The paper stresses the fact that an integrated pest management system is not a new concept leading to a more stable agroecosystem, but was practiced by the farmers for centuries; host-plant resistance is one of its major components. The system was destabilized by the introduction of new sorghum varieties and hybrids not resistant against the major sorghum insect pests like shootfly Atherigona soccata (Rondani); stem borer, Chilo partellus (Swinhoe); midge, Contarinia sorghicola (Coquillet); and headbugs, Calocoris angustatus (Lethiery). In order to balance the agroecosystem again, an integrated insect pest management system for sorghum has to be developed based on traditional pest management practices. These are host-plant resistance, cultural control and biological control. Insecticides should be used only if absolutely necessary. For a better understanding of such a control approach, a summary of the biology and population dynamics of the major insects is given, together with a brief account on the levels and mechanisms of host-plant resistance so far known. Based on this information, the rainfall pattern, plant duration to maturity, time of planting, natural enemies and insecticides, a sorghum based integrated pest management system is proposed for the monsoon and post-monsoon season in which host-plant resistance alone or in combination with the above mentioned control methods could be used.

A method is suggested for estimating sampling biases for male tsetse, based on the assumptions of the classical feeding cycle “model”. Sampling intensity for any particular method is denned as the catch of flies in different segments of the non-feeding phase of their current cycles, expressed as a percentage of the catch of flies that are just entering the feeding phase. Two examples are given, of Glossina palpalis in Nigeria and G. morsitans in Zambia. In both cases, sampling intensity is low just after feeding and increases slowly at first, and then rapidly as flies enter the feeding phase.

Flies in their feeding phase can apparently find and feed on hosts within one day. Feeding rates are inversely related to the fat content of flies entering the feeding phase of their current cycle, whilst rates of fat usage increase with this variable. These results suggest that it is important to distinguish between tsetse flight activity in general, and feeding behaviour in particular. Finally, it is emphasised that the analytical technique proposed depends on our still inadequate understanding of the field behaviour of flies.

Immune haemolymphs from the giant silkmoth, Hyalophora cecropia and, to a lesser extent, from the tsetse, Glossina morsitans morsitans possess antibacterial activities against several species of bacteria known to be pathogenic to man, animals and poultry, as demonstrated by in vitro bacterial growth inhibition assays. Cecropia immune haemolymph possesses a broader antibacterial spectrum and was found to be active against several Gram-positive and Gram-negative pathogenic bacteria. Mice injected with pre-determined lethal doses of Enterobacter cloacae, Klebsiella pneumoniae and Corynebacterium pseudotuberculosis and then treated with single intraperitoneal or subcutaneous injections of Cecropia immune haemolymph had much lower mortality than untreated controls. A possibility of developing a broad-spectrum antibiotic modelled on insect immune factors is discussed.

Occurrences of outbreaks of the African armyworm, Spodoptera exempta, between January and December for 19 years (1961–1979), and records on nightly moth catches from light trap stations located between southern Kenya and east-central Tanzania (1–8°S and 35–41°E) for 16 years (1963–1978), have been mapped, tabulated, and analysed on a standard weekly time-table, in order to compare events occurring between the ‘off-season’ months of July to November and the beginning of the ‘outbreak seasons’ from December or January. Circumstantial evidence from the above, coupled with analysis of climatological statistics for rainfall, temperature and windflow over the area and period, support the hypothesis that populations of S. exempta can survive between southern Kenya and eastcentral Tanzania during the off-season months, and contribute to the initial development of ‘outbreak seasons’ in Tanzania or Kenya. Methods for monitoring and analysing such developments for improving the efficiency of the Armyworm Forecasting Service in East Africa are suggested.

Surveys were conducted in 1981, 1983 and 1984 across the island of Puerto Rico to determine the effects of patch characteristics of the tropical herbaceous legume, Crotalaria pallida Ait. on the herbivore and its parasitoids. The mean number of flower racemes per plant and mature pods per raceme varied from year to year and appeared to be influenced by rainfall patterns, but were independent of patch size or plant density. Low larval parasitism was also observed during dry seasons. Proportion of plant patches colonized by Etiella zinckenella was smallest during dry seasons. Local extinction of C. pallida patches was found commonly, with a 77.4% extinction rate during a 6-month period. Patch extinction, which was caused mainly by human activity, was independent of patch size.

