A successful technique for feeding colonies of Glossina morsitans morsitans Westw. and G. austeni Newst. in the absence of living hosts is described. Insects are fed through membranes made of silicone rubber or of agar and Parafilm, overlying blood pools poured onto grooved glass plates. The diet employed is fresh pig blood, collected at slaughter, and aseptic procedures are adopted at every stage of diet preparation and presentation. The reproductive performance of these in vitro-fed colonies in terms of adult survival, fecundity, and offspring size, is the same as that of colonies fed on living hosts, when compared over a long period of time. The fact that the technique is successful when used with a diet of pig blood, but is not successful when used with cow blood prepared in the same way, suggests that the technique per se is adequate to elicit a normal feeding response from these tsetse species. The nature of the nutritional superiority of pig blood is not understood.

Glossina morsitans morsitans Westw. and G. pallidipes Aust. flying to and from a stationary ox were captured by a wall of electrocuting netting at distances ranging from 45 m upwind of the ox to 120 m downwind. The pattern of catches at each distance suggested that most flies not recently fed were attracted to the ox from downwind, from distances up to about 90 m. Most flies leaving the ox flew downwind. This was most marked for flies that had fed on the ox, some of which travelled 120 m within about 30 min of engorging.

Larvae of Pimpla instigator (F.) were found to pass through five larval instars during development. Initial studies of external morphology identified only three morphological groups: the primary larva (L1) the intermediate larvae (L2−4) and the final instar larvae (L5). Biometrical study of the maximum head width, width of buccal mask, inter-metatentorial distance and diameter of spiracles permitted morphological separation of all five larval instars. Under a 16 h photophase (at 25±1°C; 70±10% r.h.) and an 8 h scotophase (at 15±l°C; 80±10% r.h.) the 2nd, 3rd and 4th larval instars were found to last about 24 h each, and the 5th instar 9±1 days. Progressive development of the genital buds during the 5th instar and prepupal stages facilitated sexing of the larvae.

A latex agglutination test to identify the sources of blood-meals of Culicoides is described. The method requires the minimum amounts of reagents and apparatus and would be suitable for rapid tests under field conditions. Blood-meals of Culicoides nubeculosus (Mg.) and C. variipennis (Coq.) from laboratory cultures, which had engorged on blood from several species of domestic animals, were differentiated without difficulty. In a series of ‘blind’ tests, the blood-meals of 49 C. variipennis from seven species of domestic animal were identified with 100% accuracy. The differentiation of blood-meals from closely related hosts, such as mouse and guineapig or ox and sheep, is entirely dependent upon the specificity of the antisera conjugated to the latex. Blood-meals could be identified reliably for about 48 h after ingestion at 20–25°C, but at higher temperatures the reaction became less reliable after 24 h. Detergents commonly used in collecting midges had no effect on the latex agglutination test, but 0–01% mertbiolate, often used as a preservative, caused auto-agglutination and cannot be used when this test is to be applied.

For nine weeks from 14 April to 13 June 1975 an average of over 500 double translocation heterozygote male Aedes aegypti (L.) were released daily into a village where the estimated adult mosquito population was 1200. The two purposes of this release were to suppress the A. aegypti population by sterility induced by chromosomal translocations and to assess the effect of rainy season conditions on the population dynamics of the indigenous and released mosquitoes. As a result of the releases, hatchabirity of eggs from females collected during landing-biting catches dropped from over 93% to 30–40%. For eight weeks following the termination of the releases, hatchability remained below 70% while that of the reference village averaged between 80 and 99% hatch. A relative population decline observed in the release village is not attributed solely to the effect of the releases, partly because it occurred before sufficient time had elapsed to allow induced sterility to build up in the field population.

