Estimates of global species richness for insects are based upon extrapolations from “known” to unknown faunas and hence rely upon accurate counts of species for the referrent taxon or region. The number of reference species depends upon the species concepts employed by workers in that group combined with the degree to which nonstandard (i.e. nonmorphological) approaches have been used. Genetic data are more directly applicable to the detection of the apparent absence of gene flow, which lies at the heart of any species concept, than is morphological information. But what criteria can be used as a practical guide to suggest the absence of gene flow and define species-level units? Minimally, the phylogenetic species concept requires that there be one fixed difference between two samples for them both to be considered discrete species. The assumptions accompanying this definition include the survey of sufficient geographic locations, loci, and individuals. Based upon six studies of mostly widespread, readily identifiable and well-investigated bee “species”, we estimate that the number of species currently recognised may underestimate the true figure by half (although for at least two of the studies localities have been undersampled and more collections are needed). Even when examples for which there are fewer than five fixed differences between samples are removed from the data set, the number of recognised species increases by perhaps as much as 50% (the same caveat regarding undersampling of populations still applies). We suggest that the presence of morphologically unrecognised species may be more common among widespread, easily identified “species” than is generally accepted. Whether or not similar levels of species underestimation apply to other faunas, such as tropical rainforest canopy beetles, remains to be investigated.

Species richness of arthropods in northern temperate coastal rain forests far exceeds previous estimates, and the functional significance that these species play in ecosytem processes remains largely unknown. Examination of several species, many of which are not yet described, indicates that these intact ancient rain forests are structurally complex and act as reservoirs for biological diversity. Forest harvesting and resulting fragmentation affects arthropod diversity by altering key patterns of natural processes which are inseparably linked to habitat diversity. Consequences for arthropods may vary but those species which are endemic or inseparably linked to habitat features found only in these forests are particularly vulnerable to fragmentation-induced changes. Several important questions arise. What are the implications of forest fragmentation on ecological processes? What role does dispersal play in arthropod population viability? Given the immense biodiversity of arthropods, what are the functional roles that the species play in these forests and how are these changed when forests are harvested? The lack of empirical evidence makes it difficult to answer these questions and to quantify the functional roles of arthropods in these ecosystems.To address these questions, I suggest that studies should not rely on single-species approaches and the measurement of diversity (i.e. species richness and abundance) but should focus on addressing the functional roles of forest arthropods. To move beyond the basic description of pattern I suggest that studies concentrate on describing species assemblages while including dynamic processes such as dispersal into the framework of how we think about arthropods in ancient forests. The use of feeding guilds in the development of predictive models may give us an understanding of these factors and provide information that could be used to examine functional patterns in community structure.

Lathrolestes nigricollis (Thompson) (Hymenoptera: Ichneumonidae) was released in New England against Fenusa pusilla (Lepeletier) in 1979–1980 and again in 1989–1990. By 1991, the parasitoid was widely established in southern New England. From 1979 to 1990 the commonness of leafminers at the original release site in the first generation declined from 50–54% of leaves mined to under 3%. Over 4 years at five sites, parasitism of larval birch leafminers increased between leafminer generations from 10.3% (G1) to 27.6% (G2) to 39.2% (G3). In Massachusetts, parasitism decreased (from 28 to 14%) with increasing distance from the original release site, while corresponding damage increased (from 6% of leaves mined to 33%). In areas local to the original release site (<4 km) parasitism has reduced damage from birch leafminer to inconsequential levels. Some sites in other parts of southern New England, particularly roadside sites, still have high levels of mined leaves in spite of the parasitoid’s presence in the general area.

The formulated product of the insect-molting hormone analog, tebufenozide (MIMIC 240LV), and two aqueous formulations of Bacillus thuringiensis subsp. kurstaki, ABG6387 and ABG6414, were aerially applied over prethinned commercial stands of balsam fir infested with the eastern hemlock looper. MIMIC® was applied once at the rate of 65.1 g active ingredient (ai) in 1.86 L/ha and twice at the rate of 33.4–35.4 g ai in 1.91–2.02 L/ha. The formulations of B. thuringiensis were applied twice at rates of 19.3–24.1 billion International Units (BIU) in 1.54–1.93 L/ha for ABG6387 and 33.2–36.0 BIU in 1.67–1.80 L/ha for ABG6414.

