Cyclodiene insecticide resistance persists in field populations of Drosophila spp. at a frequency of approximately 1% (0.01), despite the withdrawal of most cyclodiene type insecticides except endosulfan. However, we have previously documented that resistance-associated amino acid replacements in the gene Rdl, a γ-aminobutyric acid receptor, can significantly affect several channel functions of the integral chloride ionophore. We were therefore interested in investigating if different resistance-associated replacements confer significant fitness disadvantages and whether the use of endosulfan could be maintaining selection for cyclodiene resistance in the field. Using PCR amplification of specific alleles (PASA) within 3000 individual flies, we report that neither the alanine302 > serine (allele 1) replacement in Drosophila melanogaster Meigen nor the alanine302 > serine (allele 1) or alanine302 > glycine (allele 2) replacements in D. simulans Sturtevant showed any reduction in frequency in cage experiments run for one year in the laboratory in the absence of selection. Further, repeated applications of endosulfan selected significantly for cyclodiene resistance in the field. Thus the apparent absence of fitness cost, combined with the continued use of endosulfan, may maintain cyclodiene resistance at this relatively high frequency in field populations.

The association of esterase (EST) activity with resistance to the organophosphorus (OP) insecticide methidathion was investigated in field-collected populations of whitefly (Bemisia tabaci (Gennadius)) in Israel. The inheritance of EST activity was studied by controlled crosses in the laboratory. Among-family variance of EST activity was highly significant although all families were maintained in the same rearing room. This indicates that genetic or common environmental effects due to rearing each family on a different caged plant must have been important. Heritability estimated from son-mother regression was h2=0.98, but this estimate seems to be unrealistically high. Daughter-mother and daughter-mid-parent regressions produced lower heritability estimates (as expected). None of the regression coefficients of daughters on mother, however, were significantly different from zero and the regression on parent explained only very small amounts of the activity variation in the offspring. Estimates obtained from intraclass correlations among offspring were higher and outside the acceptable range, reflecting the variance component due to the common environment. The predictive value of the heritability estimates appears to be very low. The frequency distribution of activity among individuals sampled from one insecticide treated site (AM) was skewed to the right, as previously reported. But samples from another site (TZ) showed a symmetrical distribution, unlike the previous pattern. When field collected individuals from AM were released on clean plants in the laboratory, samples taken from the plants one or more days later showed symmetrical EST distribution. Electro-phoretic paterns of field and laboratory samples were the same, but band intensity was stronger in samples from laboratory populations. Differences in mean EST activity between populations and sampling years were unrelated to methidathion resistance.

Seven presumed Mendelian enzyme loci (Est-2, Est-3, Gpi, Idh-l, Idh-2, Mdh-2 and Mpi) were characterized and tested for polymorphism in coffee berry borers, Hypothenemus hampei (Ferrari), sampled in Côte d′Ivoire, Mexico and New Caledonia. The average genetic diversity was H = 0.080. Two loci, Mdh-2 and Mpi were polymorphic, and thus usable as genetic markers. The population structure of H. hampei was analysed using Weir & Cockerham's estimators of Wright's F-statistics. A high degree of inbreeding (f = 0.298) characterized the elementary geographic sampling unit, the coffee field. The estimate of gene flow between fields within a country was Nm = 10.6 and that between countries was Nm = 2. The population genetic structure in H. hampei could be related to its known population biological features and history.

The possibility of density dependent feeding success in the African malaria vectors Anopheles gambiae Giles (sensu lato) and A. funestus Giles was investigated by examining the gonotrophic status of mosquitoes from 423 collections made in a Tanzanian village. Up to 500 mosquitoes were caught in any single 35 minute indoor resting collection. Estimates of the total numbers of mosquitoes resting were made by removal sampling. These indicated that the efficiency of resting collections decreased with increasing mosquito population. Of 1904 mosquito blood meals tested by ELISA, 95% were of human origin. There was no decrease in the proportion of mosquitoes fully fed at high population densities and the only demonstrable avoiding action by hosts in response to high densities of mosquitoes was the increased use of bednets during the wet season. The late biting cycle of these mosquitoes precludes many other density-dependent host defence mechanisms because the hosts are asleep at the time of attack. African malaria vectors have evolved with man and perhaps their evasion of density-dependent host responses has reinforced their preference for human blood.

