The diamondback moth, Plutella xylostellaLinnaeus (Yponomeutidae), remains one of the most serious pests of crucifers in many parts of the world, particularly in South East Asia. This paper reviews the wide array of strategies which have been used in attempts to control P. xylostella over the last 40 years and consolidates the fragmented and sometimes hard-to-access literature on two and three level trophic interactions (e.g. host plant—herbivore, herbivore—natural enemy and host plant-herbivore—natural enemy interactions) involving the pest. Special reference is made to studies and examples from South East Asia, and in particular Malaysia, where P. xylostella and its primary parasitoids have been well studied owing to the great economic significance of the pest in this region. It is suggested that a multitrophic approach to research may assist in the development of more sustainable methods for the management of P. xylostella and overcome some of the problems inherent with insecticide-intensive methods.

Mortality responses of adult New Zealand flower thrips, Thrips obscuratus (Crawford) to controlled atmospheres were assessed. Temperature (0–48°C) and time (1–24 h) had major effects on mortality. There were approximately two-fold increases in thrips mortality between atmospheres with 40 and 60% CO2, when there was 0 or 2% O2 present in treatments of 0°C for 4 h and 24 h; 12°C for all exposure times; 24°C for 1 h. Thrips mortality was significantly greater when there was no O2 in the test atmosphere, than when there was 0.25, 8 or 12 or 20% O2. The results show that there are a variety of ways produce infested with New Zealand flower thrips could be treated with controlled atmospheres to achieve high mortality. These data can be used as the basis of potential quarantine treatments depending on the produce being exported.

Age composition of the filarial vector mosquito Culex quinquefasciatus (Say) was determined by examining the number of ovariolar dilatations of 1200 adult females from Calcutta. Average duration of the gonotrophic cycle was 4.5 days. The proportion parous, the daily survival rate and daily mortality rates of the natural population were 0.53, 0.87 and 13%, respectively. The oldest mosquito sampled in the study area passed eight gonotrophic cycles in its lifetime.

The genus Stenobracon Szépligeti which contains species of economic importance for biological control of lepidopterous borers of sugarcane, rice and other graminaceous crops in Asia, is revised. Principal components analysis was employed to help delineate species. Six species, distributed from Pakistan to Australia, are recognized. Stenobracon brevis, S. clarus and S. malensis are described as new and Stenobracon nicevillei (Bingham), S. deesae (Cameron) and S. oculatus Szépligeti are redescribed. Illustrated keys for males and females are provided. Phanaulax levituberculatus Cameron is a new junior subjective synonym of S. nicevillei and S. trifasciatus Szépligeti is a new junior subjective synonym of S. oculatus. Elphea flavomaculata Cameron is synonymized with Ischnobracon (Bracon) laboriosus Cameron (comb. n). Host records are summarized.

The foraging behaviour of the redlegged earth mite, Halotydeus destructor (Tucker), in annual pasture was documented and the relative numbers of individuals foraging on subterranean clover, Trifolium subterraneum (Leguminosae), capeweed, Arctotheca calendula (Compositae) and grasses (mainly Lolium rigidum (Gramineae)), estimated in winter and spring in south-western Australia. The main feature of H. destructor foraging behaviour was the presence of aggregations ranging in size from 3–36 individuals on the adaxial surfaces of the upper canopy of the pasture. Most aggregating individuals were feeding, while most solitary and paired individuals were searching. There was no evidence that individuals aggregated for reproductive behaviour; the adult sex ratio was similar to that of the population even though immatures were under-represented in aggregations. During winter the distribution of numbers of H. destructor on subclover, capeweed and grasses in the upper canopy of pasture was similar to the estimated ground cover of these plant species. However, individuals showed host-plant selection for subclover when feeding. During spring, when the pasture plants were flowering, there were more H. destructor in capeweed flowers than in aggregations on subclover foliage. The capeweed flowers were shown to enhance the rate of H. destructor reproduction in a laboratory experiment. Most H. destructor were on or near the soil surface, with 10% at any one time on the upper canopy of the pasture. H. destructor is polyphagous, and behaved in pasture as a selective opportunist feeder.

