To explain differences between years in outbreak development of the African armyworm, Spodoptera exernpta (Walker), in eastern Africa, the availability to host-plants of soil nitrogen, phosphorus and potassium was studied in a primary outbreak area during the first month after the start of the short rainy season in relation to the drought intensity during the preceding year. The weather-soil-plant interactions observed during four successive years (1988–1991) indicated that nitrate levels, especially in soils not extremely poor in organic carbon and total nitrogen, can differ greatly between years in re sponse to differences in drought intensity and rainfall pattern. These resulted in large differences (i.e. up to 2.0% dry wt.) in organic nitrogen levels in leaves of host-plants of S. exempta. Phosphorus and potassium levels in soils and host-plants showed no consistent relationship with the weather pattern over the years and in general their differences between years were relatively small. The occurrence of S. exempta during the first two months after the start of the short rainy season was well associated with plant nitrogen levels in the primary outbreak area: outbreaks developed only in years with at some locations organic nitrogen concentrations of almost 5% in young grasses. If fitness of S. exempta is strongly dependent on the organic nitrogen content of its food within the range encountered in this study, weather conditions during the long dry season might indirectly explain the observed differences in subsequent outbreak development via their delayed effect on host-plant nitrogen levels.

The frequency of defecation in relation to constant0/simulated outdoor temperatures and larval age (instar) was examined under laboratory conditions in four species of pine sawflies (Diprionidae): Gilpinia frutetorum (Fabricius), Diprion pini (Linnaeus), D. similis (Hartig) and Macrodiprion nemoralis (Enslin). Defecation interval and total faecal production was regular under higher constant temperatures (higher production in younger instars, second and third) and production significantly decreased under varying air temperatures during the same period (24 hours) when larvae of younger instars were influenced more by lower temperatures. The influence of larval age (instar) on faecal production was insignificant if larvae were exposed to varying temperatures during 24 hours. Faecal production and temperature were positively correlated in each of the four species and instars tested, as well as in total (all the test species and instars combined). Implications of the results for estimating canopy larval density by frass-drop measurements in the field are briefly discussed.

Zygogramma bicolorata Pallister was introduced for biological control trials against the weed Parthenium hysterophorus (Asteraceae) in India. The insect entered diapause over an extended period of time between July and December in Bangalore. Diapausing adults burrowed into the soil, and emerged in May–June with the commencement of monsoon rains. Percentage diapause increased over time, peaking at 72% during November. Non-diapausing adults were capable of breeding, under laboratory conditions, during the winter. Some adults bred both before and after diapause, during two consecutive years. Soil moisture played an important role in providing the conditions for burrowing or emerging from the diapause chambers. The studies also showed that diapausing adults had to be exposed to the high summer temperatures, for termination of the diapause. It was possible to break diapause by continuous exposure to 30°C, 35°C and 40°C for 22 days, nine days and 10 hours, respectively, during February–March, about three months after its inception. This method can be used to initiate mass multiplication for carrying out releases early in the season.

Pareuchaetes aurata aurata (Butler) from Chromolaena jujuensis in northern Argentina was tested as a biological control candidate for the composite weed C. odorata. The larvae feed voraciously and complete their development on C. odorata. The host-plant feeding range of P. a. aurata is limited to the genus Chromolaena. It is suggested that the egg-laying behaviour of P. a. aurata, which scatters its eggs around the base of the host-plant, will help to overcome the problem of ant predation which prevented the establishment of Pareuchaetes pseudoinsulata Rego Barros, which lays its eggs in batches, in South Africa. Any possible benefits of the new association between P. a. aurata and C. odorata may also contribute to the success of this biological control programme.