The continual recruitment of new individuals makes it difficult to study both the survival of multivoltine mosquitoes, and the size of the infectious reservoir in narural populations of malaria vectors. During long-term surveillance of a population of Anopheles gambiae Giles sensu lato in a Tanzanian village by daily light trapping, a temporary dry spell resulted in the cessation of recruitment for a period of 33 days, and a decline in numbers of A. arabiensis Patton caught from over 2000 to less than 10 in a sentinel house. Traps placed elsewhere in the village indicated similar proportionate declines although numbers caught varied according to location. A survival rate of 83% per day was estimated from the rate of population decline. Survival was unrelated to the size of the mosquitoes. The infectious reservoir (the chance of a mosquito acquiring an infection) was estimated to be 2% per feed. The exploitation of fortuitous events which temporarily eliminate a single stage in the life cycle has general applicability in the study of the bionomics of multivoltine insects.

The daytime resting sites of adult Conopomorpha cramerella (Snellen), a serious pest of cocoa in south-east Asia, were identified as a possible target to which insecticide could be selectively applied. Field observations were made on 126 resting moths in a cocoa plantation in East Malaysia. All moths were found on the undersides of branches 45° or less from horizontal, and 1·5 cm or more in diameter; 90% were found in the lower half of trees. A field trial was conducted in which 25 ppm deltamethrin was applied with knapsack sprayers to these sites. Five applications were made at ten day intervals, each crop season (6 months), over a two year period. Mean estimated losses in the sprayed area were 12.1 to 33.7 percentage points lower than in the control area, giving an estimated net benefit of up to RM978 ha in one season, assuming a cocoa price of RM4000 per tonne. Losses in the sprayed area still ranged from 7.0% to 43·1% and did not decline over successive seasons. There was some evidence for immigration into the sprayed area, but it is suggested the rapid recovery of populations following spraying is also due to the high potential rate of increase of the insect in susceptible cocoa varieties. In susceptible cocoa varieties control is always likely to be difficult, but in less susceptible varieties, and with well maintained trees, chemical control by selective application to adult resting sites can be a practical component of integrated control.

Adults of the dung fly Neomyia cornicina (Fabricius) (Diptera: Muscidae) were allowed to feed on cattle dung containing known quantities of ivermectin for a single period of 24 hours. The concentrations required to kill 50% and 95% of the population, within one week of feeding, were calculated to be 0.139 and 0.393 μg/g (wet weight), respectively. Ingested ivermectin was found to reduce significantly the proportion of females ovipositing and delay significantly the onset of first oviposition. For those females which did oviposit, there was no effect of ivermectin ingestion on the number of eggs laid, but the eggs of females fed dung containing ivermectin at 0.15 μg/g showed a significant reduction in percentage egg hatch.

A study of the age-dependent changes in the ovarioles and the age structure of a population of adult females of Anopheles gambiae in the Amani hills of north-eastern Tanzania was undertaken using the ovarian oil injection technique. The greatest difficulties encountered in accurately determining the age of members of this species are due to small size and the low parity diagnostic index of the ovary, which decreased in successive gonotrophic cycles and could drop to zero even in the first cycles. Additional information can be obtained from basal bodies, groups of six to eight cells in the calyx wall enclosed by the end of the ovariolar sheath. The possibility exists of their use for age grading of mosquitoes. The mean age of the population during the study, which was carried out in the rainy season, was rather low. The oldest were individuals that had completed four ovipositions. Among the mosquitoes infected with sporozoites were females that had completed between ≥2 and ≥4 ovipositions.

The effects of the chitin synthesis inhibitor triflumuron, 2-chloro-N-[[[4-(trifluoromethoxy)phenyl]amino]carbonyl]benzamide, on the housefly, Musca domestica Linnaeus were examined in the laboratory. A dose of 1 μg of triflumuron applied topically to adult females of M. domestica resulted in egg hatch inhibition greater than 95%. At doses below 1 μg applied topically to adult females, subsequent hatched larvae were also highly sensitive to triflumuron; 0.5 μg applied topically produced 25% inhibition of egg hatch, but of the larvae that eclosed, only 20% survived to pupation. Hence, a dose of 0.5 μg applied to adults gave a total mortality of 85%. The effects of a single topical application of triflumuron declined over time in a dose dependent manner. After a dose of 1 μg, <5% of the first egg batch hatched, however, after 14 days egg hatch had risen to >90%. Topically applied triflumuron was most effective in inhibiting egg hatch when applied to females less than 3 days old. After 60 min tarsal contact with 20% triflumuron suspension concentrate on poly cotton cloth, 100% egg hatch inhibition was maintained for up to 1 week. Lower contact times, or triflumuron concentrations, gave lower percentage mortalities. The results suggest that triflumuron has considerable potential as a control agent for incorporation into autosterilizing devices for housefly control. However, further work is required to develop delivery systems capable of transferring an effective dose to females under field conditions.