Laboratory nests of the leaf-cutting ant Atta sexdens rubropilosa Forel fed daily with leaves of Ricinus communis showed a gradual decrease in fungal garden volume, a higher ant mortality rate, and fungal garden extinction after 6 weeks. The mean oxygen consumption rate of these ants was higher than that of control ants collected from nests fed with leaves of Eucalyptus alba (Myrtaceae) suggesting one or more components of the leaves of R. communis had a direct physiological effect on the ants, in addition to inhibiting fungal garden growth.

The Kerala isolate of Oryctes baculovirus (OBV-KI) was introduced into the Andaman Islands in May 1987 for the population suppression of Oryctes rhinoceros (Linnaeus). The virus was released initially at four locations along the island chain. In all the places of virus release, damage to coconut palms was reduced by about 90% within 43 months of release. At one place of virus release, Sipighat, the adult beetle population in the field was reduced by 80% within 18 months and by 96% by the end of 55th month of release. Sixty per cent of the adult beetle population in the field showed symptoms of infection 30 months after virus release. The decline in the numbers of breeding sites at Viper Island and Sipighat were 100% and 98% respectively at the end of 43 months of virus release. The rate of spread of the virus in the field was estimated to be about 1 km/month. The beetle population remained at low levels after the establishment of the virus in these islands.

Resistance involving penetrability of cotyledons of subterranean clover Trifolium subterraneum (Leguminosae) by the redlegged earth mite, Halotydeus destructor Tucker, was examined in this study using several approaches. Values of penetrability, presented as mechanical strength measured with a penetrometer with a tubulated probe, of cotyledons tested 11–13 days after seeding, from nine varieties of two subspecies of subterranean clover (T. subterraneum subsp. yanninicum and subsp. brachycalicinum) were negatively correlated with feeding damage caused by H. destructor in both 3 h pairwise choice tests in a Petri dish with detached cotyledons, and in 2 week multiple choice tests in containers with seedlings. Simulation tests with different artificial membranes (Parafilm and Gladwrap) making up sachets, containing 5% glucose as a feeding stimulant, indicated that within 3 h H. destructor preferred membranes with lower strength. Field-collected young adults had significantly higher physical fitness to feed and/or gather on the artificial membrane sachet containing 5% glucose, than those from the laboratory culture, when the sachet was compared with cotyledons of either resistant (DGI007) or susceptible (Dalkeith) varieties in the choice tests. Electron microscopic observations of mite infested cotyledons of the resistant variety (DGI007) reveal that upper epidermal cell walls around penetration holes are bent inwards. Anticlinal cell walls of palisade cells appear buckled following attack by H. destructor. The results from these studies were taken as further evidence for the involvement of mechanical factors in antixenotic resistance of subterranean clover cotyledons to H. destructor.

All available identifications of sibling species of the Simulium damnosum complex in southern Sierra Leone for the years 1983–1987 have been analysed for any evidence that they might provide on whether the appearance of S. sirbanum in this area in 1988 is indicative of a gradual southerly extension of range, or a unique occurrence. No evidence was found to suggest a long-term immigration trend over the five years prior to 1988, and the latter of the two possibilities is the more likely. No meteorological or other factors are known that might account for the sudden appearance of S. sirbanum in southern Sierra Leone in 1988, but the possible causes of such immigration into forested parts of West Africa and their epidemiological implications for human onchocerciasis in the region are briefly discussed.

Culex pipiens Linnaeus is represented by two forms: the form molestus which is autogenous, stenagamous and has no reproductive diapause, and the form pipiens which is anautogenous, eurygamous and diapauses. The ecological, physiological and geographical variations in the larval siphonal index of C. pipiens (the ratio of the length of the siphon to its width at the base) was investigated using both original measurements and previously published data. In all, data were used from 259 collections from open water bodies (marshes, ditches, holes, barrels, etc.) and underground waters (house basements) in areas of Europe, Asia and part of Africa. Autogeny was studied in some cases by dissection of females. In C. pipiens the mean siphonal index was strongly correlated with ecology (water body type) and adult physiology (autogeny). In spite of high geographical and individual variation, which makes the identification of individual larvae difficult, the mean siphonal index can be used with certainty for the differentiation of populations of the two forms of C. pipiens.