Bread and durum wheat varieties varying in wheat stem sawfly resistance were sown at different densities in northern Syria for two seasons to examine the effect of density on plant characteristics related to infestations of the wheat stem sawflies Cephus pygmeus (Linnaeus) and Trachelus spp. Plants sown at low densities had a longer interval between developmental stages and had higher stem solidity than those sown at high densities. Stem solidness was negatively correlated with the percentage of sawfly-infested stems. Plant height and grain yield were neither strongly nor consistently correlated with wheat stem sawfly infestation but were influenced by variety and density. These results differ from those of previous studies in North America where widely spaced plants suffered higher infestations than did closely spaced plants. Results of our study suggest that stand density should be considered when recommending wheat stem sawfly-resistant wheats to farmers in west Asia and North Africa and when designing screening trials for wheat stem sawfly resistance in the field.

The effect of cyromazine on the mechanical properties of the larval integument of Lucilia cuprina (Wiedemann) was measured using third instar larvae 2–3 h after moulting placed on liver containing 1 ppm cyromazine. Cuticle stiffness (Δ force (N)/δ strain) was approximately 20–30% higher in treated larvae (relative to controls) after 6–10 h treatment, after which time stiffness decreased to levels well below controls. Larvae placed on 1 ppm cyromazine early in the second instar also showed a short period of slightly elevated cuticle stiffness (relative to controls) after 3–5 h treatment. Third instar larvae placed on cyromazine-treated diet for 8 h then transferred to normal diet showed the same short period of elevated stiffness (at 6–10 h) followed by rapid decrease in stiffness to levels well below controls and greatly reduced growth as larvae fed cyromazine continually. A slight increase in cuticle stiffness appears to be only an early and temporary effect of cyromazine poisoning in L. cuprina. These results are discussed in relation to the previously documented effects of cyromazine on the mechanical properties of the larval cuticle of Manduca sexta (Linnaeus).

Zygogramma bicolorata Pallister (Coleoptera: Chrysomelidae) of Mexican origin, introduced for biological control trials against the weed Parthenium hysterophorus (Compositae), was found to feed on the tender leaves of the related plant sunflower (Helianthus annuus Linnaeus) at a few places in Kolar district in Karnataka State, India. The insects in these populations exhibited variability in feeding response to sunflower under laboratory and field conditions. The sesqueterpene lactone parthenin, specific to P. hysterophorus, was found to induce feeding by the adults of Z. bicolorata on treated sunflower leaves within six hours, as compared to 4–7 days on untreated leaves. Further investigations resulted in the novel finding that dusting of parthenium pollen, or smearing its extract which contains parthenin, on sunflower leaves also induced feeding responses in adults. Parthenium pollen was detected on sunflower leaves growing in close proximity to the weed stand, suggesting that this could be one of the factors influencing variability in the feeding behaviour of Z. bicolorata.

Megalurothrips sjostedti (Trybom) is an important pest of cowpea (Vigna unguiculata) in West Africa. Three key factors assumed to be responsible for its pest status are analysed, the survival on alternative host-plants during the dry season, the inefficient biotic mortality factors regulating population growth, and the effect of larval feeding on the development of cowpea flower buds. Extensive surveys indicate clearly that M. sjostedti survives the dry season on a wide range of alternative hosts all belonging to the Leguminosae, where it can feed and reproduce. Different antagonists were observed attacking eggs and larvae of M. sjostedti; their impact, however, is low and cannot prevent pest outbreaks. Two undescribed Megaphragma spp., and one Oligosita sp., all trichogrammatid egg parasitoids, were recorded for the first time. The anthocorid Orius sp. was the most important larval predator. No hymenopterous parasitoids could be reared from larvae collected on cowpea and three major alternative hosts, whereas a low percentage of the larvae collected from the flowers of Tephrosia candida, an exotic shrub native to India, were parasitized by the eulophid Ceranisus menes (Walker), also recorded for the first time in Africa. The feeding activity of six larvae of Megalurothrips sjostedti during five days induced the shedding of all flower buds of a cowpea inflorescence. The results of the analysis shed new light on the M. sjostedti pest problem, and the ways to solve it. The lack of efficient antagonists, particularly larval parasitoids known from closely related south-east Asian Megalurothrips spp., and the high damage threshold, indicate that M. sjostedti is a potential target for biological control. However, further studies are needed to investigate the migration of M. sjostedti adults to and from alternative host-plants, in order to reinforce the action of biocontrol with cultural pratices.

