The survivors of field treatments of Blattella germanica (L.) with ben-diocarb (strain D) and malathion (strain H) in the UK were bred in the laboratory. The resistance of these two strains to commercial formulations of bendiocarb, dioxacarb, propoxur, lindane, dieldrin, malathion, diazinon, fenitrothion, pirimiphos-methyl and permethrin was assessed in 24-h exposures on treated glass plates. Resistance in terms of knockdown or mortality was demonstrated for bendiocarb, dioxacarb and dieldrin in both strains, for malathion in strain H and, marginally, for permethrin in strain D. Selection with bendiocarb in strain D produced a x4 and x3 increase in resistance to bendiocarb at the LC50 and KD50 levels, respectively. There was no significant cross-resistance to propoxur, but cross-resistance to dioxacarb was evident at the LC50 and KD50 but not the 95 % levels. There was little or no cross-resistance in strain D to diazinon, pirimiphos-methyl or fenitrothion. In strain H, malathion selection increased resistance to malathion at the 50% but not the 95% levels, and there was no evidence of cross-resistance to diazinon, pirimiphos-methyl or fenitrothion. Some cockroaches from strains D and H were able to recover from knockdown during and after exposure to the carbamates and permethrin. There was no recovery from the organophosphorus or organochlorine insecticides.

Studies on the precision levels and efficiencies of available insect suction samplers (D-Vac and FARMCOP) were undertaken in the Philippines to determine the best possible sampling procedure to derive absolute population estimates of delphacids, cicadellids and their predators in flooded rice. The D–Vac suction sampler when used in conjunction with an enclosure placed over the rice hill prior to sampling was found to be the most suitable. The sampling procedure adopted is described in detail, and sampling efficiencies for the D–Vac are given for nymphs and adults of the delphacids and cicadellids and their predators. These values could be used by other workers following the sampling procedures described here on modern rice varieties grown under flooded conditions to achieve absolute population estimates.

Eighteen strains of the complex of Culex pipiens L. from Africa, Asia and Europe were bioassayed for resistance to chlorpyrifos and electro-phoresed and stained for esterases. Susceptible strains showed only low activity esterase bands. The resistant strains of C. quinquefasciatus Say from hot countries (Liberia, Nigeria, Sri Lanka, Tanzania, Thailand) all showed the same two high intensity esterase bands (Rm 0·60 + 0·82). Different patterns of high esterase were found in resistant C. pipiens strains from cooler localities in Nairobi, Kenya (Rm 100), and Mont-pellier, France (Rm 0–50). Selection experiments on strains originally polymorphic for resistance and esterase pattern showed, without exception, that high esterase remained associated with resistance, and it is concluded that the association is almost certainly causal and not merely due to genetic linkage. The high intensity esterase bands were probably due to alleles of the loci Est-l, Est-2 and Est-3, separated by crossover distances of approximately 2·4 and 5·5 units, respectively. Strains monomorphic for what appeared to be the same high esterase pattern varied markedly in resistance level. Enzyme assays showed a direct relationship between levels of enzyme activity and resistance. Bioassays with fenthion and chlorpyrifos revealed differences in the relative resistance of C. quinquefasciatus from Colombo (Sri Lanka) and Dar-es-Salaam (Tanzania). Despite these differential degrees of cross-effectiveness, it is concluded that high intensity esterases are reliable indicators of organophosphate resistance in mosquitoes of the C. pipiens complex, although the possibility of other resistance mechanisms means that the lack of abnormally active esterases does not necessarily indicate the absence of resistance.

A preliminary investigation to find an easily assayable biochemical character, varying reproducibly with age which would give a more accurate assessment of insect age than the best methods then available, had revealed the fluorescent eye pigments, the pteridines, as a promising candidate. Stomoxys calcitrans (L.) was chosen as a model. Double-blind laboratory experiments in which the age post-eclosion of females was predicted from a standard laboratory curve of pteridine accumulation with age post-eclosion, showed the method to be accurate on average to ±1·49 days. Further laboratory experiments defined the relationships between temperature and pteridine accumulation in males and females of S. calcitrans such that the method could be modified for field use. Field observations in the UK indicated that the temperature of adults in the wild is determined by ambient temperature, the number of sunlight hours and the flies' own physiological and/or behavioural capabilities. From this information, equations for the accumulation of pteridines with temperature and age post-eclosion in males and females of S. calcitrans were constructed. To test the accuracy of these equations, approximately 19 600 marked flies of known age were released in a farmyard; 126 females were recaptured over 22 days and 90 males over 19 days. The average errors of predicted age using the pteridine accumulation method were ±190 days for females and ±137 days for males over the life-span of the recaptured insects.

Manitoba traps were used in a pasture field in northern England between June and October in 1980 and 1981.Samples of Diptera were also collected from cattle, cow dung, the field surface and from flowers in the hedges. A comparison of these samples, using multivariate similarity coefficients, showed that the Manitoba trap and cattle samples were most similar. There were 12 weeks in the two years during which both Manitoba trap and cattle samples were obtained, and the accumulated catches for this period were compared. They consisted mainly of cattle-visiting species, the proportion being greatest in the Manitoba trap samples (97%). Non-cattle-visiting species in the cattle sample (25%) consisted mainly of dung flies and species associated with the field surface. It is suggested that these were taken accidentally in proportion to their abundance around the cattle. There were some differences in the proportions of various species between the samples from cattle and the Manitoba trap, and the sex ratio of the three most abundant species was significantly higher in the traps. An experiment in which a large electrocuting grid was used to intercept flies travelling towards artificial bait as used in Manitoba traps, showed that the proportion of the two most abundant species differed significantly from a sample obtained from a nearby Manitoba trap. No such difference was observed in the sex ratios of these species.