Exhibit 3b, #41 COMMUNITY AND ECOSYSTEM ECOLOGY Nontarget Hymenoptera Collected in Pheromone- and Synthetic Floral Volatile-Baited Traps ROBERT L. MEAGHER, JR. AND EVERETT R. MITCHELL Center for Medical, Agricultural and Veterinary Entomology, USDAÐARS, Gainesville, FL 32604 Environ. Entomol. 28(3): 367Ð371 (1999) ABSTRACT Monitoring traps baited with lepidopteran sex pheromones and synthetic ßoral vola- tiles were used to collect adult Hymenoptera in Þelds of cotton and corn. Species from Apoidea, Pompiloidea, Scolioidea, Sphecoidea, and Vespoidea were collected, including the genera Am- mophila, Apis, Bombus, Cerceris, Larra, Melissodes, Myzinum, and Tachytes. More Bombus spp. were collected from traps baited with Spodoptera frugiperda (J. E. Smith) sex pheromone than those baited with phenylacetaldehyde, whereas more Sphecoidea were collected in phenylacetaldehyde-baited traps. Trap design was also an important factor in capture of various species. More Sphecoidea and Tiphioidea were collected in fabric cone-shaped traps than plastic funnel traps. Efforts should be made to develop traps and lures that consistently capture the target pest but do not attract or easily capture aculeate Hymenoptera to preserve beneÞcial populations. KEY WORDS Bombus, Apis, Larra, Cerceris, nontarget insects VARIOUS TRAP DESIGNS, colors, and lures are used in dopteran pests, our objectives for this research were agricultural systems to monitor for adult lepidopteran to identify Hymenoptera that were collected in traps pests. Many of the traps also capture nontarget insects, and compare these captures among trap designs and such as Coleoptera, Diptera, and Hymenoptera, which lures that may be used for fall armyworm. include numerous beneÞcial insects (Gauthier et al. 1991). Some studies found trap design to be more Materials and Methods important than lure in capture of nontarget Hyme- noptera (Adams et al. 1989, Mitchell et al. 1989), 1997. All white bucket traps (white canopy, funnel, whereas other research has shown trap color, trap and bucket) (International Pheromone Systems, Wir- height above crop canopy, and lure inßuence Hyme- ral, Merseyside, England) were placed in northwest- noptera movement and subsequent capture (Gross ern Alachua County, FL, from 23 July to 10 October, and Carpenter 1991). to capture male fall armyworm. This part of the county Currently, populations of adult male Spodoptera was planted to over 470 ha of cotton, Gossypium hir- spp. are monitored using plastic funnel traps (Univer- sutum L., and offered Þelds separated by paved and sal Moth Traps or “bucket” traps) with a synthetic unpaved roads and forested strips. Traps baited with blend of sex pheromone components as a lure (Mitch- Tre´ce´ (Tre´ce´, Salinas, CA) red septa lures were placed ell et al. 1985, Tumlinson et al. 1986, Mitchell and along pivot roads and edges in an 80-ha Þeld from 18 Tumlinson 1994). However, chemicals other than sex June to 10 October. pheromones are being assayed as moth attractants. For Three treatments were used in the experiment: the instance, ßoral compounds that attract noctuid moths pheromone blend alone, phenylacetaldehyde (Al- have been isolated and identiÞed (Cantelo and Jacob- drich, Milwaukee, WI) in plastic caps (0.5 ml per cap) son 1979, Haynes et al. 1991, Heath et al. 1992). Male alone, or a combination of both lures. Pheromone and female Trichoplusia ni (Hu¨ bner) were attracted to lures were attached to the bottom of a cork that was synthetic phenylacetaldehyde in ßight tunnel, green- placed in a hole in the canopy of the bucket trap. The house, or screen cage bioassays (Haynes et al. 1991, cap with phenylacetaldehyde was hot-gun glued (Ar- Landolt et al. 1991, Heath et al. 1992). Synthetic phe- row Fastener, Saddle Brook, NJ) to the bottom of the nylacetaldehyde was tested in ßight tunnel bioassays cork, which was placed in the trap canopy. The com- with male fall armyworm, S. frugiperda (J. E. Smith), bination lure was composed of a cork with attached and was found to increase upwind ßight and contact cap and the pheromone lure attached to the outside of in combination with a sex pheromone lure (Meagher the cork. All traps contained insecticide strips to kill and Mitchell 1999). This material and other com- insects that were captured (Hercon Vaportape II pounds are being tested in the Þeld as additional at- [Emigsville, PA] containing 10% 2, 2-dichlorovinyl tractants. Because few studies have documented the dimethyl phosphate). Trap contents were removed 3 species and number of nontarget Hymenoptera that times per week and pheromone and phenylacetalde- are captured in traps intended for agricultural lepi- hyde lures were replaced every 2 wk. The experiment 368 ENVIRONMENTAL ENTOMOLOGY Vol. 28, no. 3 was designed as a randomized complete block with 4 Table 1. Species and number of Hymenoptera collected in the replications of the 3 treatments. The location of each 1997 and 1998 