Mononychellus tanajoa (Bondar), a spider mite new to Africa, was discovered on cassava in Uganda in 1971. This exotic mite of Neotropical origin has since spread to 27 countries, causing an estimated 13–80% reduction in cassava yield. The history of its introduction and spread in Africa is reported, and present knowledge on the taxonomy, biology, ecology and pest status of the mite, and the evolution of strategies to control it are reviewed. While attempts have been made to control M. tanajoa, it remains a serious pest in most areas. The prospects for classical biological control as an appropriate control strategy are discussed.

The braconine Palabracon diatraeaphilus gen. et sp. n. is described and illustrated from a unique female collected in Ecuador. The new species was reared as a parasitoid of an undetermined species of Diatraea.

Residues of pirimicarb or deltamethrin plus heptenophos on potatoes in field cages in southern England reduced artificial infestations of susceptible and insecticide-resistant strains of Myzus persicae (Sulzer) introduced up to 14 days after treatment. Deltamethrin plus heptenophos gave better control than pirimicarb, particularly of nymphs, but also selected more strongly for very resistant (R2) aphids. Susceptible (S) aphids survived the treatment with deltamethrin plus heptenophos better than moderately resistant (R1) aphids, probably due to a different or more pronounced behavioural response to the pyrethroid. However, the combination of the greater persistence of deltamethrin plus heptenophos and the high resistance of M. persicae to the pyrethroid led to more prolonged selection for R2 aphids, which is sometimes associated with a resurgence in aphid numbers.

From March to October 1984, targets consisting of black cloth and netting, baited with 1-octen-3-ol (released at about 0·5 mg/h), and acetone (about 100 mg/h) or butanone (15 mg/h), and coated with deltamethrin, were deployed at 3–5/km2 in 600 km2 of the Zambezi Valley of Zimbabwe where Glossina morsitans morsitans Westwood and G. pallidipes Austen were initially abundant. About 3 000 000 adults of G. pallidipes and 200 000 of G. m. morsitans originally present in the area as adults or pupae were killed by targets or removed by sampling procedures between March and December. At the end of this period, the tsetse populations in the centre of the block had declined by at least 99·99%. The targets dealt adequately with a strong invasion pressure from initially dense infestations nearby, partly because the targets reduced the abundance of tsetse up to 5–10 km outside the block. At 18 months after the start of the study, the targets were badly faded; this was corrected by spraying them with a black dye and an ultraviolet-light absorber that protected the dye and insecticide, but by then the targets had deteriorated and were without adequate odour attractant, and many were no longer being maintained. Tsetse then invaded further into the block, but only in small numbers. Tabanids and muscoids were not strongly attracted to targets; their population densities in and near the block did not change greatly. Targets offer a simple and ecologically clean method of controlling tsetse and preventing invasion.

Age structures and reproductive anomalies of two field populations of Glossina pallidipes Austen in Zimbabwe were compared. One population was treated with insecticide sprayed on odour-baited targets, while the other was left undisturbed. Both populations were sampled with F3 prism-type traps. Capture probabilities were higher for females in early stages of pregnancy than for those in later stages. Per-capita mortality rates were estimated from age distributions which had been corrected for rate of increase. The mortality rate steadily increased over time in the insecticide-treated area but remained roughly constant in the undisturbed area. Estimates of mortality had wide confidence intervals, and the mortality model is critically reviewed. No evidence was seen in the treated population for an increased incidence of abortion or other reproductive anomalies, or for a decrease in insemination rate.

