Preparations made from the Australian acridine grasshopper Caledia captiva (F.) that contained a recently discovered cytoplasmic polyhedrosis virus (CPV) were used to treat Locusta migratoria (L.). Feeding of the locusts with the preparations caused a significant increase in mortality. Although these deaths were associated with the viral treatment, their ultimate cause appeared to be uncontrolled proliferation of a bacterium, which is probably a strain of Enterobacter cloacae and which is usually found in locusts without ill effect. Four areas are identified where further work is necessary before the potential of the C. captiva CPV as an agent for the biological control of L. migratoria can be assessed.

In Tanzania, residual populations of Spodoptera exempta (Walker) may persist in highlands and river valleys, giving rise to the first outbreaks in the rainy season. To examine whether parasitoids could be a key factor in maintaining these populations at low density, samples were taken during outbreaks in eastern and north-eastern Tanzania from 1980 to 1984. Parasitism was high (around 50%) in the low density samples (below 25 larvae/m2) and low (below 17%) in the highest density ones (about 250–1200 larvae/m2). The number of outbreaks reported from the same administrative district as the sample site, and concurrent with it, was inversely related to parasitism level. The directly density dependent relationship between percentage parasitism and host density at lower densities, and the inversely density dependent one at higher densities, is similar to that found in parasitoid complexes of other Lepidoptera. Parasitoids in the rare braconid genus Meteoridea were recorded attacking S. exempta; this is the first published record from mainland Africa.

The ecology of Aedes albopictus (Skuse) was studied in a rubber estate near Kuala Lumpur, Malaysia, in 1982–84. The rainfall fluctuated from month to month and year to year. However, there was close correspondence between rainfall and the numbers of eggs, larvae and pupae, and females throughout the two years. The mortalities of larvae and pupae observed were 87·5% in 1983 and 73·5% in 1984. Larvae of the dytiscid Lacconectus punctipennis Zimmermann were the major predators of the mosquito larvae, and the spider Heteropoda venatoria (L.) was the main predator of the adults. A species of Coelomomyces was found infecting the ovaries and an unidentified species of neogregarine was seen in the Malpighian tubules. The presence of Dirofilaria immitis larvae in the proboscis of A. albopictus was an indication that it is a potential vector of diseases of veterinary importance in Malaysia.

A simple field method for direct detection of an altered acetylcholinesterase (AChE)-type organophosphate and carbamate resistance mechanism in individual insects was developed. The test will allow accurate differentiation by eye, on the basis of a colour change, of homozygous-susceptible and heterozygous or homozygous-resistant individuals. In initial field trials of this test in Sri Lanka, Anopheles nigerrimus Giles and Culex tritaeniorhynchus Giles were shown to contain this type of resistance gene, while a further nine species of mosquito had no evidence of it. Frequencies of the altered AChE gene in A. nigerrimus were estimated to be higher than those of C. tritaeniorhynchus, where the gene was found in only one of three localities in the island. The implication of the increase in prevalence of the AChE resistance gene in mosquito populations is discussed.

Four geographical strains of Ceratitis capitata (Wiedemann) were subjected as third-instar larvae to mass selection with dieldrin. Two strains, SOUTH and COSRIC, responded well but continued to show substantial heterogeneity in tolerance to the insecticide after 21 and 25 generations of selection, respectively. The technique was altered to single family sib selection, resulting in the isolation from the SOUTH strain of a dieldrinresistant strain (73) showing a more homogeneous response to the insecticide. By the same technique, a dieldrin susceptible strain (72/3) was isolated from the COSRIC strain. A decline in resistance was observed in the 73 strain between generations three and four, but the level subsequently increased again, without selection, reaching 68-fold in generation seven. It remained high for the next two years. No significant change was observed in susceptibility to dieldrin of strain 72/3; it remained highly susceptible. Resistance to solutions of technical dieldrin (96%) was shown to be mainly under the control of a single major autosomal gene, intermediate in dominance but with a modifying effect of the genetic background on the expression of the resistance gene. The resistance heterozygotes, segregating in backcrosses to susceptible individuals, were shown to be more resistant than expected, while in backcrosses to resistant ones they were more susceptible. A dose of 11 p.p.m. dieldrin was found to discriminate almost completely between heterozygous and homozygous-susceptible individuals among backcross progeny. In tests using a purer form of the insecticide (99·7%), a similar pattern of resistance was observed. The implications for genetic sexing in C. capitata are discussed.

