BIOLOGICAL CONTROLÑPARASITOIDS AND PREDATORS Host Ranges of Gregarious Muscoid Fly Parasitoids: Muscidifurax raptorellus (: ), Tachinaephagus zealandicus (Hymenoptera: Encyrtidae), and Trichopria nigra (Hymenoptera: Diapriidae)

1 2 CHRISTOPHER J. GEDEN AND ROGER D. MOON

USDAÐARS, Center for Medical, Agricultural, and Veterinary Entomology, PO Box 14565, Gainesville, FL 32607

Environ. Entomol. 38(3): 700Ð707 (2009) ABSTRACT Attack rates, progeny production, sex ratios, and host utilization efÞciency of Muscidifurax raptorellus (Kogan and Legner) (Hymenoptera: Pteromalidae), Tachinaephagus zealandicus Ashmead (Hymenoptera: Encyrtidae), and Trichopria nigra (Nees) (Hymenoptera: Dia- priidae) were evaluated in laboratory bioassays with Þve dipteran hosts: house ßy (Musca domestica L.), stable ßy (Stomoxys calcitrans L.), horn ßy (Hematobia irritans L.), black dump ßy [Hydrotaea aenescens (Weidemann)] (Diptera: Muscidae), and a ßesh ßy (Sarcophaga bullata Parker) (Diptera: Sarcophagidae). M. raptorellus killed and successfully parasitized all Þve host species and produced an average 2.6 parasitoid progeny from each host. Host attack rates were highest on stable ßy and lowest on horn ßy; there were no differences among hosts in the total number of progeny produced. T. zealandicus killed larvae of all ßy host species in similar numbers, but parasitism was most successful on H. aenescens and S. bullata and least successful on horn ßy and house ßy hosts. SigniÞcantly more parasitoid progeny emerged from S. bullata (10.2 parasitoids per host) than the other hosts; only 2.5 progeny were produced from parasitized horn ßy hosts. Most of the killed puparia that produced neither adult ßies nor parasitoids (“duds”) contained dead parasitoids; in house ßy, stable ßy, and horn ßy hosts, Ͼ30% of these dudded pupae contained adult wasps that failed to eclose. T. nigra successfully parasitized pupae of all host species except house ßy and was most successful on stable ßy. SigniÞcantly more parasitoid progeny emerged from S. bullata (30.6 parasitoids per host) than the other hosts; only 5.7 progeny were produced from horn ßy hosts.

KEY WORDS Muscidifurax raptorellus, Tachinaephagus zealandicus, Trichopria nigra, house ßy, stable ßy