This paper reports the results of a survey of parasitism of budmoths on larch in Europe as part of a programme to determine the potential of European parasitoids for the classical biological control of the North American defoliator, Zeiraphera canadensis Mutuura & Freeman. The parasitoid complexes of Zeiraphera diniana (Guenée), Exapate duratella (Heyden), Ptycholomoides aeriferanus (Herrich-Schäffer) and Spilonota laricana (Heinemann) are determined from field-collected samples of host larvae and pupae. The parasitoid complexes are characterized by their component guilds and the mean apparent parasitism of the individual species. Host phenology appears to affect the parasitoid complex of the larch budmoths. Parasitoids that attack the young larval instars are better represented from hosts that overwinter as larvae, while parasitism is dominated by the guilds that attack the late larval instars in hosts that overwinter as eggs. The study identifies the ichneumonids, Phytodietus spp., Chorinaeus cristator (Gravenhorst) and Phaeogenes osculator (Thunberg) for further experimental investigation as potential candidates for the biological control of Z. canadensis.

The impact of Sahelian grasshopper species on short cycle pearl millet (Pennisetum glaucum) was studied at two sites in north-west Mali and techniques of crop loss assessment were investigated. At one of these sites an onfarm study was carried out with the co-operation of seven village farmers. The grasshoppers Diabolocatantops axillaris (Thunberg) and Kraussaria angulifera (Krauss) were the principal destructive species, primarily causing grain loss through their feeding upon millet heads. Leaf damage was found to be sustainable to a high level before influencing yield, although direct grain loss through head damage was shown to be an important determinant of final yield. It was concluded that the best means of determining the yield loss attributable to grasshopper attack for late-season patterns of damage was to estimate head damage immediately prior to harvest. In this estimate, heads were assigned to one of ten classes according to the severity of the damage. Studies at a second site demonstrated the existence of a linear relationship between estimates of head damage and actual grain weight loss. On the basis of these findings it is concluded that the average loss of yield incurred as a result of grasshopper attack in the study village, Mamaribougou Freybe, was 25.7% + 3.0, −2.5. A system is proposed by which such village-level crop loss estimates may be undertaken.

The T:Y(ωp†)30c genetic sexing strain of Ceratitis capitata (Wiedemann), which is designed to generate males with brown (wild-type) pupae and females with white pupae, was contaminated artificially in a series of population experiments, to investigate the pattern of breakdown of the strain. The frequency of brown female pupae increased whether contamination was with males, unmated females or mated females, but the frequency of white male pupae increased only after male or mated female contamination. Contamination with either sex, mated or unmated, caused a rise in the overall frequency of brown pupae. A fitness advantage of the ωp† gene, when not attached to the Y chromosome, seemed to be indicated. The sex ratio became distorted in favour of females after contamination with females, mated or unmated, but not after male contamination which caused a temporary increase in males. The experiments revealed evidence of a low frequency of natural recombination between ωp† and the translocation breakpoint on the Y chromosome, shown by the appearance of a small number of white males that could not be attributed to the experimental procedure. These findings are shown to be relevant in determining whether wild type contamination has taken place in mass rearing of this strain for sterile male release.