Development and survival of an erigonine spider, Erigonidium graminicolum (Sundevall) (Araneae: Linyphiidae), were studied at seven constant temperatures ranging from 15°C to 35°C. Plotting development rates (reciprocal of mean times for development) and survival rates (expressed as percentages) against the different constant temperatures indicated that E. graminicolum is well adapted to low temperatures, but sensitive to high temperatures. Juvenile development occurred over a wide range of constant temperatures, but the survival rate was low at the upper and lower extremes. All eggs died before hatching at the highest tested temperature (35°C). There were statistically significant differences between males and females in juvenile and pre-imaginal development times. Juvenile development time decreased as temperature was raised to 32°C, but increased at 35°C. The mean duration of pre-imaginal development (egg to adult) varied from 94.4 d at 15°C to 23.8 d at 32°C. Considering development rate and survival, the most favourable temperature range appeared to be 15–32°C. For all life stages, graphs obtained by plotting the reciprocal of development times against temperature could be described by the six-parameter rate function of Wang et al. (1982). Temperature-independent distributions of normalized development times were approximated by a cumulative Weibull function. Frequency distributions of development times changed with temperature; they were almost uniform at 15–20°C, more symmetrical at 25–28°C, and somewhat skewed to the right at 30–32°C. The sum of effective temperatures (SET), and lower development threshold (LOT), were established for all life stages of E. graminicolum. Survivorship curves varied with temperature and, on average, were Type III.

Studies were carried out to estimate and compare mortalities of larvae and pupae of Aedes (Stegomyia) simpsoni sensu lato breeding in axils of Colocasia esculenta, Xanthosoma sagittifolium (Araceae) and Musa spp. (Musaceae) in different locations in Uganda and also to ascertain whether there are any differences in their dynamics that could be used to separate the species. The probability of larvae surviving from one day to the next (40·5 – 99·8%), and the proportions dying in each instar (0·97 – 77·3%) and dying daily in all the four instars (3·9 – 46·3%) varied between seasons at each site and between sites and species of plants. Estimates of larval and pupal mortality were generally highest for fourth instar [52·4 ± 6·0% (mean ± SE)] and lowest for the second instar larvae (16·0 ± 2·6%). There was higher larval instar mortality in axils of X. sagittifolium (96·9%) and Musa spp. (82·2 – 96·5%) than in C. esculenta (47·1 – 88·4%).

Populations of the soil insects African black beetle, Heteronychus arator (Fabricius), and whitefringed weevil, Graphognathus leucoloma (Boheman), and the damage caused to potato crops in south-western Australia were measured with and without the insecticide chlorpyrifos incorporated into the soil prior to planting. Low populations of both insects were highly damaging. Destruction of stems by H. arator reduced tuber yield and both species damaged tubers from the time of their formation. Tuber damage increased with time because both insect species on average damaged multiple tubers and caused multiple attacks on tubers. Growth of G. leucoloma larvae caused increased abundance of the more damaging later instars in spring. The insecticide reduced resident H. arator abundance and hence damage to newly-emerging potato stems in summer, but had less effect on reducing attacks on tubers in summer crops because adult beetles flew into some crops during growth. Exceptionally high tuber damage per H. arator adult in winter crops was ascribed to enhanced activity during its spring breeding season. The insecticide was inconsistent in reducing the abundance of G. leucoloma larvae. Insecticidal effects were greatest near the soil surface, resulting in an increase in the relative proportion of both insects deeper in the soil. Since a greater proportion of the tubers occurred there, the resulting greater potential for the insects to cause tuber damage tended to outweigh reduction in their abundance.