The sensitivity of males and females of Glossina morsitans morsitans Westwood to the vapour phase of bisazir was tested in the laboratory in a moving air stream at temperatures ranging from 12 to 20°C. Similar tests were conducted in still air in closed systems. Bisazir was used in its granular form or absorbed onto silicone rubber or cellulose fibre surfaces from solution in dichloromethane. As a crystalline solid, it lost only 5% of its weight in a moving air stream over 12 days. However, even in relatively still air, bisazir absorbed onto cellulose fibre discs lost 70% of its weight at 20°C and 80% at 30°C in 24 days. Estimations of the amounts of bisazir present on a surface by weight difference were higher than those estimated by gas chromatographic analysis. Hence, in addition to being volatile, bisazir must be regarded as somewhat unstable. Nevertheless, sufficient active ingredient remained over a long enough period for it to be considered as a practical means of sterilizing tsetse in the field. Females were found more sensitive to bisazir than males, but exposure times required to produce total sterility in both sexes did not affect survival. Exposure times of only a few minutes were sufficient to induce a significant reduction in female fecundity or male fertility at 30–35°C, temperatures which might be expected under field conditions in a device to sterilize automatically tsetse attracted to traps in the field. The degree of sterility induced was permanent, and at these short exposure times the temperature had no effect on female fecundity.

A time series analysis of noctuid and sphingid moths caught at light traps in East Africa revealed the existence of sources of periodic variation in catch with three different wavelengths: lunar, annual and four-year. A comparison of catches of Spodoptera exempta (Walker) at different sites showed that site characteristics influence a light trap’s catch and confirmed the northward advance in flight period with season in the early part of the year. This species apparently goes through periods of increased abundance every four years. Of eight other species examined, all but one showed signs of a four-year cycle in abundance, although the peaks occurred in different years. Catches of all nine species were reduced by moonlight, but not equally. The size of the insect influenced its susceptibility to interference by moonlight. This was apparently mediated by the height of flight. High and low flying moths were more strongly influenced by moonlight than those at intermediate heights.

Various ketones and 1-octen-3-ol were used as odour attractants with biconical traps at Nguruman, south-western Kenya, to assess their effects on the catch size and composition of Glossina pallidipes Austen. Acetone, methyl ethyl ketone and 1-octen-3-ol were found to be effective in increasing catch size by up to 2–4×, but the two ketones differed in their dose-response curves. Acetone was more effective for males than females, as was 1-octen-3-ol. The age composition of samples from baited and unbaited traps differed only slightly. The level of variability between traps was reduced by the use of odour. Acetone with cow urine produced increases in catch of 9–25×, and significant differences were found between samples in sex ratio and age composition. The index of increase for cow urine plus acetone, when used with a biconical trap, was correlated with temperature changes during the day. This could only be partially accounted for by variability in dose rates. Cow urine with acetone is considered to be a potentially useful combination of baits for community participation tsetse control schemes.

Energy dispersive X-ray spectrometry was used to make quantitative determinations of the elemental composition of plasma-ashed apterous and alate individuals of Rhopalosiphum padi (L.). The aphids were collected as fundatrigenae at various localities in England and Scotland, mostly from their main overwintering host, Prunus padus, but at one site also from P. virginiana. Robust-means principal components analysis of the data confirmed that there were differences, unrelated to source, between the elemental composition of apterae and alatae, with lower concentrations of S, K and Cl in the alatae. There were, however, no clear separations of sources in either morph except between some extreme groups; it is possible that genetic uniformity limits variation in elemental composition and, therefore, the existence of detectable source-related chemoprints in R. padi. Elements that provided some discrimination were S, Cl, Ca, Zn and possibly Al, and, except for S, their mean concentrations were all less than 1% of weight of the sample examined; difficulties in detecting differences at these low concentrations may have prevented detection of source-related chemoprints in R. padi.

Seasonal abundance of fleas infesting small mammals was studied from July 1975 to August 1976 in a grassland in Lake Nakuru National Park, Kenya. Ten species of rodents and two species of shrews were live-trapped 935 times. The Nile grass rat (Arvicanthis niloticus) was the most abundant species, with 233 individuals captured 690 times, forming 73·8% of all captures. From these grass rats, 4247 fleas were collected, including Dinopsyllus lypusus Jordan & Rothschild (n = 2596), Xenopsylla bantorum Jordan (n = 1396) and Ctenophthalmus calceatus cabirus Jordan & Rothschild (n = 255). D. lypusus and C. calceatus cabirus were more abundant on Nile grass rats during the rains, while ×. bantorum became more abundant during the dry season and peaked during the early part of the rains. The potential for plague transmission by fleas between Nile grass rats probably exists all year. The potential for transmission to people living in close proximity to these rodents may be highest during the seasonal peak of X. bantorum, at the end of the dry season.

The citrus pest Scirtothrips aurantii Faure shows a positive phototactic response to yellow surfaces, particularly fluorescent yellow with peak reflectance at about 525 nm. A fluorescent yellow sticky trap was devised. Impact of individuals on the trap surface was principally random. Taylor's power law gave a generalized value of 1·42 for slope of b. Widespread geographical locations gave similar values for biologically significant values of b. Habitat differences within one geographical area influenced b, which was highest in natural vegetation and lowest in certain citrus monocultures. Three traps per orchard (>5000 trees) were practicable, and produced statistically realistic results. The period during which S. aurantii can damage fruit was divided into a longer risk period and a shorter, central critical period. Where S. aurantii was present on the traps at less than nine individuals per set of three traps throughout the central critical period and 20 during the late risk period, the percentage of fruit so damaged as not to be of export grade was less than 1·0. This method of monitoring was sufficiently robust to allow for some human error in reading the traps under commercial conditions, and to permit variability in placement of the traps in the orchard.