Nine of the 10 plots treated with the single application of MIMIC showed reductions resulting from treatment, ranging from 3 to 93% within 9–11 days and 8 to 100% after 3 weeks. Plots treated twice with the lower dosage of MIMIC generally had larval population reductions of over 50% within 9–11 days of treatment and over 60% 3 weeks after treatment. For both treatments with MIMIC, plots with poor efficacy were clearly associated with low foliar deposition of the active ingredient. Overall, the results suggest that deposits of tebufenozide of less than 1.5 μg/g foliage do not give acceptable control of hemlock looper. Pupal population reductions resulting from treatment ranged from 8 to 99% for plots treated with the single application and 76 to 100% for plots treated twice at the half-dosage. Ground assessments of defoliation occurring after treatment ranged from 10 to 51% (year-old foliage) and 0 to 16% (current-year foliage) for the plots treated once, and 1 to 33% (year-old foliage) and 0 to 8% (current-year foliage) for the plots treated twice at the half-dosage. Thirty-five to 65% of year-old foliage and 15–39% of current-year foliage was defoliated in untreated check plots.Reductions resulting from treatment by ABG6387 ranged from 27 to 91% 10 days after the first application, from 73 to 99% 8 days after the second application, and from 90 to 100% for pupae. Plots treated with ABG6414 had larval population reductions from 76 to 93% 10 days after the first application and from 98 to 100% 7 days after the second application. No pupae were recovered in the plots treated with ABG6414. Defoliation assessed from ground level was less than 5% in the plots treated with ABG6387 and ABG6414 compared with defoliation of 10–29% of year-old foliage and 8–12% defoliation of current-year foliage in the check plots.

The dynamics of Aphis pomi De Geer (Homoptera: Aphididae) and its predator, Aphidoletes aphidimyza (Rondani) (Diptera: Cecidomyiidae), were studied over two seasons in a commercial apple orchard in Nova Scotia. Aphid and predator population dynamics were very different in the two seasons. The most numerous predator in both years was A. aphidimyza. This predator was found in the earliest colonies sampled each season and was present throughout the growing season. In the year with relatively high aphid densities a single peak in aphid and predator density was seen in mid-July, just prior to maximum predation rate. The predator’s numerical response resulted in delayed density dependence in predation rate. In the year of low aphid densities, predator and prey dynamics appeared uncoupled, and a higher rate of colony turnover may have been caused by other predators such as mirids. Aphidoletes aphidimyza tended to aggregate oviposition in high-density aphid colonies, but this aggregation resulted in predation that was only very weakly spatially density dependent.

Several horticultural oils, an insecticidal soap, neem, garlic extract, a sugar ester, and a synthetic insect growth regulator were evaluated under laboratory conditions for their ability to inhibit pear psylla’s [Cacopsylla pyricola (Foerster)] feeding and oviposition. Supreme oil significantly reduced winterform oviposition rates in both choice and no-choice assays and reduced feeding rates in no-choice assays. Supreme oil also reduced oviposition rates of summerform females in choice tests but not in no-choice tests. In addition, summerform oviposition rates were reduced significantly by application of several mineral oils (Orchex 796E, 692, and WS2928) and garlic barrier to pear foliage in no-choice tests, and by all oils, garlic barrier, and M-Pede in choice tests. Summerform feeding rates were reduced by application of mineral oils. Data suggest that "summer type" oils and other compounds may have utility in pear psylla management programs by decreasing feeding and oviposition.