Parasitoids of muscoid ßy immatures have been stud- feed, and lay a single egg between the puparium and ied extensively for many years. (Rutz and Patterson the developing host within. Parasitoid larvae consume 1990). Efforts to use parasitoids for operational ßy the killed host externally within the comparative control have often followed an algorithm involving safety of the puparium. These characteristics place the surveys to determine locally dominant species fol- pteromalid ßy parasitoids within the group described lowed by augmentative releases of that species (Mor- as idiobionts (Askew and Shaw 1986). gan and Patterson 1990; Geden et al. 1992; Weinzierl Gregarious parasitoids received less attention until and Jones 1998; Crespo et al. 1998, 2002; Skovgard and the discovery that the gregarious form of Muscidifurax Nachman 2004). In nearly every case, the parasitoids aptorellus Kogan and Legner, once thought to be na- selected for release have been solitary pupal parasi- tive to Chile, had become established in parts of the toids of the pteromalid genera Muscidifurax and Spal- United States (Kogan and Legner 1970, Legner 1987, angia. Although different species in this group show Petersen and Cawthra 1995, Antolin et al. 1996, Taylor authentic differences with regard to environmental et al. 1997, Taylor and Szalanski 1999). Gregarious M. and behavioral determinants (Geden 1996, 1997, 1999, raptorellus is now available from commercial insecta- 2002), these species differ little in their fundamental ries, and it has shown potential as a biocontrol agent bionomics. All attack pupae of a prescribed age from under a range of production systems and in different a wide range of host species (Geden et al. 2006), host geographic areas (Petersen and Cawthra 1995, Pe- tersen and Currey 1996, Floate et al. 2000, Kaufman et al. 2001, Geden and Hogsette 2006). 1 Corresponding author, e-mail: [email protected]. 2 Department of Entomology, University of Minnesota, 1980 Fol- Another gregarious species that has received re- well Ave., St. Paul, MN 55108. newed attention is the encyrtid larval parasitoid Tachi- June 2009 GEDEN AND MOON:HOST RANGES OF GREGARIOUS FLY PARASITOIDS 701 naephagus zealandicus Ashmead. T. zealandicus is (22Ð23 d) at 25ЊC. Sarcophaga bullata was the most thought to be indigenous to the Southern Hemisphere commonly used host for colony maintenance, but the (Olton 1971). It was imported from Australia and New other species used in this study were used occasionally Zealand and released in California poultry houses in as well. T. nigra were held under the same environ- 1967 (Legner and Olton 1968, Olton and Legner 1974). mental and food conditions as T. zealandicus. Pupae of In subsequent years, this species was recovered in S. bullata or stable ßy were exposed to cages of adult otitid ßies in Ohio (Downing 1975) and from house parasitoids for 3Ð4 d of exposure to oviposition and ßies, calliphorids, and sarcophagids in South Carolina removed and held for ßy and parasitoid emergence (Ables 1977). Although the status of this species in the (26Ð27 d) at 25ЊC. United States at present is unknown, there has been Fly Colonies. House ßies and stable ßies were from considerable interest in T. zealandicus in Brazil, where long-established colonies maintained at the Center for it seems to be an important natural enemy of ßies in Medical Agricultural and Veterinary Entomology poultry houses (Costa 1989; Monteiro and Pires do (CMAVE), Gainesville, FL, and were reared using Prado 2000; Ferreira de Almeida et al. 2002a, b). standard methods and diets (Hogsette 1992). Black During a collecting trip to Russia and Kazakhstan, dump ßies, Hydrotaea aenescens, were from a colony one of the authors (R.D.M.) collected specimens of a established from Florida poultry farms in 1989 and diapriid found attacking stable ßy pupae on dairy reared using the methods described in Hogsette and farms. A colony was established at the USDA labora- Washington (1995). Two-day-old pupae were sepa- tory in Gainesville for further study. This species has rated from rearing media by water ßotation and dried since been identiÞed as Trichopria nigra (Nees), a in a forced air blower designed for this purpose palearctic species that occurs in Russia, Moldavia, Ka- (Bailey 1970). zakhstan, Greece, Germany, Denmark, and Sweden Horn ßies were from a colony established in the (Petrov 2002, Medvedev 1988, Anonymous 2007). In 1970s. Adults were reared in cages at 26ЊC, 60Ð80% preliminary laboratory tests, we found that this was a RH, and constant light and given bovine blood daily by gregarious species that looked promising as a biocon- placing blood-soaked pads on the tops of the cages. trol agent for several species of pest ßies (C.J.G., un- Blood for the colony was collected every 2 wk at a local published data). To our knowledge, the life history, abattoir in 8.5-liter batches and treated with 30 g of host range, and habits of this species are unknown. sodium citrate, 1.5 g of kanamycin sulfate, and 250,000 Tachinaephagus zealandicus and T. nigra are koino- U of nystatin. Eggs were collected daily on water- biont gregarious endoparasitoids whose host range has soaked cotton pads placed 1 cm below the screened not been examined. The objective of this study was to cage bottoms. Larvae were reared by placing 1 ml of examine the performance of the ectoparasitoid M. eggs on a rearing medium composed of 1 liter of steer raptorellus and these two endoparasitic species on a manure, 1 liter of pelletized peanut hulls soaked over- range of host species that are commonly found in night in 750 ml water, and 115 g of a prepared mixture association with production systems. of wheat ßour (44%), Þsh meal (33%), alfalfa meal (18%), and baking soda (5%). The manure was col- lected from pastured and frozen before use to Materials and Methods kill any present. Larval trays were kept in Parasitoid Colonies. The M. raptorellus colony was the same room as the adult ßies, and pupation oc- established from specimens collected on a New York curred on day 5 under these rearing conditions. Pupae poultry farm where commercially produced parasi- were separated from media by water ßotation and toids had been released in previous years. The com- dried in a forced-air blower 7 d after egg placement. mercial source is believed to have originated from The colony of the ßesh ßy, S. bullata Parker, was material collected from Nebraska feedlots. The T. zea- founded with ßies obtained from Carolina Biological landicus colony was established from samples col- Supply (Burlington, NC). Adult ßies were given water lected from a poultry farm in Santa Cruz de Concieri- and moist sugar-yeast cakes. The cakes were prepared cao, Sao Paulo, Brazil. Two T. nigra colonies were by mixing three parts sugar and one part yeast hydro- tested; these had been established from samples col- lysate (by volume) and holding the mixture in pans at lected on dairy farms near Kraznodar, Russia, and 90% RH for 3 d. Flies were presented with fresh beef Almaty, Kazakhstan. All of the colonies were 1Ð2 yr old liver for larviposition. Larvae were reared on beef liver at the time of testing. and pupated Ϸ7 d after larviposition on paper towels Muscidifurax raptorellus were maintained by pro- or vermiculite placed under larval rearing pans. viding parasitoids with 2-d-old house ßy pupae three Samples of 100 pupae of each species were weighed times per week at a host: parasitoid ratio of Ϸ10:1 in for each cohort used in the bioassays and averaged as chambers maintained at 25ЊC and 60Ð80% RH under follows: 106 (S. bullata), 18 (house ßy), 15 (stable ßy constant darkness. T. zealandicus were maintained at and H. aenescens), and 5.5 mg (horn ßy). 25ЊC and 60Ð80% RH under constant light and were Bioassay Procedure. Tests were conducted by ex- provided with water and honey. Mature, wandering posing groups of 100 hosts of a given species to 5 (M. stage host larvae were exposed to 2- to 5-d-old T. raptorellus and T. zealandicus)or10(T. nigra) female zealandicus females for oviposition. The exposed lar- parasitoids (3Ð5 d old) in 30-cm3 cups with screen- vae were removed from the cages after 24 h, allowed topped lids. Tests with M. raptorellus and T. nigra were to pupate, and held for ßy and parasitoid emergence conducted with host pupae. Fly pupae were Ϸ48 h old 702 ENVIRONMENTAL ENTOMOLOGY Vol. 38, no. 3

Table 1. Mean (SE) no. host attacks and progeny production by M. raptorellus using five species of hosts

No. killed No. parasitized No. Percent Progeny per Host Host use hosts hosts progeny females parasitized host Horn ßy 21.2 (3.2)b 10.5 (1.9)b 24.9 (5.2) 45.8 (2.6) 2.4 (0.2) 43.0 (6.6)b House ßy 24.7 (3.7)ab 17.7 (3.6)ab 42.9 (9.1) 40.6 (5.1) 2.4 (0.1) 69.1 (7.9)a Stable ßy 37.9 (5.1)a 21.2 (3.7)a 45.6 (7.3) 39.4 (4.5) 2.2 (0.1) 62.4 (6.8)ab S. bullata 21.1 (2.1)b 9.3 (1.7)b 26.9 (6.1) 45.9 (6.5) 2.9 (0.3) 44.9 (6.3)ab H. aenescens 29.9 (5.3)ab 16.9 (5.3)ab 38.0 (7.9) 54.3 (4.3) 2.3 (0.1) 56.5 (3.7)ab ANOVA F 4.76a 3.50b 2.12c 1.51c 2.23c 3.33b