Pods of C. pallida within patches were inspected for lima bean pod borer, E. zinckenella (Lepidoptera: Phycitidae) injury. The number of pods bored per mature raceme, and the per cent of bored pods were positively correlated with the number of host-plants per patch and the (surface) area of the patch. No correlation was found between per cent parasitism and pod borer larval density. However, per cent parasitism was influenced directly by C. pallida plant density and patch surface area. Given the transient nature of C. pallida patches and the changes that occur over a season, these positive responses of parasitoids to plant density and patch surface area are important components for their successful exploitation of larval hosts.

A situation report on the Ixodoidea infesting livestock in India is presented. Of the nearly 120 tick species known from this country, information regarding the hosts and distribution of 46 ixodid species belonging to eight genera and six argasid species belonging to three genera, parasitizing livestock has been provided. The ecology and disease relationship of the economically more important species have been reviewed. Notwithstanding the rich tick fauna, the native cattle seldom carry excessive tick burdens. Control is greatly facilitated by the high degree of natural resistance possessed by indigenous cattle, zebu. Chemical acaricides are the maincomponent of control strategies which are largely aimed at protecting the at-risk exotic and crossbred animals from ticks and the diseases transmitted by them. Exploitation of immunity and possibly some biological control may be integrated with the use of chemicals in future.

In 1996, many swarms of red locust, Nomadacris septemfasciata (Serville), formed in the Buzi-Gorongosa outbreak area in Mozambique. Some of these swarms migrated and crossed international boundaries, on a scale not seen since the last plague of 1929–1944.

The swarms did not breed in countries that they invaded, namely, Zimbabwe, Zambia, Botswana and South Africa. On the other hand, in Malawi, successful breeding by the swarms resulted in hopper bands. Filial generation outbreaks continued into 1997 in the breeding area, but at a reduced scale.

An analysis is made of the possible causes of the upsurges of 1996 and the likely implications to the preventive control strategy that has been the principle for red locust plague prevention.

There is need for a further comprehensive study of the Buzi-Gorongosa plains so as to understand better the factors that contribute to its high potential as an outbreak area.

A large pond, up to 2000 m2 area, adjacent to the sea at Jimbo, near Vanga, on the south coast of Kenya was studied for salinity and the presence of mosquito larvae 1980–1981. Rainfall peaked in November–December 1980 and again in March–May 1981, while salinity of the pond-water fluctuated inversely from dryness in February, more than 100% seawater equivalent in months before main rains, to below 13% seawater after the onset of rain in March.

Five types of mosquito larvae occurred in the pond: Aedes albocephalus mostly at 21–30% salinity (range 13–30%); Culex sitiens mostly at 31–40% salinity (range 13–54%); Culex tritaeniorhynchus and Culex thalassius mostly at 41–50% salinity (range 13–50%) and the Anopheles gambiae complex from 13 to 54% salinity, at least those at higher salinities being Anopheles merus. No larvae were present when salinity surpassed 54% through evaporation, or fell below 13% after rain. The most reliable way to limit or control these pest and vector mosquitoes at Jimbo would be to drain the pond into the sea after rainfall, so that mosquito populations could not increase by breeding sequentially, while salination proceeds.

Juvenile hormone analogues (JHAs), JHA I (6,7′-epoxy-3′,7′-dimethyl-dec-2′-enyl 6-ethyl-3-pyridyl ether), JHA II (3′,7′-dimethyl-undec-2,6′-dinyl 6-ethyl-3-pyridyl ether), MV-678 (2-methoxy-9-[4-isopropyl phenyl]-2, 6-dimethylnonane), and R-20458 (4-ethylphenyl-6, 7′-epoxy-geranyl ether) in acetone solution were applied to insect diets at 10ppm (v/v), and their effects on four species of stored products beetles observed. The larvae of the flour beetles, Tribolium confusum Jacquelin duVal, and Tribolium castaneum (Herbst), reared in media treated with JHAs I and II failed to pupate. At the end of the 6-week observation period, these larvae had not yet pupated. MV-678 and R-20458 produced ecdysial failures, lethal morphological abnormalities and larval/pupal and pupal/adult intermediates. Only a few apparently normal adult flour beetles emerged from these treatments. The lesser borer grain beetle, Rhizopertha dominica (Fabricus), completely failed to reproduce in a medium treated with JHA I. JHA II, MV-678 and R-20458 affected the reproduction of the two grain beetles, R. dominica and Sitophilus oryzae (L.), to varying extent, with MV-678 being the least potent JHA.

