The North American species of the genus Cosmia Ochsenheimer are reviewed and four species are recognized: Cosmia calami (Harvey), C. praeacuta (Smith) comb. nov., C. epipaschia (Grote) comb. nov., and C. elisaesp. nov.Achytonix McDunnough syn. nov. is synonymized with Cosmia subgenus Calymnia Hübner. Namangana praeacuta nigramaculata Barnes et Benjamin and Namangana praeacuta orae Barnes et Benjamin are synonymized with Cosmia praeacuta. Illustrations of adults and genitalia are provided.

Studies based on presence/absence of a species may provide insight into habitat associations, allowing the distribution of species to be predicted across the landscape. Our objective was to characterize the carabid fauna in three mature boreal forest types (conifer, mixedwood, and deciduous) and a disturbed habitat (clearcut) to provide baseline data on the carabid species inhabiting the major forest types of boreal northwestern Ontario. Only spring-active carabids were considered for logistical reasons. We further identified the coarse woody debris structure and microhabitat characteristics within these stand types to try to refine our ability to predict the within-stand occurrence of carabids. We found the carabid composition of deciduous stands (mixedwood and aspen) similar and typical of the carabid fauna found in mixedwood forest in other nearby studies. The important feature of these forest types may be the presence of a deciduous leaf litter layer. In contrast, spruce stands contained fewer species and individuals and a distinct carabid fauna. As predicted, the clearcut habitats were dominated by open habitat species, although the fauna was dissimilar to what was recorded in clearcut habitats in other studies within northwestern Ontario. Coarse woody debris appeared to be an important habitat attribute for several carabid species in clearcut habitats.

Furia gastropachaecomb. nov. (= Furia crustosa MacLeod et Tyrrell) has long been associated with declines in populations of the forest tent caterpillar, Malacosoma disstria Hübner; however, its prevalence and impact on host populations have never been documented. We clarify the nomenclature for this species, which has previously been called F. crustosa, and describe the morphology of F. gastropachae isolates from Maryland and New York. Epizootics of F. gastropachae were studied in M. disstria populations in New York and Maryland, United States of America. Prevalence (mean ± SE) of F. gastropachae in late fifth instar larvae in New York was 25.6 ± 7.8% compared with 22.2 ± 11.3% infection by virus and 23.3 ± 8.4% parasitism. In Maryland, F. gastropachae was not found in early instars and its incidence in fourth and fifth instars was 14.5 ± 7.3 and 21.6 ± 17.6%, respectively. Dipteran parasitoids were also important natural enemies with 38.3 ± 3.0 and 17.9 ± 1.4% parasitism in fourth and fifth instars, respectively. Virus was found in fewer than 5% of larvae collected. Furia gastropachae showed a marked tendency toward resting spore production in infected larvae, with 100% of larvae collected in New York and >80% of larvae collected in Maryland producing resting spores, either alone or in combination with conidia. The appearance of cadavers of larvae dying from fungal and viral infections was similar. Of 13 species of Lepidoptera exposed to F. gastropachae conidia in host-range bioassays, only 3.3% of Danaus plexippus (L.) (Lepidoptera: Danaidae), 5.6% of Pieris rapae (L.) (Lepidoptera: Pieridae), and 3.7% of Manduca sexta (L.) (Lepidoptera: Sphingidae) were successfully infected by this fungus.

Dusky wireworms, Agriotes obscurus (L.), aggregated in similar numbers at wheat [Triticum aestivum L. (Gramineae) ‘Max’], oat [Avena sativa L. (Gramineae) ‘Walderen’], barley [Hordeum vulgare L. (Gramineae) ‘Verdin’], and fall rye [Secale cereale L. (Gramineae) ‘Wheeler’ and ‘Prima’] cultivar bait stations containing 100 seeds planted 3 cm deep in 127-cm2 circular bait stations. Similar levels of aggregation also occurred at 11 varieties of wheat planted at 100 seeds/127 cm2. When wheat, oat, barley, and the fall rye cultivars were planted at increasing density (0–180 seeds per bait station), aggregation by A. obscurus increased initially, but reached a plateau at numbers and at seeding rates specific to each grain variety as determined using the asymptotic equation y = B0(1 – e–B1x). Except for barley, this equation predicted wireworm densities within 11% of the densities actually observed at bait stations with 100 seeds/127 cm2. It was concluded that any of the wheat, oat, barley, or fall rye varieties would be suitable for monitoring A. obscurus wireworm populations if planted in bait stations at 100 seeds/127 cm2, as well as for aggregating wireworms by means of a trap crop.