With the decline in production of new insecticides, the development of resistance is likely to become more of a problem than at present. Models of the spread of resistance genes may become integral parts of many management programmes. A critical examination is made of the classical theory of inheritance of resistance, i.e. the spread of a single resistance- conferring allele through a population. The theoretical considerations suggest that simple Mendelian inheritance is unable always to account for the development of resistance. It is suggested that some resistance factors may be under multigenic control and that such inheritance could account for the rate at which resistance sometimes develops. Following on from the theory, an analysis was made of time to knockdown and dry body weight in Sitophilus oryzae (L.). A half-sib experimental design was employed which enabled quantitative analyses of the characters to be made. Both characters had substantial heritabilities and were phenotypically correlated. The analysis also indicated a positive genetic correlation between the characters. The implications for the development of tolerance in the light of the genetic analysis are discussed.

Three populations of Aonidiella aurantii (Maskell) were studied on orange trees at two sites in the hot Transvaal lowveld in South Africa. The first two were on the same trees, which had severe scale in the tree crowns and a low infestation in the bottoms. The third population, at a second site, was high and relatively evenly-spread over the whole tree. In previous years, the parasitoids Aphytis africanus Quednau and A. melinus DeBach had been unable on their own to reduce the high scale populations. Once a certain scale density level on the tree framework had been breached, a proportionately large number of susceptible scales escaped parasitism and reached the gravid female stage, which is invulnerable to attack. The ineffectiveness of parasitoids in these circumstances, resulting partly from their narrow niche breadth, was compounded by peak numbers only being reached late in the season. By this time, the large number of adult scales had already rendered the fruit unmarketable. In both sites, the predator Chilocorus nigritus (F.) was introduced and the trees also received commercial applications of petroleum oil. In one site, this oil treatment was restricted to the tree crowns. Although overall parasitism of vulnerable stages was relatively unaffected by these abiotic and biotic controlling factors, there was a shift to increased percentage parasitism of the virgin female away from the other susceptible stages. In addition, it was confirmed that the broad attack niche provided by C. nigritus, which preys on all developmental stages of the scale, readily complements parasitoid activity and can be of major economic importance.

In Zimbabwe, the catches of Glossina pallidipes Austen and G. morsitans morsitans Westwood in traps were roughly doubled by the presence of two bushpig bedding sacks or jars containing 200 ml of urine from ox or buffalo. The main effect of these residues was to increase the trap-entering responses of tsetse once they reached the trap; in addition, the urine brought G. pallidipes to the traps. The efficacy of the residues persisted for at least two months after initial exposure, although urine on soil was ineffective after two days. Efficacies were not affected by the presence of 1-octen-3-ol, acetone or butanone as additional attractants but were lower in the presence of carbon dioxide and lower with electrified visual targets than with traps. A ten times increase in dose caused no clear increase in catches, ten jars of urine being repellent, but catches increased when the urine and sacks were used together.

Rastrococcus invadens sp. n. has been accidentally introduced to West Africa where it is causing considerable damage to many plants, especially mango and citrus, in Togo and Ghana. The origin of the mealybug is the Oriental Region, where it is known from a wide area, but it has always been mistaken for R. spinosus (Robinson), a closely related species. Serious outbreaks on mango in some areas of the Oriental Region suggest recent introductions also, although records from other parts of the Oriental Region are as early as 1900.

Cryptolaemus montrouzieri Mulsant is a well-known and historically important predator of mealybugs. It is indigenous to eastern Australia, and has been introduced throughout the world for biological control. In contrast, the other members of the genus, including those which resemble C. montrouzieri, remain poorly known. The genus Cryptolaemus, with seven previously described species, is redefined and two species-groups are recognized. The first, the montrouzieri group, comprising C. montrouzieri, C. affinis Crotch (Indonesia and Papua New Guinea and northern Queensland, Australia). C. wallacei Crotch (Indonesia) and C. simplex Blackburn (northern Australia), is revised; C. crotchi sp. n. (Indonesia) and C. sinestria sp. n. (Indonesia, Papua New Guinea and Solomon Islands) are described as new, and C. simplex is reduced to a subspecies of C. montrouzieri. A key to these species is provided. The second group comprises two of the remaining species. C. subviolaceus Crotch and C. concinnus Weise, but lack of sufficient material precludes a revision of this group at present. The final species, C. engelhardi Gorham. is transferred to the chilocorine genus Brumoides Chapin. Lectotypes are selected where appropriate.

Measurements of several parameters of demographic performance for 16 field-derived Australian populations of Phthorimaea operculella (Zeller) were obtained under laboratory conditions. Fecundity, female weight and low temperature development rates were the most important parameters contributing to differences between populations. It is suggested that the increased fecundity and female weights found in warmer regions may be selected for in part by the continuous high mortality imposed by parasitoids, which have a reduced impact in the colder parts of P. operculella's range. A more rapid development rate in populations from colder areas is probably due to selection for a decreased lower threshold temperature in these areas, indicating better survival characteristics under cold conditions.