In experiments to select a character for identifying larval instars in Sesamia calamistis and Eldana saccharina, body length, body width and head capsule width were measured in populations reared on artificial diet. Seven instars were found in S. calamistis and 5 in E. saccharina. For both species, body length was determined to be the best characteristic to distinguish instars because it was easy to measure and had the smallest error.

The sternal gland in Macrotermes michaelseni is located in the anterior half of the fifth abdominal sternite between the third and fourth abdominal ganglia and overlapped by the fourth sternite. It is found in all castes including newly-swarmed de-alates. The gland is largest in major workers followed by major soldiers, minor soldiers and minor workers and is the only source of trail pheromone. Major workers have the highest potential trail pheromone activity followed by minor workers, male and female de-alates, minor soldiers and major soldiers.

There is a change in trail laying behavior with time and direction of travel, from the culture or from the arena and it can be correlated with the deposition of a light or heavy trail, or no trail at all.

The trail of one major worker leaving the culture is attractive to a significant number of workers for 1 hr 30 min and that of five major workers for 1 hr 45 min. The trail of workers leaving the arena is not as strong, one worker trail is attractive for 1 hr and five worker trails for 1 hr 15 min. The well-established trail is attractive to a significant number of workers for 9 hr. All trails lose attractiveness at a constant rate.

A study of the resting site preferences of Glossina palpalis gambiensis in an area of a gallery forest in the Northern Guinea Savanna of Mali, infested exclusively by G. p. gambiensis was conducted in February, 1980 (hot dry season). Five site-types were compared: fallen logs, boles, undergrowth, tree canopies and branches. Distribution of resting flies was as follows: fallen logs (55.62%), boles (35.24%), undergrowth (8.4%), tree canopies (0.61%) and branches (0.00%). Most of the flies were collected during the periods 06.00–09.00 (37.47%) and 09.00–12.00 hr (46.5%). On boles, 86.06% of flies were found below 1.0 m and no flies were collected above 2.0 m. These results are compared to those obtained during simultaneous studies on the resting site preferences of G. p. gambiensis in an area along the same gallery forest, with mixed G. p. gambiensis and G. morsitans submorsitans infestation. Possible reasons for the differences observed are discussed.

The efficacy of a field vacuum unit for the management of Bemisia tabaci was evaluated on melon (Cucumis melo L.) during two summer seasons. The tractor-mounted unit was designed to dislodge insects by blowing air from lateral vents onto the plants while simultaneously vacuuming from above. Field observations, collection of leaf samples, and hand-vacuum sampling before and after the field vacuuming were used to evaluate the efficacy of removal. Bemisia tabaci population reductions were significant (30–60%) and lasted from week to week when vacuumed weekly. Parasitoid populations were slightly affected, but were not significantly different from populations in the untreated control plot. Field vacuuming was found to be as effective as insecticide applications for the control of B. tabaci in melons.

Phenology of Arboreum varieties: G 27, LD 133 and LD 230 differed from Hirsutum, F 414 under unsprayed conditions in Ludhiana. Protection to Arboreum varieties at seedling stage against grasshopper and to bolls formed between 1 August to 15 September against bollworm damage was found necessary. In Hirsutum variety, F 414 bolls formed between 16 August to 15 September needed maximum protection from bollworm damage.

At vegetative phase, jassid population was positively correlated with the leaf size and negatively with number of gossypol glands and leaf thickness, and that of whitefly was positively correlated with number of leaves per plant. Six plant morphological characters: number of leaves, squares, bolls, vegetative and reproductive branches and shoot pubescence-affected bollworm infestation significantly and the effect was related to and varied with the age of the plant. Anthocyanin pigment was not detected in any plant part in the four cotton varieties. At vegetative and reproductive phases, gossypol, total phenols, condensed tannins, leuco-anthocyanins and flavenols in leaf, shoot and boll per se did not influence insect pest infestations under field conditions. The parasite population was not related to pest infestations throughout the plant growth. Prédation rates, however, on pest populations seemed to be considerable. Sclereids and lignified cells affected penetration of first instar Earias vittella larvae into bolls.