The seven-spotted lady beetle, Coccinella septempunctata L., was first reported in Manitoba in 1988. The effect of this introduced species on the relative abundance of the more common native coccinellines was determined from D-Vac Insect Net® and sweepnet samples in alfalfa in 1983–2001, by sweepnet and visual sampling in field crops and other vegetation in 1989–2001, and by transect sampling of aggregations of coccinellines in spring and autumn on the shore of Lake Manitoba from 1989 to 2001. The trends of annual changes in the abundance of six species in southern Manitoba were similar, whether based on the mean density per 200 sweeps in alfalfa or on their relative abundance in alfalfa, on other vegetation, and in aggregations on the beach. Before 1988, Hippodamia tredecempunctata tibialis (Say), Coccinella transversoguttata richardsonii Brown, and Hippodamia convergens Guerin were the most abundant species, followed by Hippodamia parenthesis (Say) and Coccinella trifasciata perplexa Mulsant. By 1992, C. septempunctata had become the dominant species, but H. tredecempunctata has since been the dominant species in most collections. The relative abundance of C. transverso guttata, H. convergens, H. parenthesis, and C. trifasciata has decreased since the establishment of C. septempunctata. The decline in abundance of these species seems to be caused by their competitive displacement by C. septempunctata. Their displacement also might be a contributing factor in the slight increase in abundance of H. tredecempunctata because C. septempunctata may compete less directly with this species than with other native coccinellines.

Meligethes viridescens (Fabricius), bronzed or rape blossom beetle, is a widespread and common pest of oilseed rape [Brassica napus L. and Brassica rapa L. (Brassicaceae)] in the western Palaearctic subregion. The establishment of M. viridescens in eastern North America has raised concern that its presence is a potential risk to the Canadian canola industry, especially to the prairie ecozone of western Canada where up to 4 million ha of summer canola (B. napus and B. rapa) are grown annually. Study of museum specimens indicated that M. viridescens was first recorded in Nova Scotia in 1947. Field surveys indicated that, as of 2001, M. viridescens was established as far west as Saint-Hyacinthe, Quebec. A CLIMEXTM model for M. viridescens in Europe was developed and validated with actual distribution records. In Canada the model predicted that once introduced, M. viridescens would readily survive in the canola-growing areas. The actual distribution of M. viridescens in eastern Canada matched the predicted distribution well. The westward dispersal to and establishment of M. viridescens in canola-growing areas of Ontario and western Canada, particularly southern Manitoba, appear to be inevitable. Establishment in these areas presents the risk of substantial production losses to canola producers.

The relative importance of exploitative competition for resources on grasshopper reproductive allocation has not been fully examined. Given the large fluctuations in grasshopper densities that periodically occur in western North America, an increased understanding of how grasshopper survival and reproduction vary in response to intraspecific densities and per capita resource availability is important. I examined if exploitative resource competition could explain variation in reproductive allocation in Melanoplus sanguinipes (Fabricius) in response to resource availability and grasshopper population density. I also examined whether individual differences in competitive ability resulted in increased variance in egg production with low per capita resource availability. As expected with exploitative resource competition, per capita resource availability explained a significant amount of the variation in all reproductive characteristics examined. There was no effect of per capita resource availability on survival. Residuals of the regressions of egg production and vitellogenesis versus per capita resource availability did not differ for resource or density treatments, indicating that exploitative competition for resources played a more important role than interference competition in determining reproductive allocation in M. sanguinipes. Individual differences were evident, as variation around the mean of egg production increased with resource limitation. Exploitative competition for resources was important in determining both individual and population-level reproductive responses of grasshoppers to resource availability.