experiments, Alachua County, FL trap within each replication was changed weekly. No. collected 1998. In late March, the same area in northwestern Superfamily/Family/Tribe/Genus/Species Alachua County was planted to silage corn, Zea mays 1997 1998 L. An experiment was designed to compare the effect Apoidea of trap design on the collection of Hymenoptera. The Anthophoridae Anthophorinae experiment contained 4 treatments: bucket traps with Melissodes bimaculata (Lepeletier) 1 2 either Tre´ce´ or Scentry (Ecogen, Langhorne, PA) S. Melissodes spp. 1 3 frugiperda sex pheromone lures, and Heliothis “cone” Svastra o. obliqua (Say) 1 0 traps (Ecogen) with either lure. Cone traps are made Xylocopinae Xylocopa virginica (L.) 2 1 of fabric mesh and are designed in the style of hard- UnidentiÞed anthophorids 6 12 ware cloth traps (Hartstack et al. 1979). Heliothis cone Apidae traps are composed of 2 cones: the base cone measures Bombinae 80 cm long, with a bottom opening of 34 cm that Bombus bimaculatus Cresson 1 0 B. fraternus (Smith) 2 1 narrows to 15 cm at the top; the apex cone measures B. impatiens Cresson 0 4 27 cm long, with a bottom opening of 15 cm that B. pennsylvanicus (De Geer) 57 59 narrows to 6 cm at the top. The bottom portion of the Apinae apex cone is secured to the top portion of the base Apis mellifera L. 17 6 Colletidae cone with Velcro material. The lure is placed in the Colletes sp. 0 1 middle of the base cone; insects ßy into this opening Halictidae and up through the 15-cm opening toward the apex Halictinae cone, eventually going through the 6 cm opening and Agapostemon splendens (Lepeletier) 9 84 Pompiloidea being trapped in a fabric container around the apex Pompilidae cone. No insecticide strips were used with the cone Pompilinae traps. Traps were placed in similar locations as in 1997, Anoplius sp. 3 1 and the experiment was designed as a randomized Scolioidea Scoliidae complete block with 4 replications of the 4 treatments. Campsomerinae Trap location within a replication was randomized Campsomeris (Dielis) 12 weekly, and trapping began 8 April and ended 12 June. plumipes fossulana (F.) Individuals were identiÞed by using keys (Stange C. (Pygodasis) quadrimaculata (F.) 3 0 Scoliinae 1992) and by comparing with identiÞed specimens. Scolia (Discolia) n. nobilitata (F.) 0 5 Voucher specimens (FLDA, V01ÐV30) were placed in Sphecoidea the Florida State Collection of Arthropods, Florida Sphecidae Department of Agriculture and Consumer Services, Larrinae Larra bicolor F. 55 45 Division of Plant Industry, Gainesville. Species were L. analis F. 1 0 placed within superfamilies [using the classiÞcation of Tachytes sp. 0 41 Borror et al. (1989)] for analysis. Numbers per day for Tachysphex sp. 0 2 each superfamily in each Þeld were compared across Philanthinae Cerceris bicornuta Gue´rin 2 73 treatments using analysis of variance (ANOVA). To Sphecinae satisfy ANOVA assumptions, counts were log (x 1 1) Ammophila procera Dahlbom 1 0 transformed before analysis. Seasonal means and Ammophila sp. 1 42 treatment combinations were separated using a least Chalybion californicum (Saussure) 0 1 Prionyx parkeri Bohart & Menke 1 2 signiÞcant difference (LSD) mean separation test or Sphex (S.) ichneumoneus (L.) 1 0 orthogonal comparisons (PROC GLM, CONTRAST Tiphioidea statement, SAS Institute 1996). Untransformed means Tiphiidae (6SE) are given in text and Þgures, whereas statistical Myzininae Myzinum sp. 1 36 results refer to transformed data. Vespoidea Vespidae Eumeninae Results Euodynerus megaera (Lepeletier) 0 3 Polistinae 1997. Various aculeate Hymenoptera including spe- Polistes fuscatus (F.) 0 10 cies from Apoidea, Pompiloidea, Scolioidea, Sphe- Polistes sp. 2 14 coidea, and Vespoidea were collected (Table 1). More than 50 individuals of the bumblebee Bombus penn- sylvanicus (De Geer) and the sphecid Larra bicolor F. spp. were analyzed separately, more were found in were collected. pheromone-baited traps than the combination or phe- As a group, numbers of Apoidea were not different nylacetaldehyde-baited traps (Tre´ce´ 0.15 6 0.06 per among treatments (F 5 4.9; df 5 2, 6; P 5 0.0557), day, Tre´ce´ 1 phenylacetaldehyde 0.03 6 0.02, phe- although there was a trend for more Apoidea to be nylacetaldehyde 0.03 6 0.02) (F 5 7.4; df 5 2, 6; P 5 collected in pheromone-baited traps than in pheny- 0.0241). The reverse was true for Sphecoidea (mostly lacetaldehyde-baited traps (Fig. 1). When Bombus L. bicolor), where more wasps were collected in traps June 1999 MEAGHER AND MITCHELL:NONTARGET HYMENOPTERA IN TRAPS 369 and Myzinum sp. (Tiphiidae), were collected only from cone traps. There was a trend for more Apoidea to be found in bucket compared with cone traps (F 5 4.0; df 5 1, 9; P 5 0.0758). No other superfamily showed differences between trap treatments. Discussion Several studies have documented the presence of aculeate Hymenoptera in monitoring traps (Adams et al. 1989, Mitchell et al. 1989, Gauthier et al.