The effect of dung burial by the African dung beetle Catharsius tricornutus De Geer on the survival and size of the dung-breeding African fly Haematobia thirouxi potans (Bezzi) was examined in the laboratory at 25°C using 1-litre pads of cattle dung. The experiment included a set of dung pads which were exposed to infrared radiation for 12 h per day to mimic the heating effect of solar radiation. C. tricornutus is a large (1–2-g live weight) dung beetle, and moderate numbers can completely bury dung pads within 24 h of pad colonization. Dung burial was proportional to the number of pairs present (range 0–8 pairs per pad), but the level of burial was reduced in the presence of infrared radiation. Dung pads lost moisture with time, and the rate of desiccation of unburied dung increased with the amount of dung burial and under infrared radiation. Mortality in the immature stages of H. t. potans increased with increasing levels of dung burial and with exposure of pads to infrared radiation. The timing of addition of dung beetles (i.e. before or after egg hatch) did not affect fly survival. At moderate levels of dung burial, fly larvae were killed directly by beetle activity and at high levels of dung burial, fly larvae also perished due to an unfavourable physical environment associated primarily with moisture stress. Infrared radiation killed immature flies and increased the level of moisture stress. There was a low level of survival in dung buried by the beetles under these unnatural laboratory conditions. Stunting of surviving flies was associated with reduced levels of moisture in the core of the unburied dung and was observed when the level of dung burial was >60% in ambient conditions and >40% under infrared radiation.

Clones of Sitobion avenae (F.) collected in southern England in 1983 and 1984 were subjected to sexual-inducing conditions in the laboratory (12 L:8 D photoperiod at 15°C and/or 8 L:16 D photoperiod at 12°C). Most clones were continuously parthenogenetic under these conditions; some clones produced males only (androcyclic), and the same behaviour was observed under field conditions in the autumn of 1983. In contrast, over 90% of the clones collected from Scotland in 1984 were holocyclic (i.e. cyclically parthenogenetic). Fundatrices from intra- and interclonal matings were obtained. Those from interclonal matings were significantly more fecund than fundatrices from intraclonal matings. Examination of the descendants of these fundatrices indicated that both life-history strategy and migratory tendency were genetically-determined.

Inundative releases of the indigenous egg parasitoid Trichogrammatoidea cryptophlebiae Nagaraja were made against Cryptophlebia leucotreta (Meyrick) in two sweet orange orchards in the Transvaal lowveld, South Africa. The rate of parasitism was raised to similar levels in the two release areas, but there was a significant decrease in crop damage in only one of them. Although the larval population was twice as large in a control orchard, the losses there were compensated for by a larger total crop. Three explanations for the different performances of the parasitoids are examined. Movement away from release areas was not apparent. Rates of parasitism at the tops or bottoms of trees or at the four cardinal aspects were not significantly different overall but should have been to compensate for the linear increase in numbers of host eggs with height, and the larger numbers on the warmer northern and eastern sides of trees. Parasitism was higher at the tree tops than at the bottoms in the orchard that lost the least fruit. The stochastic processes governing host-egg distribution on fruit were also important. Pest density was higher in the release areas than the control, and eggs were distributed more contagiously in the orchard with the most crop loss.

Through repeated sampling in maize planted on different dates over different seasons in South Africa, it was shown that damage by Busseola fusca (Fuller) to plant parts other than the whorl had an important influence on yield. Damage to plants in the period after tasselling was shown to be important. The number of larvae in the plant was a weak estimator of expected yield losses.

The influence of natural enemies on the population growth of Diuraphis noxia (Mordvilko) in wheat fields in eastern South Africa was investigated. Since the natural enemies were not present in the initial stages, the apparent mortality observed must have been due to other causes. During the logarithmic phase of aphid population growth, predators, in particular the coccinellid Hippodamia variegata (Goeze), had a slight inhibitory effect on growth. During the decline phase of the aphid population, the potential population size was reduced by as much as 52% by predation. The most abundant parasitoids, Aphidius colemani Viereck and Diaeretiella rapae (M'Intosh), had a negligible effect. Although natural enemy populations showed functional responses to increases in Diuraphis noxia numbers, they did not prevent the latter from attaining and exceeding economic injury levels.

Suction traps sampling at a height of 12·2 m have been operated by the Rothamsted Insect Survey since 1964 to monitor the aerial movement of aphids and other insects. Many variants of the trap have evolved in other countries due to the lack of a detailed published description. A standard for a low maintenance structure of the ‘12-metre' suction trap now being adopted by the Rothamsted Insect Survey and some other users is described. This should lead to better uniformity and comparability of data from such traps.