n ϭ 15 sets of Þve females (Þve separate tests of three sets each) with 100 host pupae of each species for a 24-h exposure. Host use is the percent of killed hosts from which adult parasitoids emerged. Means within columns of the same strain followed by the same letter are not signiÞcantly different at P ϭ 0.05 (TukeyÕs HSD). ANOVA df ϭ 4,70. a P Յ 0.01. a P Յ 0.05. c P Ͼ 0.05. with the exception of S. bullata. Because of the long The long pupal stage of S. bullata made such daily duration of the pupal stage in this species (12 d under pupal age assessments impractical. For this species, our conditions), pupae of S. bullata were tested when three pupal ages were selected (Յ1, 4, and 7 d after they were 4 d old. Tests with T. zealandicus were pupariation), representing newly formed, young, and conducted with mature, wandering stage larvae (Ϸ6d old pupae, respectively. Six sets of 10 parasitoids were after larviposition) in cups containing vermiculite as a used for each host species and pupal age combination. pupation substrate. House ßies were not included because they were not For each test date and host species, sets of 100 pupae successfully parasitized by this species. or larvae with no parasitoids were set up and handled The ability of T. nigra to locate pupae of horn ßy, in the same manner. After exposure to parasitoids, stable ßy, and H. aenescens in natural media was as- hosts were removed and held for ßy emergence. Un- sessed as well. In these tests, 100 third instars of each eclosed pupae were counted after ßy emergence held host were placed in cups of rearing medium and al- for parasitoid emergence, after which emerged para- lowed to pupate. Two days after pupariation, the pu- sitoids were sexed and counted. The number of pupae pae were removed from half of the cups, cleaned, and with evident exit holes was counted to determine the presented to parasitoids in petri dishes for 24 h. For the number of successfully parasitized hosts. This allowed remaining samples, parasitoids were introduced into a determination of the percentage of killed pupae from the cups containing medium and host pupae and held which parasitoid progeny emerged. This value, termed for 24 h. Pupae were removed by hand from the “host use” as described elsewhere (Geden et al. 2006), medium after exposure to parasitoids and held for ßy would be 100 if every killed host produced adult para- and parasitoid emergence. Six sets of 10 females with sitoids. Actual values reßect the efÞciency with which 100 host pupae were tested for each host species and the parasitoids use available hosts, host-feeding events presentation method combination. without oviposition, host rejection after stinging, and Data Analysis. Separate one-way analyses of variance mortality of parasitoid immatures. (ANOVAs) were performed for each parasitoid colony In the tests involving T. zealandicus, those pupae to evaluate the effect of host species on the number of that produced neither ßies nor parasitoids, or “duds” host pupae killed, the number of parasitoid progeny as deÞned by Petersen (1986), were dissected to de- produced, the proportion of females among progeny, the termine their disposition. The dud pupae were clas- number of progeny produced per parasitized host, and siÞed into four groups: those containing no visible sign host use as deÞned in the previous section. The propor- of parasitism and those containing dead larvae, pupae, tions of dud hosts in different categories in tests with T. or adults of T. zealandicus. zealandicus were also analyzed by one-way ANOVA. Three sets of hosts and parasitoids were tested on The effects of pupal age and presentation method on T. each of Þve separate occasions, for a total of 15 ob- nigra attacks were evaluated by separate one-way ANO- servations for each host species and parasitoid com- VAs for each host species. Data representing proportions bination. Host mortality among controls was used for (sex ratio, host utilization, disposition of dud pupae) quality control, and tests were rejected and repeated were subjected to arcsine transformation before analysis if control mortality exceeded 10%. and are presented as percentages in the tables. Data were Effect of Host Age and Presentation on Parasitism analyzed using the GLM Procedure of SAS (SAS Insti- by T. nigra. Because so little is known about T. nigra, tute 1992). two additional experiments were conducted. In the Þrst, sets of 100 pupae of varying ages were exposed to Results groups of 10 females for 24 h using the same methods as described previously. For horn ßy, stable ßy, and H. Muscidifurax raptorellus attacked and produced aenescens, pupal ages tested were Յ1, 2, 3, and 4 d after progeny from all Þve species of hosts (Table 1). Dif- pupariation at 25ЊC. These times represent most of the ferences among hosts were small for all of the char- time between pupariation and adult ßy emergence. acteristics that were measured. SigniÞcantly more sta- June 2009 GEDEN AND MOON:HOST RANGES OF GREGARIOUS FLY PARASITOIDS 703

Table 2. Mean (SE) no. host attacks and progeny production by T. zealandicus using five species of hosts

No. killed No. parasitized Percent Progeny per Host No. progeny Host use hosts hosts females parasitized host Horn ßy 49.9 (4.4)b 1.4 (0.5)d 4.0 (1.6)c 36.8 (7.5) 2.5 (0.3)b 3.3 (0.7)d House ßy 70.9 (3.0)a 9.3 (1.6)c 42.7 (7.5)b 49.0 (6.5) 5.3 (0.9)b 12.3 (1.9)c Stable ßy 62.7 (7.6)ab 15.0 (2.3)b 55.3 (9.3)b 46.2 (3.0) 3.6 (0.2)b 21.4 (3.1)bc S. bullata 68.0 (3.3)ab 16.3 (2.0)b 182.2 (36.1)a 38.2 (5.0) 10.2 (0.7)a 24.7 (2.5)b H. aenescens 50.6 (7.2)b 29.7 (6.1)a 158.5 (31.5)a 43.3 (7.8) 5.1 (0.3)b 55.1 (5.2)a ANOVA F 3.31a 15.02a 13.2a 1.86b 13.12c 49.8a