Eggs, larvae and pupae of Spodoptera littoralis (Boisd.) were exposed to five non-freezing constant temperatures (13°, 10°, 7°, 4° and 1°C). The period of exposure required to kill eggs and larvae, and to affect pupae seriously, was determined at each temperature. Distinction was made between dead pupae and pupae which were not killed but developed into deformed adults incapable of mating or laying fertile eggs. Probit line analysis was applied to the mortality results and response lines were plotted for all developmental stages at each constant temperature. Observed 100% mortality and computed LD 99 estimates showed that at all temperatures, except 1°C, resistance to cold was minimal at the egg stage, increased in maturing larvae and was at a maximum in the pupal stage, so that the order of the LD 99 values was: pupae (dead)>larvae>eggs. At 1°C the order was: pupae (dead)>30-day-old larvae>eggs>15-day-old and 1-day-old larvae. The LD 99 (days) value decreased with decreasing temperature for all larval stages and for pupae; and for eggs it was greater at 1°C than at 4° and 7°C.

The larvae and pupae of Elachiptereicus abessynicus Becker and of Pachylophus lugens Lw. are described and illustrated and those of the latter species are compared with those of Pachylophus beckeri Curr. A previous misidentification of P. beckeri is recorded and additional records of host plants and distribution are given for all three species.

A new electric trap for tsetse flies (Glossina spp.), based on a circuit producing the necessary high voltages from a small 2-V battery, is described. Power consumption is reduced to less than 1% of its level in other circuits designed for the same purpose so that in theory up to 300 h operation per charge is available from a single 2-V dry cell with a 7-Ah capacity (weight approximately 460 g). In practice this time is more likely to be limited by the shelf-life of the cell in tropical climates. The new circuit and a modified grid were tested against G. morsitans submorsitans Newst. during field trials in Ethiopia and the results are discussed. A high percentage of females was obtained only when a stunning capacitor was introduced across the output. The capacitor stores the charge from the circuit and gives a bigger impulse as flies land across the wires. It seems that, in the absence of the capacitor, females that visit the catching device escape from the electrified grid. When they are caught they are stunned for about the same length of time as male flies.

A highly significant negative linear correlation (r=– 0·911, P=0·001) was observed between Xyleborus fornicatus Eichh. population levels in tea plants and the crop size in the preceding month. A highly significant negative linear correlation (r=– 0·858, P=0·001) was also observed between Xyleborus population levels and the percentage of healed galleries of the respective samples. Cambial activity, which has a direct relationship with crop size, determines the rate of healing of brood galleries, and this seems to be the critical factor that regulates population size within a pruning cycle. Since the population has an inverse relationship to crop size and as the latter could be predicted in advance using weather parameters, it should be possible to predict population levels of Xyleborus in advance.

A general survey of coconut plantations in Guyana in 1976 indicated that in the total area of about 18 700 ha, 1000 of the 3000 ha of coconut palms on the east coast of Demerara were infested by Azteca sp. Detailed crop loss assessments on four plantations in 1970–72 indicate that the overall loss of crop to this pest in Guyana in 1976 was about 3·7 million coconuts, valued at G$372 000. Damage is partly due to mealybug, Nipaecoccus nipae (Mask.), and scale insect, Aspidiotus destructor Sign., colonies maintained by the ants and may also be partly attributed to interference with pollinating insects. Chemical control with aldrin and dieldrin sprays applied to tree trunks to create insecticide barriers and applied to whole trees with a power-sprayer and extension lance did not give adequate long-term control. Baits based on minced bovine intestine with 2% v/w ethyl alcohol were therefore assessed in field trials. Aldrin at 0·3–2·4% a.i., nonachlor at 0·3–2·0% and mirex at 0·05–0·6% were tested. Bait containing 0·1% mirex gave best results with complete kill of ants' nests 60 days after application at about 5 kg bait/ha. Mealybug and scale infestations were also eliminated by this treatment with subsequent reductions of leaf-yellowing and increases of leaf number 100 days after treatment. Up to September 1976, 0·15% mirex bait had been used to control Azteca sp. on 200 ha of coconut palms in Guyana.