The distribution and existing control measures for three important Chilo spp. (C. partellus, C. orichalcocilliellus and C. sacchariphagus) to cereal crop production in 10 east African countries and their neighbouring Indian Ocean Islands are reviewed. Various cultural, biological, host-plant resistance/tolerance and chemical control methods developed in the past 30 years of cereal research programmes are discussed. Cultural methods are very effective on Chilo spp. control, but are not widely practised. Although several effective natural enemies have been identified for Chilo spp., very few systematic programmes of their utilization have been effected. Several resistant/tolerant maize and sorghum lines have been reported, but they are not yet fully used. Chemical control though effective, has proved rather unsuccessful in traditional agriculture. Suggestions to incorporate these control methods in Integrated Pest Management (IPM) programmes for Chilo spp. in different places in the region are given.

Bruchidius atrolineatus Pic is a Coleopteran Bruchid developing in cowpea (Vigna unguiculata Leguminosae) seeds. This insect causes important damage in tropical countries during storage of cowpea seeds. The influence of some volatile sulfur compounds is studied on adults and larvae of B. atrolineatus. Dimethyl trisulfide and diethyl trisulfide produce the highest mortality to adults, eggs and larvae. The toxicity of the sulfur compounds is higher than that of carbon tetrachloride but is less than that of Dichlorvos at a dosage of 500 g/l. It is suggested that the sulfur compounds produced by the leaves of some tropical plants (Alliaceae, Bignonlaceae, Mimosaceae) could be used for control of Bruchid populations in the stored products.

Seven benzyl-1,3-benzodioxole analogs were tested for possible effects on reproduction in females of the Mediterranean fruit fly. Single topical doses of J2581 (5-ethoxy-6-[4-methoxyphenyl]methyl-1,3-benzodioxole) and J3263 (5-methoxy-6-(1-[4-ethoxyphenyl]ethyl)-1,3-benzodioxole) were found to cause a temporary delay in ovarian growth in female medflies by 9 and 13 days respectively when compared to acetone-treated controls. Continued exposure of the analogs at the target site(s) appears to be necessary for more permanent effects. Nurse cell development appears not to be affected, but oocyte growth and subsequent egg production and egg hatch were significantly reduced. Methoprene treatment appeared to obviate the effect of J2581 in arresting ovarian growth.

Field experiments were conducted to investigate the attractiveness of fish bones to predatory ants in a maize agroecosystem. The fish bones were applied crushed in shallow furrows together with dry maize stalks, or broadcast on the soil surface at 3 application rates: 0, 40 and 80 kg/ha; subsequent reduction in termite attack on maize plants was then assessed. Burying the fish bone powder caused higher ant nesting (10–78 %) than surface broadcasting (12–39 %). Attractiveness of the fish bones and the predatory efficiency of the ants against termites in maize was also highest with a dose of 80 kg/ha (1.0 kg/plot) causing higher nesting of ants (>90 %) and 54 % lower termite damage to maize plants compared to control plots. Dose responses indicated significant relationships between amount of fish bone powder applied and termite activity and also between termite damage and maize yield. The mechanism by which the ants are able to protect maize against termite damage may be twofold: direct kill as a result of their predatory action and termite avoidance of maize plants with ant nests. Further studies are required to assess the potential benefits of an integrated use of this technique with other non-chemical options for termite control in maize, such as intercropping.

The development time and survival of the castor whitefly, Trialeurodes ricini Misra were measured at four constant temperatures. The duration of the egg stage varied from 15 days at 20°C to 4.6 days at 35°C. No eggs hatched at a constant 40°C. The minimum temperature threshold for egg development was calculated to be 17°C. The total development time from egg to adult ranged from 54.4 days at 20°C to 16.8 days at 30°C. Development of the egg and the first-to fifth-instar larva required 63 and 180 degree-days respectively above the threshold value. Immature survival increased from 64.5% at 20°C to 92.8% at 30°C. Trialeurodes ricini is well adapted to high temperatures and may extend its distribution if mean world temperatures increase as a result of global warming.