Monitoring of Glossina austeni Newstead populations in the forested areas of Unguja island of Zanzibar has since November 1990 routinely been carried out with the sticky panel trap because commonly used tsetse traps (Biconical, Epsilon, F3) have proved to be unsuccessful in catching G. austeni. Initial studies on the catching ability of various types of sticky panels for G. austeni indicated that the monopanel was as efficient in catching flies as the 3-dimensional version and the smaller legpanel. No significant differences in catch rate and sex ratio were observed with monopanels in various colours and colour combinations. Legpanels coloured white on one panel side and blue on the other side caught significantly more flies compared with other colour combinations, but female flies were under-sampled (32%). The type of sticky material applied on the panel influenced significantly the catch rate and female ratio. During long-term trapping with baby blue and white monopanels, females were under-sampled (38–46%) except when polyisobutyleneLMW was used as sticky material. Analysis of the age composition of the sampled G. austeni females revealed that tenerals and nulliparous were well represented (11–24%). More than 20% of the trapped females were older flies, i.e. females with four or more ovulations, but this percentage dropped to 10% when Tanglefoot was used as sticky material.

Predatory feeding on Tetranychus urticae Koch (Acari: Tetranychidae) populations on cotton by phytophagous thrips, Thrips imaginis Bagnall, T. tabaci Lindeman and Frankliniella schultzei Trybom (Thysanoptera: Thripidae), was investigated in the field and laboratory. Phytophagous thrips are a common early season pest of cotton in Australia, though their true pest status is undefined. In California, the phytophagous thrips (Frankliniella occidentalis (Pergande)) is regarded as an opportunistic predator of mite eggs, their consumption of which increases fitness over a diet of leaf tissue alone. Thrips are among the most abundant of insects on young cotton. If they consume mite eggs, even at relatively low rates, they could have a significant influence on the probability of survival of early season spider mites. Consumption of eggs of T. urticae by thrips was investigated in the laboratory. Second instar F. schultzei consumed more eggs per day (ca. 4 eggs per day) than did second instar T. tabaci or T. imaginis (ca. l egg per day). Consumption by first instar F. schultzei was much lower than for second instars. Adult T. tabaci consumed ca. l egg per day whilst adults of F. schultzei consumed only ca. 0.5 eggs per day, although some individuals of this species did consume substantial numbers of eggs. Larvae of all thrips species showed a type II functional response to prey density. In the field, adults and larvae of T. tabaci and F. schultzei showed a preference for cotton seedlings that were also infested by spider mites. In a glasshouse, larvae of T. tabaci showed a highly significant preference for feeding within mite colonies. In the field, suppression of predators, predominantly T. tabaci and F. schultzei, with a broad spectrum insecticide (dimethoate) contributed to outbreaks of mites occurring earlier than they would have otherwise. The results show that phytophagous thrips eat mite eggs and that they are potentially important predators of spider mites in the field, especially given their abundance on young cotton and preference for inhabiting situations in which mite colonies are found.

Factors (developmental state, starvation time, host local density and host distance from insects) affecting the response of the ragwort flea beetle, Longitarsus jacobaeae (Waterhouse), to upwind plants of tansy ragwort, Senecio jacobaeae (Compositae), were examined in a wind tunnel. There was seasonal variation in beetle response to host odours. Individuals collected during spring and early summer (6 May to 14 July 1993) showed directed movement toward the upwind plants. However, individuals collected in mid-summer (23 July 1993) showed no significant response to upwind host plants and walked randomly in the wind tunnel. Individuals collected during late summer to autumn (4 August to 23 October 1994) once again showed response to upwind plants, whereas those collected in winter (January 1994) walked randomly in the wind tunnel. The absence of beetle response to plant odours coincided with potential summer aestivation and winter ‘hibernation’. For spring and early summer beetles that responded to plant odours, the strength of the response did not change significantly with the number of plants (1, 2, 4 or 6) presented upwind in the tunnel nor with the starvation time (2, 6, 10, 12, 24 or 36 h) of the tested beetles, and these patterns were consistent for male and female beetles. Individuals responded to plants from a distance of 60 and 300 cm. The speed of movement, similar for males and females, increased slightly as L. jacobaeae oriented more directly toward host plants.