Neem seed extract (20, 60, and 180 ppm azadirachtin) applied to the crowns of black spruce trees immediately prior to female reproductive bud flush, resulted in high mortality of early instar spruce budworm larvae. Dose-dependent reductions of 40–93%, as compared with check trees, were observed 2 and 4.5 weeks after treatment. Seed cone protection from grazing damage, however, was moderate; at 60 ppm azadirachtin, the incidence of cone grazing was reduced 29%, and the proportion of severely grazed cones reduced 39%, as compared with check trees. Larval mortality and bud damage observed in the laboratory corroborated field results. High toxicity to second-instar larvae, and performance during adverse weather conditions, supports the potential of neem seed extract for control of spruce budworm defoliation. However, both laboratory and field results indicate that feeding occurs before the effects of neem seed extract manifest, resulting in moderate protection of cones. Whereas foliage can tolerate low levels of feeding, female strobili are damaged significantly. Excellent protection of foliage was achieved from systemic injections of neem seed extract to white and black spruce trees (1.5 g azadirachtin/tree). Neem seed extract, formulated for systemic application, could provide a botanical alternative for control of forest pests in urban environments.

Aphodius hyxos Petrovitz, 1962 was reared in the laboratory and the third-stage larva is described and illustrated. The third-stage larvae and life histories of Aphodius granarius (L.), 1767 and A. hyxos are compared. Adults of A. hyxos are active from late autumn until early spring, whereas adults of A. granarius are active chiefly from early spring to late summer. The larvae of both species may be separated by differences in the distribution of setae on the raster, the shape of the anal lobes, and the size and number of holes in the respiratory plate.

Trials using four different sex combinations of Colorado potato beetle, Leptinotarsa decemlineata Say, demonstrated that male beetles move upwind towards females, whereas no attraction was seen between any of the other combinations. Attraction was not found to increase significantly after the female reached 10 days old. Visual stimuli appear to be unimportant from a distance of 50 cm. Orientation behaviour by the male beetle was observed as it walked upwind towards the female; no identifiable calling behaviour by the female was observed. The data indicate a sex pheromone is being produced by the female and is used by the male for mate location from a distance.

A new genus and species of grylline cricket are described from Baja California, Mexico. This cricket, unlike most others, is reddish in colour.

Strains of workers that were high or low in their retinue attraction response to queen mandibular pheromone (QMP) in a laboratory bioassay were fostered in queenless colonies to measure their differential queen-rearing behaviors. High-strain workers spent proportionately more time working on and in queen cells than low-strain workers, and there were significant age by strain effects for time spent rearing queen cells. No interindividual differences were detected among the strains in the tendency to rear queens. Results from this experiment suggest that QMP retinue attraction response may be a mechanism upon which selection acts for division of queen-rearing labor.

The wide distribution of obligate thelytoky in oribatid mites (Acari: Oribatida) raises the question of the mechanisms that allowed genetic diversity to be maintained or even promoted in these taxa. We tested whether or not endosymbiotic bacteria Wolbachia are associated with thelytoky in eight species of oribatid mites (as it is for most thelytokous parasitic Hymenoptera studied so far) by using a polymerase chain reaction assay. Our negative results in the selective amplification of Wolbachia DNA suggest that these bacteria are not responsible for the thelytokous mode of reproduction of oribatid mites. On the basis of known or suspected cytogenetic mechanisms in oribatid mites (automictic thelytoky, inverted meiosis), we discuss an alternative hypothesis for the inducement of thelytoky, and its relevance to the observed diversification of thelytokous oribatid mites over evolutionary time.

Eight species are recognized within the New World fauna of the tribe Pogonini: Thalassotrechus barbarae (Horn, 1892), Ochtozetus bicolor (Brullé, 1843), Ochtozetus inexspectatus sp.nov. (type locality: Santa Isabel, Mato Grosso, Brazil), Diplochaetus rutilus (Chevrolat, 1863), Diplochaetus megacephalus sp.nov. (type locality: Bottomless Lakes State Park, New Mexico), Diplochaetus emaciatus Bates, 1891, Diplochaetus planatus (Horn, 1876) [new combination] and Pogonus texanus Chaudoir, 1868. Two names are listed for the first time in synonymy: Pogonus parallelus LeConte, 1874 (= Pogonus lecontei Horn, 1876) with Diplochaetus rutilus (Chevrolat, 1863) and Diplochaetus desertus Van Dyke, 1953 with Diplochaetus emaciatus (Bates, 1891). Lectotypes are selected for Trechus barbarae Horn, 1892, Pogonus emaciatus Bates, 1891 and Pogonus depressus LeConte, 1874. A key to the species, illustrations of the most important characters and distributional maps of most species are included.