n ϭ 15 sets of Þve females (Þve separate tests of three sets each) with 100 host pupae of each species for a 24-h exposure. Host use is the percent of killed hosts from which adult parasitoids emerged. Means within columns of the same strain followed by the same letter are not signiÞcantly different at P ϭ 0.05 (TukeyÕs HSD). ANOVA df ϭ 4,70. a P Յ 0.01. b P Ͼ 0.05. c P Յ 0.05. ble ßy (37.9) than horn ßy (21.2) pupae were killed, Host use was signiÞcantly lower with horn ßy (3.3) and signiÞcantly more stable ßy pupae (21.2) were than the other species. successfully parasitized than were horn ßy (10.5) or S. Dissection of dud pupae from the T. zealandicus bullata (9.3). There were no signiÞcant differences tests showed the presence of parasitoids in Ͼ72% of among host species for total progeny production, the the pupae of all species except H. aenescens (Table 3). sex ratio of progeny, or the number of parasitoids About one third of the horn ßy, house ßy, and stable produced per parasitized host (2.2Ð2.9). Host use was ßy pupae contained fully formed adult parasitoids that signiÞcantly higher with house ßy than with horn ßy did not successfully exit the host puparia. Aborted pupae, with 69.1 and 43.0% of the killed hosts produc- parasitism of S. bullata and H. aenescens occurred at an ing adult parasitoids, respectively. earlier stage; Յ10% of the dud pupae of these species Tachinaephagus zealandicus killed larvae of all Þve contained parasitoid adults. host species, with signiÞcantly higher kill rates on The Russian strain of T. nigra killed signiÞcantly house ßy larvae (70.9) than on H. aenescens (50.6) or more stable ßy (57.4), H. aenescens (56.5), and horn ßy horn ßy (49.9); there were no signiÞcant differences (54.2) pupae than house ßy (41.1); lowest kill rates among the nonÐhouse ßy hosts (Table 2). This species were observed with S. bullata (18.2) (Table 4). Suc- successfully parasitized more H. aenescens (29.7) than cessful parasitism was signiÞcantly higher with horn any of the other species, followed by S. bullata and ßy (35.6) and stable ßy (33.9) than with H. aenescens stable ßy (16.3 and 15.0, respectively). Few of the horn (10.0) or S. bullata (4.7). None of the killed house ßy ßy hosts (1.4) were parasitized. Progeny production pupae produced any T. nigra progeny. Progeny pro- was much higher from S. bullata (182.2) and H. aene- duction was highest using stable ßy hosts (257.9), scens (158.5) than from stable ßy (55.3) and house ßy followed by horn ßy (193.3), S. bullata (141.6), and H. (42.7). An average of only four parasitoid progeny aenescens (73.7). There were no differences in sex emerged from horn ßy hosts. There were no differ- ratios among the progeny that emerged from the dif- ences in sex ratios among the progeny that emerged ferent host species. Parasitoid load was signiÞcantly from the different host species. Parasitoid load was higher in S. bullata (30.6 parasitoid progeny per par- signiÞcantly higher in S. bullata (10.2 progeny per asitized host) than the other four species, which did parasitized host) than the other four species, which not differ signiÞcantly from each other. Approxi- did not differ signiÞcantly from each other. Host use mately 60% of the horn ßy and stable ßy pupae that was most efÞcient with H. aenescens, with 55.1% of the were killed by T. nigra produced adult progeny; sig- killed hosts producing parasitoid progeny, followed by niÞcantly lower host use rates were observed with S. S. bullata (24.7), stable ßy (21.4), and house ßy (12.3). bullata (39.3) and H. aenescens 16.6). Nearly identical

Table 3. Results of dissections of host pupae killed by T. zealandicus from which neither adult flies nor parasitoids emerged (“dudded pupae”)

Percent of dudded pupae with evidence of parasitisma No. dudded Host pupaea Dead larvae Dead pupae Dead adults Any para stages (DL) (DP) (DA) (DL ϩ DP ϩ DA) Horn ßy 48.5 (4.2)b 20.4 (2.6)b 18.5 (2.3)b 36.1 (4.1)a 75.0 (4.0)a House ßy 61.6 (2.5)a 28.8 (2.8)ab 18.3 (2.0)b 32.4 (3.3)a 79.5 (2.9)a Stable ßy 47.7 (5.7)b 22.6 (3.3)b 17.3 (2.2)b 32.3 (6.5)a 72.2 (6.9)a S. bullata 51.7 (2.8)b 43.7 (4.4)a 27.6 (4.9)a 6.5 (1.4)b 77.8 (3.8)a H. aeneascens 20.9 (3.4)c 26.8 (4.9)b 13.7 (2.8)b 10.0 (2.0)b 50.5 (5.8)b ANOVA F 4.27b 7.90b 2.89c 19.84b 7.19b

a Mean (SE). b P Յ 0.01. c P Յ 0.05. 704 ENVIRONMENTAL ENTOMOLOGY Vol. 38, no. 3

Table 4. Mean (SE) no. host attacks and progeny production by T. nigra using five species of hosts

No. killed No. parasitized Progeny per Host No. progeny Percent females Host use hosts hosts parasitized host Kazakhstan strain Horn ßy 54.2 (3.6)a 35.6 (3.0)a 193.3 (15.9)b 59.0 (2.6) 5.7 (0.4)b 63.8 (2.7)a House ßy 41.1 (2.8)b 0.0 (0.0)c 0.0 (0.0)d Ñ Ñ Ñ Stable ßy 57.4 (3.4)a 33.9 (2.9)a 257.9 (22.4)a 47.4 (5.9) 8.2 (0.7)b 57.9 (3.6)a S. bullata 18.2 (2.4)c 4.7 (0.8)bc 141.6 (25.2)b 50.6 (4.0) 30.6 (4.4)a 39.3 (5.0)b H. aenescens 56.5 (5.9)a 10.0 (1.7)b 73.7 (12.3)c 52.8 (5.1) 7.5 (0.7)b 16.6 (1.8)c ANOVA F 24.45a 60.47a 28.70a 1.86b 29.76a Russian strain Horn ßy 56.4 (3.0)a 36.6 (2.4)a 179.9 (14.7)b 52.9 (2.9) 5.0 (0.3)b 64.2 (1.8)a House ßy 40.8 (3.2)b 0.0 (0.0)c 0.0 (0.0)d Ñ Ñ Ñ Stable ßy 56.2 (4.6)a 37.6 (3.0)a 262.2 (18.7)a 46.4 (4.4) 7.4 (0.5)b 76.5 (11.7)a S. bullata 17.8 (2.0)c 6.2 (0.9)bc 171.1 (26.3)b 52.1 (4.1) 27.9 (2.1)a 34.8 (4.8)b H. aenescens 63.3 (4.6)a 11.5 (2.0)b 89.7 (15.8)c 45.4 (4.8) 8.9 (1.2)b 17.3 (2.3)c ANOVA F 32.39a 85.86a 30.13a 1.85b 83.59a 28.53a

n ϭ 15 sets of 10 females (Þve separate tests of three sets each) with 100 host pupae of each species for a 24-h exposure. Host use is the percent of killed hosts from which adult parasitoids emerged. Means within columns of the same strain followed by the same letter are not signiÞcantly different at P ϭ 0.05 (TukeyÕs HSD). ANOVA df ϭ 4,70. a P Յ 0.01. b P Ͼ 0.05. results were observed with the Kazakhstan strain of successful parasitism and progeny production over the this parasitoid (Table 4). four pupal age classes. S. bullata were more susceptible Horn ßy pupae were most susceptible to attack and to attack and parasitism when they were 4 d old than parasitism by T. nigra when they were Յ1 d old (Table when they were either Ͻ1 or 7 d old. Older S. bullata 5). Similarly, attack rates and parasitism of H. aenescens pupae also produced a lower proportion of females were higher with younger than older pupae, and a than did 4-d-old pupae. signiÞcantly higher proportion of females was pro- Trichopria nigra that had to search through rearing duced from younger host pupae. Results with stable ßy medium to locate horn ßy pupae killed and parasitized were comparatively uniform, with little difference in more hosts than did parasitoids that were provided

