Currently, soil insecticides are frequently applied for wireworm control when various crops are planted in the EAA. In many cases, soil insecticides are not needed since wireworm populations are too low to cause economic damage (Cherry et al. 1993). However, few growers sample for wireworms since this procedure is difficult due to digging, sorting through soil, and low numbers of wireworms normally found. Cherry (1993) and this study have shown that rolled oats baits are attractive to wire-

552 Florida Entomologist 78(3) September, 1995

Figure 1. Regression equation for M. communis. Y = the total number of wireworms at baits and X = the duration of bait exposure in days. worms in the EAA and that the time of bait exposure affects the number of wireworms found at baits. These latter two studies provide basic data for using rolled oats baits as a sampling tool to determine the necessity of applying insecticides for wireworm control when planting various crops in the EAA. Florida Agricultural Journal Series No. R-04466.

Figure 2. Regression equation for Conoderus sp. Y = the total number of wireworms at baits and X = the duration of bait exposure in days.

Scientiﬁc Notes 553

SUMMARY

Our data show that the duration of exposure of rolled oats baits is important in determining the numbers of wireworms found at baits under ﬁeld conditions. Wire- worms of both Conoderus sp. and M. communis were found in increasing numbers from 0 to 21 days and then declined in number at 28 days.

REFERENCES CITED

CHERRY, R. 1993. Baits for sampling wireworms (Coleoptera:Elateridae) in organic soils (Histosols) of southern Florida. J. Agric. Entomol. 10: 185-188. CHERRY, R., G. POWELL, AND M. ULLOA. 1993. Reduced soil insecticide use in sugarcane planted after rice. Sugar y Azucar. 88: 37-38. RILEY, T. J., AND A. J. KEASTER. 1979. Wireworms associated with corn: identiﬁcation of larvae of nine species of Melanotus from the North Central States. Annals Entomol. Soc. America 72: 408-414. SAS LANGUAGE. 1990. Version 6. First Ed. Cary, North Carolina. 1042 pages. SEAL, D., R. CHALFANT, AND M. HALL. 1992. Effectiveness of different seed baits and baiting methods for wireworms (Coleoptera: Elateridae) in sweetpotato. Envi- ron. Entomol. 21: 957-963. SNYDER, G., H. BURDINE, J. CROCKET, G. GASCHO, D. HARRISON, G. KIDDER, J. MISHOE, D. MYHRE, F. PATE, AND S. SHIH. 1978. Water table management for organic soil conservation and crop production in the Florida Everglades. Univ. of Florida, Agricultural Experiment Station Bulletin 801, Gainesville, 63 pp. TOBA, H., AND J. TURNER. 1983. Evaluation of baiting techniques for sampling wireworms (Coleoptera: Elateridae) infesting wheat in Washington. J. Econ. Ento- mol. 76: 850-855.

♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦♦

Scientiﬁc Notes 553

SOME CORRECTIONS TO THE GENERIC RECORDS OF CENTRAL AMERICAN EUMOLPINAE (CHRYSOMELIDAE: COLEOPTERA)

R. WILLS FLOWERS1 1Agricultural Research Programs, Florida A&M University, Tallahassee, FL 32307

Seeno & Wilcox (1982) list 56 genera of Eumolpinae (Coleoptera: Chrysomelidae) as occurring in Central America. As part of an ongoing project on the Eumolpinae of Costa Rica, I have found that eight of these generic records are probably incorrect since they are based on species that have been transferred to other genera or are based on very doubtful records. Since study and identiﬁcation of Central American Eumolpinae is difﬁcult enough without worrying about genera that do not in fact occur there, I feel it worth while to review these questionable records.

Agbalus Chapuis 1874:242. Type species Agbalus sericeus Chapuis by monotypy. Valid genus, South America. Hylax Lefèvre 1884:xlv. Type species Amasis chalcaratus Chapuis by monotypy. Valid genus, Central and South America. Bechyné (1950a) restricted this genus to its type species, A. sericeus Chapuis from Brazil, transferring all other species to Hylax Lefèvre. Those other species, including

554 Florida Entomologist 78(3) September, 1995 chalybaeus chiriquiensis Jacoby, hoegei Jacoby, mexicanus Jacoby, puncticollis Jacoby, quadriplagiatus Jacoby, tenebrosus Jacoby, and viridis Bowditch from Central Amer- ica are listed under Hylax in the Bechyné (1953) catalog of the Eumolpinae.