Uscana lariophaga Steffan is an egg parasitoid of the bruchid species Callosobruchus maculatus (Fabricius) and Bruchidius atrolineatus Pic in West Africa, which cause damage in stored cowpea (Vigna unguiculata (Leguminosae)). In a no-choice situation, the attractiveness of some odour sources to the parasitoid was studied in a simple diffusion olfactometer. Cox's proportional hazards model was used for data analysis. The results showed that cowpea beans were more attractive to naive U. lariophaga females than glass beads. Baits with host eggs were more attractive than baits without eggs. Experienced females did not perform any better than naive ones. The relationship between the inclination to fly of the parasitoids, and the baits used, was difficult to interpret. Changes in searching activity between the different weeks of experimentation were observed. Only a small part of this variation could be explained by (absolute) changes in barometric pressure.

The spatial dispersion patterns of larvae of the Palaearctic mosquito Aedes cantans (Meigen) were studied in two temporary woodland pools in England, between 1988 and 1990. Variance to mean ratios indicated that all larval instars were aggregated, but the degree of aggregation differed between ponds and between years, depending on the population density. The negative binomial model was found to fit the majority of the data, and the values for the parameter k varied between 0.05 and 0.96. Taylor's power law gave a good fit to the data with values of b, an index of aggregation, varying between 1.42 and 1.91. Iwao's patchiness regression yielded values for α between –1.63 and 6.92, and β from 1.49 to 3.25. These values indicate that the population is aggregated due to habitat heterogeneity, and younger instars may demonstrate a propensity to congregate with one another. A relationship between water temperature and mean number of larvae sampled at each transect point was observed in Pond One in 1989, suggesting that larvae aggregate where water temperatures are higher.

The African species of Eurystylus are revised and their status as pests of important agricultural crops is reviewed. A key is provided for the 12 recognized species, one of which, E. lestoni, is described as new. A diagnosis and description are provided for each species, as well as illustrations of diagnostic taxonomic characters and information on host-plant associations and distribution. Lectotype designations are made for eight nominal species of Eurystylus and two species are removed from the genus. The following new synonymies are established: E. annulipes (Poppius)= E. annulipes impunctatus (Poppius), syn. n.; E. bellevoyei (Reuter)= E. parvulus(Reuter), syn. n., E. occidentals Wagner, syn. n., E. gallouedeci Villiers, syn. n. and E. argenticeps Odhiambo, syn. n.; E. lineaticollis Poppius=E. pallidiventris Poppius, syn. n. and E. mutabilis Bergoth, syn. n.; E. oldi Poppius=E. kivuensis Schouteden, syn. n., E. risbeci Schouteden, syn. n., E. phaiopterus Odhiambo, syn. n., E. immaculatus Odhiambo, syn. n. and E. bilqis Linnavuori & Al-Safadi, syn. n. Two species are removed from the genus Eurystylus(E. sobrius (Walker)= Plesiocapsus sobrius (Walker), comb. n. and E. sericeus (Walker)=Proboscidocoris sericeus (Walker), comb. n.) and one species, E. rufocunealis Poppius, is assigned the status of nomen dubium.

In Zimbabwe, studies were made of the levels of known tsetse attractants present in natural ox odour. Typically an ox (400 kg) produced phenol (0.1 mg/h), 3-methylphenol (0.09 mg/h), 4-methylphenol (0.7 mg/h), 3-ethylphenol (0.01 mg/h), 4-ethylphenol (0.02 mg/h), 3- and 4-n-propylphenol (0.02 mg/h), l-octen-3-ol (0.01 mg/h), carbon dioxide (140 l/h), acetone (5 mg/h) and butanone (0.3 mg/h). Of these, only phenol, 4- and 3-methylphenol and carbon dioxide were always detected in ox odour. Studies were made of the numbers of Glossina pallidipes Austen and G. morsitans morsitans Westwood attracted to natural ox odour and synthetic odour, the latter consisting of blends of identified attractants dispensed at the doses naturally present in ox odour. Natural ox odour caught twice (P < 0.05) as many G. pallidipes and 1.5 (P < 0.05) times as many G. m. morsitans as the synthetic blend suggesting the presence of an unidentified attractant in ox odour. Passing ox odour through filters indicated that all attractants can be trapped on a combination of charcoal and sodalime filters but the unidentified attractant(s) may pass through a sodalime filter, and break through a charcoal filter used for more than 6 h. Increasing the dose of ketones in the synthetic odour from 2 to 100 mg/h doubled the catches at the source. Increases in ketone levels in hosts, induced by starvation or possibly trypanosomiasis, may increase attraction of tsetse to such animals.