The differences between the eggs of A. gambiae and A. melas form an absolutely reliable diagnostic character for distinguishing the two species in the Lagos area.

Identification of eggs laid by wild-caught females has formed the basis of work on the seasonal incidence of gambiae and melas adults in houses in different localities in and around Lagos.

In the area as a whole both melas and gambiae are abundant in most localities, but the greater part of Lagos town on Lagos Island is an almost pure gambiae area.

The sporozoite rate of gambiae—10·0 per cent.—is consistently much higher than that of melas—3·5 per cent.—in all catching stations. Fluctuations in sporozoite rate in relation to population density have been studied in both species.

The composition of the mosquito population in houses, and in outside resting places has been studied in relation to the movements of mosquitos into and out of houses. Melas makes use of outside resting places to about ten times the extent of gambiae.

In African village houses about 90 per cent. of the blood feeding by these two species takes place after midnight, the peak of activity being the hour or two before dawn.

An improved design of experimental hut and window trap is described. It has been used in studying the exodus of mosquitos from houses after feeding, and on experiments on host selection by Anopheles.

Swarms of male “gambiae” in nature have been described. In captivity both melas and gambiae have been persuaded to mate by using artificial light.

By cross fertilisation between melas and gambiae a generation of healthy hybrid adults can be obtained. Of these the females are evidently normal sexually, while most if not all of the males are sterile.

The breeding places of melas and gambiae in and around Lagos have been studied in relation to control. In this lagoon area, where conditions are very different from those of Freetown Estuary, melas breeds under a variety of conditions, and Avicennia mangrove is only of limited value in indicating major breeding grounds.

Field observations on C. cilicrura and C. trichodactyla have shown that in Britain the larvae feed on beans, peas, cabbages, kale, cauliflowers, leeks, onions, cucumbers and lettuce. The importance of C. cilicrura as a pest of beans in spring was well known; its attack on cucumbers was suspected but had not been verified; but the infestation of leeks and brassicas in summer and spring cabbages in autumn had not hitherto been recorded in Britain. An association of larvae of C. cilicrura with rotting vegetation has not so far been established.

Descriptions of adult and larval stages are given. Diagrams are given to illustrate specific differences between larvae of C. cilicrura and those of the onion fly, Hylemyia antiqua.

Field observations have shown that C. cilicrura is on the wing from March to October and there is some evidence of slight activity during the winter months. There are 3–4 generations a year, and peak periods of fly activity occur during mid-April to mid-May, late June to early July, and in fine weather throughout the period August to October.

Larvae have been taken throughout the period May to December. Peak periods of larval activity occur from mid-May to early June on beans, peas, cucumbers and cruciferous crops, during July on leeks and cruciferous crops, and during September to November on spring cabbages.

The close relationship between cultural practice in south Worcestershire and the occurrence of high populations of C. cilicrura is discussed.

A summary of control measures devised by other investigators is given. Modifications of cultural practice that would be likely to reduce attack by C. cilicrura are suggested. Results of preliminary trials with naphthalene and DDT emulsion are given.

Concerning the well-known powerful action of modern contact insecticides the question arises whether these substances exert a specific action on the tissues of insects. It seems quite possible that they possess a general internal toxicity to various groups of animals, but that the insect differs from other animals in that the cuticle is sufficiently permeable to allow the absorption of toxic quantities when these substances are applied to the skin.

In the coastal area of Palestine, Empoasca lybica has eight annual generations. The average number of offspring per female is 16, 81 being the maximum. The adults live in summer up to seven weeks whilst those of the autumn-generation overwinter and live up to five months. The pre-oviposition period, the development of the egg and larva and the total development last from 2 to 23, 5 to 49, 8 to 50 and 15 to 72 days, respectively, according to the season of the year. The threshold of development is between 10 and 12°C. in the various stages of development. The duration of development depends mainly on the temperature. Air humidity has no influence on the duration of development so long as the leaves are sufficiently succulent. Activity of the adults ceases below 14°C. and a temperature above 46°C. results in immediate death. During khamsin-days no oviposition takes place, and many larvae die, especially those of the first stage. The most important local hosts are Solanaceous plants. The characteristic damage is the appearance of yellowish stains on the leaves and leaf-curling; the plants are delayed in development, and shed their flowers and small fruits. Young plants wither when heavily attacked. Damage is most severe in young plants in mid-summer. So far no satisfactory control measures have been found in Palestine.

A method for testing the physical stability of emulsions is described.

Since the persistent toxicity of residual films is so important in the control of mosquitos and flies, this biological activity of stable DDT emulsions was measured by the response and susceptibility of Armigeres obturbans and Culex gelidus over a protracted period of time.

These discriminating tests were carried out before emulsion concentrates were tried out in the field.

Field experiments have been conducted in Uganda to investigate the use of the insecticides DDT and Benzene Hexachioride for the control of tsetse. Since the Tsetse Research Department of Shinyanga has developed a line of investigation into the possibility of using cattle sprayed with DDT as poison bait for tsetses, it was considered that the alternative method of attack, i.e., treatment of vegetation, should be tried. The preliminarywork has been against Glossina palpalis, R.-D. Results of experiments against this species should provide a basis for work against other species. Experiments have been carried out as far as possible on small islands and headlandson Lake Victoria where the flies exist in isolated communities and where results are nottherefore influenced by infiltration of flies from surrounding bush.

A full consideration of the systematic position of Dacus armatus F. (the genotype of Dacus) and of Dacus bivittatus (Big.) is not attempted here, but it is necessary, owing to the economic importance of the latter, to clear up the confusion that has arisen between them. I am indebted to and thank Dr. A. J. Hesse, South African Museum, Capetown; Dr. John Smart, British Museum; Sir Guy A. K. Marshall, K.C.M.G., and Dr. F. van Emden, Imperial Institute of Entomology, and Dr. S. L.Tuxen, Zoological Museum, Copenhagen, for help in unravelling the problem.