Table 5. Mean (SE) no. host attacks and progeny production by T. nigra on host pupae of different ages

Host Progeny per species, No. killed hosts No. parasitized hosts No. progeny Percent females Host use parasitized host pupal agea Horn ßy Յ1 61.7 (4.3)a 39.3 (3.3)a 210.7 (13.1)a 51.7 (2.4)ab 5.4 (0.3) 63.6 (2.5) 2 27.2 (4.5)b 14.8 (3.2)b 72.8 (15.7)b 49.3 (5.5)b 4.8 (0.4) 53.5 (7.0) 3 32.0 (5.5)b 16.7 (0.9)b 72.7 (4.4)b 56.7 (2.5)ab 4.4 (0.1) 61.1 (10.4) 4 15.3 (2.2)b 10.7 (1.8)b 45.2 (8.8)b 64.5 (1.5)a 4.3 (0.5) 68.0 (4.2) ANOVA F 20.99b 26.12b 43.54b 4.09c 2.41d 0.81d Stable ßy Յ1 71.4 (4.0)a 40.6 (4.5)a 291.6 (9.5)a 28.7 (6.9) 7.5 (6.9)b 56.8 (4.8)a 2 56.8 (2.3)b 18.8 (1.6)b 199.2 (19.9)b 39.9 (4.2) 10.8 (1.4)a 32.9 (1.8)b 3 52.0 (2.8)b 31.0 (2.4)ab 255.2 (20.8)ab 42.4 (4.2) 8.2 (0.3)ab 59.5 (2.8)a 4 47.0 (4.4)b 31.1 (3.7)ab 225.6 (22.3)ab 50.4 (7.7) 7.4 (0.4)b 65.4 (3.6)a ANOVA F 8.11b 7.58b 4.45c 2.29d 3.74c 17.14b H. aenescens Յ1 70.3 (7.0)a 19.5 (2.4)a 154.8 (15.2)a 61.0 (6.2)a 8.2 (0.9) 28.3 (3.3)a 2 65.5 (2.5)ab 12.2 (2.2)b 88.7 (24.3)b 59.8 (4.0)a 7.3 (1.4) 18.6 (3.1)ab 3 65.2 (4.1)ab 6.2 (0.8)b 45.2 (6.1)b 59.2 (3.8)a 7.4 (0.5) 9.8 (1.6)b 4 52.5 (1.1)b 5.3 (1.2)b 50.2 (12.0)b 36.7 (7.7)b 9.3 (0.6) 10.1 (2.2)b ANOVA F 3.18c 13.51b 10.22b 4.81c 1.08d 10.92b S. bullata Յ1 13.6 (0.9)a 3.8 (0.4)b 215.8 (36.1)b 61.4 (4.8)ab 58.3 (6.4) 28.9 (3.6)b 2 15.2 (1.2)ab 9.1 (1.0)a 419.2 (49.9)a 76.3 (5.1)a 50.6 (5.1) 62.7 (6.1)a 3 10.8 (0.7)b 1.8 (0.4)b 65.9 (12.9)c 44.4 (5.0)b 42.0 (6.2) 16.2 (2.2)b 4 5.50b 31.03b 23.82b 9.76b 1.84d 29.64b

n ϭ 6 sets of 10 females with 100 host pupae of each species of each age for a 24-h exposure. Host use is the percent of killed hosts from which adult parasitoids emerged. Means within columns of the same strain followed by the same letter are not signiÞcantly different at P ϭ 0.05 (TukeyÕs HSD). ANOVA df ϭ 3,20 (horn ßy, stable ßy, H. aenescens) or 2,15 (S. bullata). a Days since pupariation. b P Յ 0.01. c P Յ 0.05. d P Ͼ 0.05. June 2009 GEDEN AND MOON:HOST RANGES OF GREGARIOUS FLY PARASITOIDS 705

Table 6. Host attacks by T. nigra on host pupae in rearing medium or in Petri dishes after pupae were removed from media and cleaned

Host species Progeny per No. killed No. parasitized No. and pupal Percent females parasitized Host use hosts hosts progeny exposure host Horn ßy In medium 61.3 (2.7) 48.2 (2.7) 212.7 (12.5) 53.5 (2.9) 4.4 (0.2) 78.5 (3.1) In Petri dish 29.3 (0.9) 19.0 (1.4) 97.2 (9.5) 49.4 (3.0) 5.2 (0.5) 64.5 (3.4) ANOVA F 122.55a 90.23a 54.05a 0.94b 1.79b 9.3c Stable ßy In medium 38.7 (3.1) 26.2 (1.5) 320.0 (20.9) 51.9 (2.1) 12.2 (0.3) 68.4 (2.0) In Petri dish 42.3 (5.2) 24.2 (5.1) 205.3 (26.3) 46.4 (2.7) 9.8 (1.2) 53.5 (6.3) ANOVA F 0.36b 0.14b 11.62c 2.65b 3.58b 5.08c H. aenescens In medium 46.7 (3.0) 17.2 (2.1) 148.5 (21.8) 55.9 (2.3) 8.4 (0.6) 36.3 (3.3) In Petri dish 38.5 (2.6) 14.2 (0.8) 109.2 (13.1) 63.7 (6.0) 8.0 (1.3) 37.2 (2.3) ANOVA F 4.14b 1.83b 2.40b 1.47b 0.10b 0.05b