Chalcoplacis Chevrolat 1837:433. Type species Colaspis fulgurans Klug by monotypy. [C. incerta Dejean, designated as type species by Chevrolat 1845, is a nomen nudum. Subsequent designation of Lamprosphaerus abdominalis Baly as type species by Baly 1875 is unnecessary and invalid.] Valid genus, South America. Parachalcoplacis Bechyné 1950a:268. Type species Lamprosphaerus mrazi Bechyné by original designation. Synonym of Chalcoplacis (Monrós & Bechyné 1956). Lamprosphaerus Baly 1859:124. Type species Lamprosphaerus abdominalis Baly by original designation. Valid genus, Central and South America. Antitypona Weise 1921:17. Type species Lamprosphaerus collaris Baly by original designation. Valid genus, Central and South America. Neochalcoplacis Bechyné 1950a:268. Type species Chalcoplacis dimidiata Lefèvre by original designation. Valid genus, South America. Chalcoplacis is in the center of a very tangled nomenclatural web. Baly began it by designating the species Lamprosphaerus abdominalis Baly as the type species first for his genus Lamprosphaerus (Baly 1859), then for Chevrolat’s Chalcoplacis (Baly 1865). Weise (1921) resolved this problem by restricting the definition of Chalcoplacis to Chalcoplacis incerta Chevrolat, the species designated as type species by Chevrolat (1845), and establishing the genus Antitypona with Baly’s Lamprosphaerus collaris as the type species. Weise also stipulated that all Lamprosphaerus then listed in the Coleopterorum Catalogus (Clavareau 1914) are Antitypona and all Chalcoplacis in that catalog are Lamprosphaerus. Bechyné (1950a) erected a new genus Neochalcoplacis, designating Chalcoplacis dimidiata Lefèvre as the type species, including two species from Central America [N. constituta Bechyné and N. fulvipes (Jacoby)], and erected the subgenus Parachalco- placis for a South American species. In the Bechyné’s 1953 catalog, Parachalcoplacis became a valid genus including numerous South American species described by Bechyné or transferred by him from Antitypona and Lamprosphaerus. However, both of the Central American species (constituta and fulvipes) were transferred to Lam- prosphaerus. Neochalcoplacis was now restricted to its type species. Monrós & Bechyné (1956) identified Colaspis fulgurans Klug as the type species of Chalcoplacis (by monotypy), stated that Chalcoplacis incerta Chevrolat was a nomen nudum, and synonymized Parachalcoplacis with Chalcoplacis (but they did not syn- onymize Neochalcoplacis, as stated by Seeno & Wilcox 1982). Colaspis fulgurans had previously been designated as the type species of Parachalcoplacis (Bechyné 1950a). Thus, both Chalcoplacis and Neochalcoplacis are valid genera but neither is known from Central America at present. On the other hand, Lamprosphaerus and Antity- pona are valid and widespread Central American genera.

Corysthea Baly 1865:336. Type species Corysthea ferox Baly by monotypy. Valid genus, South America. The single Central American species, Corysthea violacea described by Jacoby (1882), was transferred to Hylax by Bechyné (1953) and renamed pseudoviolaceus (Bechyné 1955) since the speciﬁc epithet violaceus was preoccupied by Hylax auratus violaceus (Jacoby). No species of Corysthea are presently known from Central Amer- ica.

Coytiera Lefèvre 1875:116. Type species Coytiera marginicollis Lefèvre by monotypy. Valid genus, South America.

Scientiﬁc Notes 555

Euphrytus Jacoby 1881:124. Type species Euphrytus aeneus Jacoby, designated by Bechyné (1950a). Valid genus, Central America. Bechyné (1953) transferred Colaspis melancholica Jacoby to this genus. In 1957, he transferred three other Jacoby Coytiera species, costata Jacoby, fulvipes Jacoby, and rugipennis Jacoby to the genus Euphrytus Jacoby, thus leaving melancholica as the only valid Central American Coytiera. In 1969, Bechyné & Bechyné again moved melancholica, this time to the genus Freudeita Bechyné, based, in their opinion, on the structure of the antenna (see below for further discussion of this species). No valid species of Coytiera are presently known to exist in Central America.