Metaphycus baruensis sp. n., an encyrtid parasitoid of Coccus spp. on coffee in Kenya, is described and compared with related species. The parasitoid is to be introduced into Papua New Guinea in an attempt to control green scales which are serious pests of coffee there.

Introduction of parasitoids gave improved biological control of Planococcus citri (Risso) in a large glasshouse stocked with a variety of ornamental plants, supplementing that achieved by the predator Cryptolaemus montrouzieri Mulsant. Following parasitoid release, there was evidence of pest population regulation on guava and coffee bushes with reduced and stabilized mealybug numbers and stable percentage parasitism. The mean temperature during one sampling period was significantly correlated with the percentage parasitism two months later, indicating that temperature has a major impact on parasitoid efficiency. Leptomastidea abnormis (Girault) was responsible for about 90% of the parasitism observed; the remainder was by Leptomastix dactylopii Howard. The effects of supplementary inoculative releases of Leptomastidea abnormis were variable, but resulted in increased percentage parasitism only when the number released contributed significantly to those already established. No evidence of pest population regulation was found on Aristolochia or Passiflora vines nor on potted Gesneriaceae plants of the genera Streptocarpus and Saintpaulia. Peak mealybug populations on these last plants coincided with periods of new growth, fruiting or flowering. Host-plant quality thus seems to be a major factor inducing mealybug outbreaks.

Females of Glossina tachinoides Westwood captured three-hourly in biconical traps deployed from 16.30 to 18.30 h weekly in three peridomestic agroecosystems in southern Nigeria from April 1984 to March 1987, were dissected and classified as nulliparous, pregnant, aborted or parous-but-not-pregnant. In all three tsetse populations, the monthly pregnancy rate ranged from 76 to 94% and the abortion rate from 3 to 8% throughout the year. No seasonal or locational differences or evidence of seasonal increase, reduction or cessation in breeding were demonstrated. A significant difference between years with respect to monthly abortion rate during the wet season was possibly due to changes in vegetation structure which is dependent on crop types. It is concluded that the low population density and pattern of seasonal fluctuation in apparent density of G. tachinoides in these peridomestic agroecosystems are due to causes other than seasonal differences in its breeding rate.

The shoot-fly Atherigona ramu sp. n. is described from Papua New Guinea. It was reared from sugarcane at the agronomy centre of Ramu Sugar Ltd, Gusap, Papua New Guinea, and is also known from other parts of Papua New Guinea.

Several published reports have presented estimates of the rate of increase, r, based on sampled ovarian age distributions from Glossina populations throughout Africa. These estimates are invalid, because an age distribution sampled at one point in time can be equated to a survivorship curve only if r = 0. When such a survivorship curve and a corresponding fecundity schedule are then used to estimate r via the Euler-Lotka equation, the result is a value of r near zero, regardless of the population's true rate of increase. Synthetic sampling from a hypothetical tsetse population confirmed that estimates computed in this fashion are entirely the products of sampling error. Valid estimates of r can sometimes be obtained from an age distribution, using an alternative method, but such estimates are highly sensitive to sampling errors in the distribution.

An immunoassay for the carboxylesterase E4, which confers insecticide resistance in Myzus persicae (Sulzer), was used to monitor the distributions of E4 activity within samples comprising more than 5000 aphids from sites in eastern and southern England. The quantitative precision of this assay is an improvement on the subjective assessment of E4 band intensity on electrophoresis gels previously used. Activity distributions did not always show discrete components corresponding to S, R1, R2 or R3 variants and thus, though convenient, this classification should be used cautiously. Distributions derived by immunoassay of many insects from the field are best compared to those of laboratory-reared standards classified according to E4 quantity (V1-V64). Data from untreated crops are presented in this way and discussed in the light of previous surveys. A large proportion of aphids corresponding to the V8 variant (intermediate between ‘R1’ and ‘R2’ categories) were found, and frequencies of resistance differed on both a local and regional scale. The majority of aphids from treated crops had E4 activities equivalent to, or above, that of V16 (i.e. ‘R2’).