n ϭ 6 sets of 10 females (Þve separate tests of three sets each) with 100 host pupae of each species for a 24-h exposure. Host use is the percent of killed hosts from which adult parasitoids emerged. ANOVA df ϭ 1,10. a P Յ 0.01. b P Ͼ 0.05. c P Յ 0.05. with pupae in the absence of rearing medium (Table (Kaufman et al. 2001). Despite the importance of this 6). With stable ßies, there was no difference in the species in biocontrol programs, nothing is known numbers of killed or parasitized hosts, but signiÞcantly about host and habitat associations of the gregarious more parasitoid progeny were produced from pupae form of M. raptorellus in its native home range in Chile. in rearing medium than in its absence. There were no Tachinaephagus zealandicus has long been regarded as differences in any of the measured variables for H. a potential biological control agent for house ßy, and it aenescens in the two pupal presentation conditions. was with this intent that it was imported and released on California poultry farms (Olton and Legner 1974). In our tests, however, house ßy was a relatively poor host for Discussion this species compared with S. bullata and H. aenescens. T. In a previous study, we examined the host range of zealandicus killed large numbers of house ßy larvae, but six species of solitary pupal ectoparasitoids, including only 12% of the killed hosts produced adult parasitoids. Muscidifurax raptor (Geden et al. 2006). M. raptor Dissection of house ßy hosts that were dudded by this killed and successfully parasitized somewhat fewer S. species showed the presence of dead parasitoid imma- bullata than the other species, and the dudding rate tures and adults in ratios that were similar to those in was higher with this host than the others. In contrast, more favorable hosts, indicating that encapsulation of M. raptorellus killed and parasitized S. bullata at rates parasitoid eggs or young larvae was not occurring at comparable to all of the other hosts except stable ßy, higher rates in house ßy hosts. which sustained signiÞcantly higher parasitism. More- Tachinaephagus zealandicus in nature is often found over, the total progeny production by M. raptorellus parasitizing calliphorid and sarcophagid hosts (Olton did not differ along host lines. We had expected this and Legner 1974, Costa 1989, Gold and Dahlsten 1989, species to show plasticity with parasitoid loading in Bishop et al. 1996, Bishop 1998, Monteiro and Pires do these hosts of greatly dissimilar sizes. This was not the Prado 2000). Ables (1977) found much higher rates of case. The number of progeny produced from horn ßy parasitism in Phaenecea sp. than in house ßy hosts in pupae was not signiÞcantly different from S. bullata, a South Carolina. House ßies may only be secondary hosts host that is 20 times larger. Harvey et al. (1998) found of this species, but the high attack rates of T. zealandicus that clutch size of this species was typically one to four suggest that augmentative releases could have a sup- eggs and that larger clutch sizes resulted in smaller- pressing effect on house and stable ßy survival in areas bodied adult progeny, an indicator of reduced Þtness. where larval breeding sites can be identiÞed. Lysyk (2004) found rates of progeny production per To our knowledge, nothing is known about the host pupa that were similar to ours in both stable ßy biology of T. nigra. Although there are no published and house ßy hosts. Higher rates were observed only records of this species in the United States, there have when host: parasitoid ratios were Յ10 (Lysyk 2004), been several reports of Trichopria sp. recovered from well below the ratios used in our tests (20 pupae: Þlth ßy pupae during parasitoid surveys (Legner and parasitoid). These results suggest that M. raptorellus Olton 1971, Bradley et al. 1984, Harris and Summerlin shows little ßexibility in adjusting clutch size to ac- 1984, Smith et al. 1987, McKenzie and Richerson commodate differences in host quality. 1991). T. nigra seems to differ from Trichopria sto- Once native to South America, M. raptorellus has moxydis, another gregarious endoparasitoid of stable spread widely since its initial release in California ßies (Morgan et al. 1990) in several important regards. (Petersen and Cawthra 1995, Antolin et al. 1996). This First, stable ßy pupae parasitized by T. stomoxydis process has been accelerated by commercial produc- produced Ϸ13 females per parasitized pupa, whereas tion and release in many parts of the United States we only observed an average of 7.4 progeny of both 706 ENVIRONMENTAL ENTOMOLOGY Vol. 38, no. 3 sexes with T. nigra. Second, Morgan et al. (1993) re- Copenhagen (ZMUC). (http://www.zmuc.dk/EntoWeb/ ported that T. stomoxydis performed best on very collections-databaser/Hymenoptera/Hymenoptera.htm). young stable ßy pupae, whereas T. nigra readily par- Antolin, M. F., D. S. Guertin, and J. J. Petersen. 1996. The asitized hosts over all of the pupal ages that we ex- origin of gregarious Muscidifurax (Hymenoptera: Ptero- amined. Third, T. stomoxydis produced a total of Ϸ10 malidae) in North America: an analysis using molecular progeny per female over her lifetime, whereas groups markers. Biol. Control 6: 76Ð82. Ϸ Askew, R. R., and M. R. Shaw. 1986. Parasitoid communities: of 10 T. nigra females produced 260 progeny (26/ their size, structure and development. pp 225Ð264. In J. female) during a 24-h oviposition test. Trichopria Waage and D. Greathead (eds.), parasitoids. Ac- painteri, a solitary African endoparasitoid of stable ßy, ademic, London, United Kingdom. has even lower fecundity than T. stomxydis and was Bailey, D. L. 1970. Forced air for separating pupae of house considered by Huggert and Morgan (1993) to be of ßies from rearing medium. J. Econ. Entomol. 