Imidacloprid-treated spheres and several classes of insecticides were evaluated in field and laboratory experiments to determine their effects on Rhagoletis mendax Curran adult behavioural activity and larval infestation in highbush blueberries, Vaccinium corymbosum L. (Ericaceae). In field tests, three treatments representing an attract-and-kill system (imidacloprid-treated spheres) and two classes of insecticides, including a naturalyte spinosad (SpinTor 2 SC) and organophosphates (Imidan®/Malathion), were evaluated against an untreated control. Significantly more blueberry maggot puparia were collected from untreated (control) plots than from other treatments evaluated. There were no significant differences in larval infestation of berries collected from plots treated with imidacloprid-treated spheres, SpinTor 2 SC, and Imidan®/Malathion. Less than 1% of the berries collected from imidacloprid-treated sphere plots and SpinTor 2 SC treated plots contained blueberry maggot larvae. In toxicity and fruit-injury studies, SpinTor 2 SC as well as additional compounds from various classes were evaluated, including a botanical [azadiractin (Ecozin 3%)], a particle film [kaolin clay (SurroundTM WP)], and three neonicotinoids [imidacloprid (Provado 1.6 F), thiocloprid (Calypso 480 SC), and thiamethoxam (PlatinumTM 2 SC and ActaraTM 25 WG)]. SpinTor 2 SC exhibited a time lag (18 h) in reducing R. mendax activity. Ecozin 3% and SurroundTM WP were ineffective in suppressing R. mendax adult activity, but numbers of larvae in SurroundTM WP treated fruit were significantly reduced. The effectiveness of neonicotinoid insecticides varied initially but resulted in equivalent levels of mortality after 48 h compared with our conventional organophosphate treatment of Guthion 50 WP. Oviposition scars on Provado 1.6 F treated plots were significantly more numerous than on plots treated with SpinTor 2 SC, Calypso 480 SC, Actara 25 WG, and Guthion 50 WP.

To effectively monitor forest tent caterpillar, Malacosoma disstria Hübner (Lepidoptera: Lasiocampidae), populations using sex pheromone baited traps, we field-tested pheromone dispenser (lure) type, lure age, and trap design using (Z,E)-5,7-dodecadienal:(Z,Z)-5,7-dodecadienal (100:1). Rubber septa lures, polyurethane lures, and two trap types [sticky-type pheromone traps (Wing Trap I) and bucket-type pheromone traps (Universal Moth trap)] were evaluated. Traps baited with polyurethane lures produced higher trap catches and lower zero-catch frequencies than did rubber septa traps. There was no detectable difference in trap catch among polyurethane lures aged 0–28 days. Wing traps reached a functional saturation point in outbreak M. disstria populations and caught fewer moths than Universal traps in nonoutbreak populations. A nonsaturating trap such as the Universal trap in conjunction with the polyurethane lure should be effective for monitoring M. disstria populations.

Receptor characterization (i.e., identifying what will be affected by an activity) is the first step in a risk assessment of biocontrol agents for insects. Development of a representative list of species at risk, based on ecological vulnerability, enables host-range screening of potential biocontrol agents on a manageable group of nontarget insects. A database of 153 species was used to characterize the butterflies potentially at risk from an inundative release of the egg parasitoid Trichogramma minutum Riley. Risk criteria for butterfly species included North American, Canadian, and Ontario geographic distributions; oviposition phenology; number of generations per year; overwintering stage; host-plant preferences; and egg mass type and location. Ecological vulnerability lists of butterfly species were generated for northern and southern Ontario; areas where there have been recent experimental inundative releases of T. minutum for the suppression of forest pests. Based on the above criteria, 2 species and a maximum of 27 species would be potentially at risk, and thus requiring host-range testing if an inundative release were considered for northern and southern Ontario, respectively. The number of species on the ecological vulnerability list for southern Ontario could be reduced to 12 species depending on the specific geographic location in southern Ontario of the inundative release. The six criteria used for receptor characterization for T. minutum, associated primarily with host-habitat location and host-location, can also be used for other parasitoids. They are components of any target host's biology, and thus will affect the scale and impact of any parasitoid attacking eggs, larvae, or pupae. Additional criteria for receptor characterization may also be added that will relate to the specifics of a parasitoid's biology and are associated with host acceptance and host suitability. Development of ecologically based vulnerability lists should become standard practice in determining which nontarget species require host-range testing, for both inundative and classical biocontrol agents targeting insects, and for the potential impact of invasive species.