Although Amphiagrion abbreviatum (Selys) and Argia vivida Hagen often occur at the same geothermally heated springs in western Canada and the United States, they differ markedly in their abundance at any particular site. There is no relationship between crude data on water temperature, conductivity, or aquatic vegetation and the relative abundance of the two species, but there is a striking correlation with presence or absence of trees. The absence of A. abbreviatum from heavily treed areas is associated with the paucity of suitable daytime perching sites, and there may be competitive pressure exerted by A. vivida for the perching sites that are available. Argia vivida does not live at open sites because it requires trees for night-time roosts. Argia vivida roosted higher than A. abbreviatum in cages and held the body at a greater angle from the cage wall. The roosting posture of A. vivida is probably related to interception of solar radiation in the morning, and the body positions of both species possibly provide defence against predation.

The distribution of the yew big bud mite, Cecidophyopsis psilaspis (Nalepa), a pest causing severe damage to Pacific yew, Taxus brevifolia Nutt., was determined within British Columbia. From 1992 to 1994, vegetative and reproductive buds on branch samples collected from 38 interior and 43 coastal populations of Pacific yew were examined for the occurrence of the mite. Mite damage was characterized by marked swelling and browning of both vegetative and reproductive buds. Interior Pacific yew populations were found to be mite free, whereas coastal yew populations were mite infested, except at three high-elevation (>700 m) coastal locations which were apparently mite free. At three other sites, individual trees without symptoms of mite attack did harbour low-density, nondamaging mite populations. The hypothesis that this mite is indigenous is less supported than the hypothesis that it was introduced to British Columbia on English yew, Taxus baccata L., and is limited to its present distribution by climatic and mountain barriers.

A second species of the dry wood termite genus Marginitermes Krishna is described from Baja California del Sur, Mexico, and a redescription of the genus is provided. Soldier head capsules, mandibles, antennae, labra, postmenta, and pronota are illustrated for Marginitermes cactiphagus Myles sp.nov. and Marginitermes hubbardi (Banks). Both species occur in dead skeletons of columnar cacti and other wood species. Both species differ from related drywood termites in the structure of the proventricular armature, suggestive of unique adaptation for intestinal manipulation of cactus wood fibres. Marginitermes is shown to be related to a western continental subgroup of Incisitermes comprised of I. minor (Hagen), I. marginipennis (Latreille), and I. fruticavus Rust.

Ptomaphagus inyoensis is described from Poleta Cave, Inyo County, California. Its reduced, but present, eyes and reduced flight wings suggest that it is a deep soil species. The spermatheca shape and other characters suggest that it is in a group with Ptomaphagus fisus Horn, a species widespread in animal nests and burrows in the southwest, Ptomaphagus cocytus Peck, a cave-restricted species from the Grand Canyon of Arizona, and Ptomaphagus manzano Peck, a montane litter species from New Mexico. A comparative morphological analysis of the group is presented.

The development of adult and nymphal populations of Lygus lineolaris (Palisot de Beauvois), L. elisus Van Duzee, and L. borealis (Kelton) was studied in relation to seeding date and stage of plant development in four seedings of canola (cv. Westar) in field plots at Glenlea, Manitoba, in 1988 and 1989. Lygus lineolaris was the dominant species in all seedings and in both years. Colonizing adults (first-generation adults) of the three Lygus spp. first invaded the plots at about the same time from the late rosette to the early flowering stages of plant development, and Lygus adult numbers reached maxima during flowering in all seedings. The stage of plant development, and not seeding date, was important in attracting colonizing adults of Lygus to canola. Second-generation nymphs first appeared at the flowering stage of canola. All nymphal instars reached maxima and median abundances during the late flowering and pod development stages in all seedings. Second-generation adults were present from early to late pod development stages. Median abundances of second-generation adults of L. lineolaris, L. elisus, and L. borealis usually were reached at the same plant growth stage in each seeding.