63: 400Ð405. only minor importance as a natural enemy. Our results Bishop, D. M. 1998. Parasitic Hymenopter reared from suggest that T. nigra is an attractive novel biological dung-breeding Diptera in New Zealand. New Zealand control agent for stable ßy. Further work is needed to Entomol. 21: 99Ð106. assess this more fully and to determine whether it can Bishop, D. M., A.C.G. Heath, and N. A. Haack. 1996. Dis- meet the criteria necessary for release from quarantine. tribution, prevalence and host associations of Hymenop- tera parasitic on Calliphoridae occurring in ßystrike in Trichopria nigra that were forced to seek pupae New Zealand. Med. Vet. Entomol. 10: 365Ð370. distributed naturally in the hostsÕ rearing medium Bradley, S. W., D. C. Booth, and D. C. Sheppard. 1984. Parasit- were at least as effective at parasitizing hosts as those ism of the black soldier ßy by Trichopria sp. (Hymenopter: that were presented with “naked” pupae in assay Diapriidae) in poultry houses. Environ. Entomol. 13: 451Ð454. dishes. In fact, horn ßy pupae in medium were at- Costa, V. A. 1989. Parasito´ides pupais (Hymenoptera, Chal- tacked to a greater degree than when they were pre- cidoidea) de Musca domestica L., 1758, Stomoxys calci- sented without medium. The presence of natural me- trans (L., 1758) e Muscina stabulans (Falle´n, 1816) dia may cause heightened searching behavior of the (Diptera, Muscidae) em avia´rios de Echapora˜, SP. MS females. Another possibility is that pupae that are thesis, ESALQ, Piricicaba, Brazil. anchored in place within medium may be easier for Crespo, D. C., R. E. Lecuona, and J. A. Hogsette. 1998. Bi- ological control: an important component in integrated the parasitoids to handle than pupae placed in petri management of Musca domestica (Diptera: Muscidae) in dishes. Finally, the greater three-dimensional struc- caged-layer poultry houses in Buenos Aires, Argentina. ture of rearing medium may reduce the likelihood of Biol. Control 13: 16Ð24. competitive interactions among host-seeking females. Crespo, D. C., E. E. Lecuona, and J. A. Hogsette. 2002. Strat- In contrast to M. raptorellus, T. zealandicus and T. egies for controlling house ßy populations resistant to nigra seemed to assess host size or quality and make cyromazine. Neotrop. Entomol. 31: 141Ð147. commensurate adjustments in parasitoid load rates. Downing, W. 1975. New occurrence and host for Tachina- Both of the latter species produced four to Þve times ephagus zealandicus Ashmead (Hymenoptera: Encyrti- as many progeny from parasitized S. bullata as from dae). Ohio J. Sci. 75: 62. horn ßy hosts. These results were somewhat surpris- Ferreira de Almeida, M. A., A. Pires do Prado, and C. J. Geden. 2002a. Inßuence of temperature on development time and ing, because we expected that the idiobiont M. rap- longevity of Tachinaephagus zealandicus (Hymenoptera: torellus would have a wider host range and be more Encyrtidae), and effects of nutrition and emergence order adaptable in its response to different hosts than the on longevity. Environ. Entomol. 31: 375Ð380. koinobionts T. zealandicus and T. nigra (Askew and Ferreira de Almeida, M. A., C. J. Geden, and A. Prese do Shaw 1986). With the exception of the T. nigra--house Prado. 2002b. Inßuence of feeding treatment, host den- ßy combinations, all three species attacked and de- sity, temperature and cool storage on attack rates of Tachi- veloped successfully on all of the hosts presented. It naephagus zealandicus. Environ. Entomol. 31: 732Ð738. may be that these parasitoids have specialized in par- Floate, K., P. Coghlin, and G.A.P. Gibson. 2000. Dispersal of ticular habitats such as carrion and dung and are com- the Þlth ßy parasitoid Muscidifurax raptorellus (Hyme- patible with many of the dipterans present in that noptera: Pteromalidae) following mass releases in cattle niche. Additional testing with hosts from other eco- conÞnements. Biol. Control 18: 172Ð178. Geden, C. J. 1996. Modeling host attacks and progeny pro- systems would be necessary to test this hypothesis. duction of Spalangia gemina, S. cameroni, and Muscidi- furax raptor (Hymenoptera: Pteromalidae) at constant and variable temperatures. Biol. Control 7: 172Ð178. Acknowledgments Geden, C. J. 1997. Development models of the Þlth ßy para- We thank H. McKeithen and A. Campbell for assisting sitoids Spalangia gemina, S. cameroni, and Muscidifurax with ßy rearing and bioassays and D. Notton and J. Noyes for raptor (Hymenoptera: Pteromalidae) under constant and identifying T. nigra and T. zealandicus, respectively. variable temperatures. Biol. Control 9: 185Ð192. Geden, C. J. 1999. Host location by house ßy parasitoids in poultry manure at different moisture levels and host den- References Cited sities. Environ. Entomol. 28: 755Ð760. Geden, C. J. 2002. Effect of habitat depth on host location Ables, J. R. 1977. The occurrence of an imported ßy par- by Þve species of parasitoids (Hymenoptera: Pteromali- asite, Tachinaephagus zealandicus Ashmead, in South dae, Chalcididae) of house ßies (Diptera: Muscidae) in Carolina. J. Ga. Entomol. Soc. 12: 114Ð117. three types of substrates. Environ. Entomol. 31: 411Ð417. Anonymous. 2007. Collection inventory of the Hymenop- Geden, C. J., and J. A. Hogsette. 2006. Suppression of house tera collection of the Zoological Museum, University of ßies (Diptera: Muscidae) in Florida poultry houses by June 2009 GEDEN AND MOON:HOST RANGES OF GREGARIOUS FLY PARASITOIDS 707