Freudeita Bechyné 1950b:241. Type species Colaspis parallina Erichson, by original designation. Valid genus, South America. Bechyné (1953) and Wilcox (1985) list the species F. dentifera Bechyné from Costa Rica. Bechyné & Bechyné (1969) also transferred Coytiera melancholica (Jacoby) and Colaspis balyi Jacoby to Freudeita, and noted the similarity of both species to F. dentifera. Blake (1976) treated melancholica, balyi, and six closely related species as a species complex within Colaspis. She ignored (as was her custom) all Bechyné’s work and mentioned neither F. dentifera nor other South American Freudeita that Bechyné described as similar to dentifera. In Bechyné’s (1950b) original description of Freude- ita, the deﬁning character is described as antennae with middle segments noticeably wider than either basal or apical segments. This character is readily apparent in the South American species originally placed in Freudeita (Bechyné 1950b) but is very weakly developed in melancholica—as the Bechynés themselves admitted. Although Blake’s across-the-board dismissal of everything Bechyné wrote was, on occasion, excessive and incorrect, I agree in this case that the melancholica complex is better left in Colaspis. Based on published descriptions, dentifera is close to C. melancholica and C. spinigera Blake (all three have a subapical spine on the hind tibia of the male); ac- cordingly Bechyné’s species is now Colaspis dentifera (Bechyné) new combination.

Ischyrolamprina Bechyné 1950a:265. Type species Spintherophyta lampros Lefèvre by original designation. Valid genus, South America. Ischyrolampra Lefèvre 1885:166 [Name for Eulampra Baly which was preoccupied] Type species Eulampra batesi Baly by monotypy. Valid genus, South America. Agrosterna Harold 1875:103. Type species Agrosterna buphthalma Harold by monotypy. Valid genus, Central and South America. This was originally described as a subgenus of Ischyrolampra Lefèvre, which included I. panamensis Jacoby from Panamá. In his catalog, Bechyné (1953) transferred I. panamensis to the genus Agrosterna Harold. Wilcox (1983) apparently missed this change and listed panamensis as Ischyrolampra, and this was perpetuated in the Seeno & Wilcox list. All valid species now included in both Ischyrolamprina and Is- chyrolampra are South American.

Longeumolpus Springlová 1960:5. Type species Eumolpus imperalis Baly, by original designation. Valid genus, South America, Martinique. I have found no records from continental Central America but Springlová (1960), in her revision of the genus, lists the type species, L. imperalis (Baly), as occurring on Martinique in the Lesser Antilles. To be consistent with other listings in Seeno & Wil- cox (1982), this record should be considered “W. Indies”.

Lycaste Gistel 1848:123. Type species Lycaste trichoa Gistel by monotypy. Callicolaspis Bechyné 1950a:275. Junior synonym. Bechyné (1950a) mentioned a specimen of Lycaste cuneiformis (Bechyné) in the Frey Museum from Mexico, a record he considered erroneous for this otherwise South American genus. This appears to be the only published Central American record. As 556 Florida Entomologist 78(3) September, 1995

Lycaste species are among the largest and showiest of the Neotropical Eumolpinae, it is not likely that collectors in Mexico have been missing them all these years.

SUMMARY

Eumolpinae genera Agbalus, Chalcoplacis, Corysthea, Coytiera, Freudeita, Ischy- rolamprina, Longeumolpus, and Lycaste do not occur in Central America, contrary to previous records in the literature. Current synonymies and known ranges for each of these genera are given.