Food-baited dishes were used to monitor the impact of ground-sprayed DDT (4% WP @ approx. 200 g/ha) and deltamethrin (0.05% SC @ 2.6 g/ha) for tsetse fly (Glossina spp.) control, on epigeal ants in mopane woodland in north-western Zimbabwe. Twenty species of ants were recorded at dishes baited with various types of food. Pheidole spp. and Monomorium opacum Forel (Formicidae: Myrmicinae) were predominant, both in the number of baits attended and in the number of workers present at individual baits. Season apparently had a greater effect on the composition and activity of the ground foraging ant assemblage than did either pesticide. There was a decline in the rate at which baits were found and in species richness of the ant assemblage as the dry season progressed and several species also showed changes in their ability to locate baits. There was no major pesticide induced disruption of the species richness or food finding ability of the diurnal ground foraging ant assemblage sampled. However, foraging success by Platythyrea cribrinodis (Gerstaecker) (Formicidae: Ponerinae) was reduced immediately after spraying with deltamethrin. Subtle effects of either insecticide on rarer species cannot be discounted from the results of this study.

The effects of nitrogen and silica application to maize plants on various aspects of the bionomics of the maize stem borer, Sesamia calamistis Hampson, were studied in the laboratory. Increasing nitrogen doses significantly increased larval survival (from 18.7% (control) to 37.3% at 2.25 g N/plant), larval weight (from 49.0 mg (control) to 99.5 mg at 2.25 g N/plant), and female fecundity (from 77 eggs per female (control) to 365 eggs per female at 1.69 g N/plant). A significant, positive relationship was found between moth fecundity and female pupal weight. Nitrogen had no effect on larval and pupal developmental time but increased adult longevity. Intrinsic rate of increase, rm, and the net reproductive rate, Ro, were positively related to leaf and stem nitrogen, while generation time, G, was negatively related. Silica had the opposite effect from nitrogen on larval survival. Increasing silica supply reduced ultimate larval survival from 26.0% (control) to 4.0% at 0.56 g Si/plant. Immature developmental time, larval and pupal weight, pupal survival, female fecundity, egg viability and adult longevity of S. calamistis were not affected by silica application.

An account is given of the development of aqueous gum-chloral mounting media used for mounting small arthropods on microscope slides. Berlese’s fluid is shown to have never been used by Berlese and other formulae are shown to have been randomly attributed to various authors, often incorrectly. Erroneous formulae and modifications to formulae have been followed by subsequent workers without reference to their origin. Details of the five formulae currently recommended in the literature are given and serious problems are shown to have been encountered with all of them by many workers. In many collections throughout the world gum-chloral slides are steadily deteriorating and specimens becoming irretrievably lost. Those workers advocating the use of gum-chloral aqueous media continue to propose alternative techniques in an attempt to overcome the problems they admit still occur. It is recommended that these media be used only for temporary mounts and never for specimens of taxonomic significance.

A method was devised for spatial sampling within the soil in potato crops where the soil insects African black beetle, Heteronychus arator (Fabricius) (Coleoptera: Scarabaeidae) and whitefringed weevil, Graphognathus leucoloma (Boheman) (Coleoptera: Curculionidae) are pests. The number and position of potato plant parts, the soil insects, and their attacks on potatoes were defined within the hilled-up soil profile of sections of the planting rows. Neither potato tubers, nor insects nor insect attacks were uniformly distributed. Tubers occurred mainly in the middle of the hill, while the insects contrasted in their distribution in the soil. Most of the damaging H. arator adults and larvae occurred in upper middle areas; the damaging G. leucoloma larvae occurred deeper and were more damaging on the under-side of tubers. The method provides a means of simultaneously sampling soil insect populations in a crop and obtaining a direct assessment of their damage and the effectiveness of control measures.

A new field-based method for sorting insects collected in suction samplers is proposed. Test sampling on grassland showed that for most taxonomic groups it was over 80% efficient. The efficiency of sorting for certain insect groups was influenced by the amount of debris in the sample. However, it was possible to determine a relationship between these two variables which enabled a correction factor to be calculated and used to compensate for this. It is suggested that popular use of this technique would radically increase the amount of research performed in the field.