sustained releases of Muscidifurax raptorellus and Spal- Monteiro, M. R., and A. Pires do Prado. 2000. Trichopria sp. angia cameroni (Hymenoptera: Pteromalidae). Environ. (Hymenoptera: Diapriidae) attacking pupae of Chry- Entomol. 35: 75Ð82. somyia putoria (Wiedemann) (Diptera: Calliphoridae) in Geden, C. J., D. C. Steinkraus, R. W. Miller, and D. A. Rutz. a poultry facility. An. Soc. Entomol. Bras. 1: 159Ð167. 1992. Suppression of house ßies on New York and Mary- Morgan, P. B., and R. S. Patterson. 1990. EfÞciency of target land dairies using Muscidifurax raptor in an integrated formulations of pesticides plus augmentative releases of management program. Environ. Entomol. 21: 1419Ð1426. Spalangia endius Walker (Hymenoptera: Pteromalidae) Geden, C. J., R. D. Moon, and J. F. Butler. 2006. Host ranges to suppress populations of Musca domestica L. (Diptera: of six solitary Þlth ßy parasitoids (Hymenopter: Ptero- Muscidae) at poultry ranches in the southeastern United malidae, Chalcididae) from Florida, Eurasia, Morocco, States, pp. 69Ð78. In D. A. Rutz and R. S. Patterson (eds.), and Brazil. Environ. Entomol. 35: 405Ð412. Biocontrol of arthropods affecting livestock and poultry. Gold, C. S., and D. L. Dahlsten. 1989. Prevalence, habitat Westview, Boulder, CO. selection, and biology of Protocalliphora (Diptera: Calli- Morgan, P. B., J. A. Hogsette, and R. S. Patterson. 1990. Life phoridae) found in nests of mountain and chestnut- history of Trichopria stomoxydis (Hymenoptera: Proc- backed chickadees in California. Hilgardia 57: 1Ð19. totrupidae: Diapriidae) a gregarious endoparasite of Sto- Harris, H. L., and J. W. Summerlin. 1984. Parasites of horn ßy moxys calcitrans from Zimbabwe, Africa. Fla. Entomol. 73: pupae in east central Texas. Southwest. Entomol. 9: 169Ð173. 496Ð502. Harvey, J., L. Vet, N. Jiang, and R. Gols. 1998. Nutritional Olton, G. S. 1971. Bioecological studies of Tachinaephagus ecology of the interaction between larvae of the gregar- zealandicus Ashmead (Hymenoptera, Encyrtidae), para- ious ectoparasitoid, Muscidifurax raptorellus (Hymenop- sitoid of synanthropic Diptera. PhD dissertation, Univer- tera: Pteromalidae), and their pupal host, Musca domes- sity of California, Riverside, CA. tica (Diptera: Muscidae). Physiol. Entomol. 23: 113Ð120. Olton, G. S., and E. F. Legner. 1974. Biology of Tachinaepha- Hogsette, F. A. 1992. New diets for production of house ßies gus zealandicus (Hymenoptera: Encyrtidae), parasitoid of and stable ßies (Diptera: Muscidae) in the laboratory. J. synanthropic Diptera. Can. Entomol. 106: 785Ð800. Econ. Entomol. 85: 2291Ð2294. Petersen, J. J. 1986. Evaluating the impact of pteromalid par- Hogsette, J. A., and F. Washington. 1995. Quantitative mass asites on Þlth ßy populations associated with conÞned live- production of Hydrotaea aenescens (Diptera: Muscidae). stock installations. Misc. Publ. Entomol. Soc. Am. 61: 52Ð56. J. Econ. Entomol. 88: 1238Ð1242. Petersen, J. J., and J. K. Cawthra. 1995. Release of a gregar- ious Muscidifurax species (Hymenoptera: Pteromalidae) Huggert, L., and P. B. Morgan. 1993. Description and biol- for the control of Þlth ßies associated with conÞned beef ogy of Trichopria painteri n. sp. (Hymenoptera: Diapri- cattle. Biol. Control 5: 279Ð284. idae), a Solitary parasitoid of Stomoxys calcitrans Petersen, J. J., and D. M. Currey. 1996. Timing of release of (Diptera: Muscidae) from Harare, Zimbabwe. Med. Vet. gregarious Muscidifurax raptorellus (Hymenoptera:Pte- Entomol. 7: 358Ð362. romalidae) to control ßies associated with conÞned beef Kaufman, P. E., S. J. Long, and R. A. Rutz. 2001. Impact of cattle. J. Agric. Entomol. 13: 55Ð63. exposure length and pupal source on Muscidifurax rap- Petrov, S. 2002. New species of genus Trichopria Ashmead torellus and Nasonia vitripennis (Hymenoptera: Pteroma- 1893 (Hymenoptera: Proctotrupoidea: Diapriidae) to the lidae) parasitism in a New York poultry facility. J. Econ. fauna of Greece. Acta Zool. Bulg. 54: 107Ð109. Entomol. 94: 998Ð1003. Rutz, D. A., and R. S. Patterson (eds.). 1990. Biocontrol of Kogan, M., and E. F. Legner. 1970. A biosystematic revision arthropods affecting livestock and poultry. Westview, Boul- of the genus Muscidifurax (Hymenoptera: Pteromalidae) der, CO. with descriptions of four new species. Can. Entomol. 102: SAS Institute. 1992. SAS users guide: statistics. SAS Insti- 1268Ð1290. tute, Cary, NC. Legner, E. F. 1987. Inheritance of gregarious and solitary ovi- Skovgard, H., and G. Nachman. 2004. Biological control of position in Muscidifurax raptorellus Kohan and Legner (Hy- house ßies Musca domestica and stable ßies Stomoxys menoptera: Pteromalidae). Can. Entomol. 110: 791Ð808. calcitrans (Diptera: Muscidae) by means of inundative Legner, E. F., and G. S. Olton. 1968. Activity of parasites releases of Spalangia cameroni (Hymenoptera: Pteroma- from Diptera: Musca domestica, Stomoxys calcitrans and lidae). Bull. Entomol. Res. 94: 555Ð567. species of Fannia, Muscina, and Ophyra. II. At sites in the Smith, J. P., R. D. Hall, and G. D. Thomas. 1987. Field Eastern Hemisphere and PaciÞc area. Ann. Entomol. Soc. parasitism of the stable ßy (Diptera: Muscidae). Ann. Am. 61: 1306Ð1314. Entomol. Soc. Am. 80: 391Ð397. Legner, E. F., and G. S. Olton. 1971. Distribution and rel- Taylor, D. B., and A. L. Szalanski. 1999. IdentiÞcation of Mus- ative abundance of dipterous pupae and their parasitoids cidifurax spp. by polymerase chain reactionÐrestriction in accumulations of domestic animal manure in the south- fragment length polymorphism. Biol. Control 15: 270Ð273. western United States. Hilgardia 40: 505Ð535. Taylor, D. B., R. D. Peterson, A. L. Szalanski, and J. J. Pe- Lysyk, T. J. 2004. Host mortality and progeny production by tersen. 1997. Mitochondrial DNA variation among Mus- solitary and gregarious parasitoids (Hymenoptera: cidifurax spp. (Hymenoptera: Pteromalidae), pupal para- Pteromalidae) attacking Musca domestica and Stomoxys sitoids of Þlth ßies. Ann. Entomol. Soc. Am. 90: 814Ð824. calcitrans (Diptera: Muscidae) at varying host densities. Weinzierl, R. A., and C. J. Jones. 1998. Releases of Spalangia Environ. Entomol. 33: 328Ð339. nigroaenea and Muscidifurax zaraptor (Hymenoptera: Pte- McKenzie, C. L., and J. J. Richerson. 1993. Parasitoids of the romalidae) increase rates of parasitism and total mortality of horn ßy in rangeland ecosystems of trans-Pecos Texas. stable ßy and house ßy (Diptera: Muscidae) pupae in Illinois Southwest. Entomol. 18: 57Ð59. cattle feedlots. J. Econ. Entomol. 91: 1114Ð1121. Medvedev, G. S. 1988. Key to the of the European parts of the USSR. Oxonian Press, New Delhi, India. Received 22 July 2008; accepted 26 December 2008.