REFERENCES CITED

BALY, J. S. 1859. Descriptions of new genera and species of phytophagous insects. Ann. Mag. Nat. Hist., ser. 3 4:55-61, 124-128, 270-275. BALY, J. S. 1865. Descriptions of new genera and species of Phytophaga. Trans. Ento- mol. Soc. London, ser. 3 2:333-357. BECHYNÉ, J. 1950a. Les générotypes des Eumolpides de l’Amérique du Sud et Centre avec les diagnoses des formes nouvelles (Col. Phytoph. Chrysomeloidea). Mitt. München Entomol. Ges. 40: 264-292. BECHYNÉ, J. 1950b. Notes sur les Chrysomeloidea de l’Amérique du Sud et du Centre. (Col. Phyotph.). Entomol. Arb. Mus. G. Frey 1:237-269. BECHYNÉ, J. 1953. Katalog der neotropischen Eumolpiden (Col. Phytoph. Chrysome- loidea). Entomol. Arb. Mus. G. Frey, 4:26-304. BECHYNÉ, J. 1955. Reise des Herrn G. Frey in Südamerika: Eumolpidae (Col. Phy- tophaga). Entomol. Arb. Mus. G. Frey 6:569-657. BECHYNÉ, J. 1957. Eumolpides Neotropicaux de la collection du Museo Civico de Sto- ria Naturale “Giacomo Doria” di Genova. Ann. Mus. Genova 69:226-247. BECHYNÉ, J. B., AND B. SPRINGLOVÁ DE BECHYNÉ, J. 1969. Notas sobre Phytophaga Americanos (Coleoptera). Rev. Fac. Agron. (Maracay). 3:5-64. BLAKE, D. H. 1976. Colaspis melancholica Jacoby and its close relatives (Coleoptera: Chrysomelidae). J. Washington Acad. Sci. 65: 158-162. CHAPUIS, F. 1874. In Histoire Naturelle des Insectes. Genera des Coléoptères. T. La- cordaire and F. Chapuis, eds. 10:1-455. CHEVROLAT, L. A. A. 1837. In Dejean, P. F. M. A., Catalogue des coléoptères de la collection de M. le comte Dejean. livr. 5, pp. 361-443. CHEVROLAT, L. A. A. 1845. In d’Orbigny, C. 1841-1849. Dictionnaire universel d’His- toire Naturelle, Paris, 5:1-792. CLAVAREAU, H. 1914. Coleopterorum Catalogus, Chrysomelidae: Eumolpinae. pars 59(11):1-215. GISTEL, J. 1848. Naturgeschichte des Thierreichs für höhere Schulen, 216pp. HAROLD, E. VON. 1875. Über Chrysomelidae aus Cordova. Coleopterologische Hefte XIV:95-106. JACOBY, M. 1881. Biologia Centrali-Americana, Insecta, Coleoptera, Eumolpidae, Chrysomelidae. 6:73-144. JACOBY, M. 1882. Biologia Centrali-Americana, Insecta, Coleoptera, Eumolpidae, Chrysomelidae. 6:145-224. LEFÈVRE, E. 1875. Descriptions d’eumolpides nouveaux ou peu connus. Revue et Ma- gasin de Zoologie, ser. 3:102-138. LEFÈVRE, E. 1885. Eumolpidarum hucusque cognitarum catalogus, sectionum con- spectu systematico, generum sicut et specierum nonnullarum novarum de- scriptionibus adjunctis. Mém. Soc. roy. Sci. Liège, ser. 2, 11:1-172. MONRÓS, F., AND J. BECHYNÉ. 1956. Über einige verkannte Chrysomeliden-Namen. Entomol. Arb. Mus. G. Frey 7:1118-1137. SEENO, T. N., AND J. A. WILCOX. 1982. Leaf beetle genera (Coleoptera: Chrysomel- idae). Entomography 1:1-221. Scientiﬁc Notes 557

SPRINGLOVÁ, B. 1960. Essai monographique du genre Eumolpus. Mem. Inst. Roy. Sci. Nat. Belgique 2nd Ser. Fas. 60. 79 pp. WEISE, J. 1921. Wissenshaftliche Ergebnisse der schwedischen entomolgischen Reise des Herrn. Dr. A. Roman in Amazonas 1914-1915. 6. Chrysomelidae. Ark. Zool. 14:1-205. WILCOX, J. A. 1983. Checklist of the beetles of North and Central America and the West Indies. Vol. 8. The leaf beetles and the bean weevils. Family 129. Chry- somelidae. E. J. Brill. New York. 166 pp.

Erratum 633

ERRATUM

L. S. Bauer—Resistance: A Threat to the Insecticidal Crystal Proteins of Bacillus thuringiensis.

Change ﬁrst sentence in conclusions (p. 434) to read: The genetic capacity of insect populations to evolve resistance to Bt δ-endotoxins is now well documented in many species within three different insect orders.