Threshold temperatures, growth rates and physiological development times were calculated for the egg, larval and pupal stages of the groundnut leaf-miner, Aproaerema modicella (Deventer), a key pest of groundnut (Arachis hypogaea) and soyabean (Glycine max) in India. The life cycle of Aproaerema modicella required 660 degree-days (DD) above threshold temperatures (12.4°C for eggs, 11.3° for larvae and 14.7° for pupae). A function fitted to the adult longevity and fecundity data describes the effect of temperature and female age on fecundity. Egg production was greatest at 30°C and declined at lower and higher temperatures. Head capsule width measurements indicated that five larval instars are typical in India. The results obtained are discussed in the context of earlier work.

Field studies in Zimbabwe showed how targets 1 m tall and 25–400 cm wide, composed of vertical panels of visual material and net and baited with odours of acetone, 1-octen-3-ol, and phenols, could be refined for controlling Glossina morsitans morsitans Westwood and G. pallidipes Austen. Attraction from a distance was not affected by swivelling movement but improved several-fold as the width of visual panels increased from 25 to 200 cm. Blue or blue/black targets were about as effective as black for G. m. morsitans but slightly more effective for G. pallidipes. Fine and coarse net was unattractive at long range but the flies avoided coarse net when close. The percentage of tsetse alighting on visual panels before flying round increased up to several times when the panels were widened, and when non-shiny black replaced blue or shiny black, but the percentage of flies alighting before departure was not much affected expect by panel width, being near 0% at 25 cm and 85% at 2 m. Most tsetse alighted first near the centre of targets, especially where black contrasted with nearby blue. Studies of various fabrics and paints and of restricted deposition of insecticide suggested several disposable targets that could have the costs of materials and insecticide, with efficacy preserved for G. m. morsitans and increased by up to 50% for G. pallidipes.

Field and laboratory trials were undertaken to determine effects and efficacy of neem extracts on the Sahelian grasshopper, Kraussaria angulifera (Krauss). Formulations of aqueous extracts were sprayed in field experiments. Concentrations of 0.5% and 1% were effective in reducing grasshopper feeding. Neem kernel powder applied systematically, partially protected millet against K. angulifera. The residual activity of the botanical insecticide was at least seven days. Analysis of plant selection behaviour showed that neem is primarily an antifeedant rather than a repellent for this insect. On first to third instar nymphs, growth regulating activity, expressed as difficulty in moulting or an incompleteness of sclerotization, may occur.

Cuticular hydrocarbons were analysed by gas chromatography (GC) in 564 specimens of the Anopheles gambiae Giles complex from several West African sites, to determine whether individual A. gambiae sensu stricto and A. arabiensis Patton can be reliably identified, and to investigate the extent to which distinct chromosomal forms of A. gambiae sensu stricto, which are ecologically restricted as well as in some cases sexually isolated, can be distinguished by their cuticular hydrocarbons. Sympatric A. arabiensis and A. gambiae sensu stricto at Banambani, Mali could be distinguished with 90% correct identifications using the concentrations of four hydrocarbons in a linear discriminant function, but at a second site in Moribabougou, Mali, A. arabiensis was indistinguishable from a small sample of Bamako form of A. gambiae sensu stricto. Sympatric chromosomal forms of A. gambiae sensu stricto could be separated and clearest differences were found between the Mopti and Bamako forms. Direct gene flow between these forms has been found to be completely lacking despite partial intergradation of each form with the savanna form. Ethological isolating mechanisms between these forms have not however been demonstrated. Estimates of the rates of misclassification between the savanna form and the Mopti and Bamko forms reflect the degree of integradation observed amongst these forms by analysis of karyotype frequency in the wild. Discrimination was poor when an allopatric sample of the Mopti form was compared with other samples. An overall test shows that the proportion of correct classifications in discriminant analysis tends to be higher between sympatric than between allopatric populations; however, more extensive sampling would be needed for a rigorous test. The involvement of cuticular hydrocarbons in specific mate recognition systems is discussed.