A large collection of mosquitos made in five atolls of the Maldive Islands has revealed the presence of 15 species. These are listed and notes of taxonomic interest on seven of them are given.

Field and laboratory studies have shown that the several root-infesting Anthomyiid flies differ greatly in the number of generations composing their annual cycle and in the minimum temperature conditions necessary for full activity.

Onion flies (Delia antiqua (Mg.) ) were mainly univoltine in laboratory cultures. Of 369 first-generation pupae studied in 1953, only 141 (about 40 per cent.) gave rise to flies in July-August and 228 (about 60 per cent.) entered a diapause, in which 196 (about 85 per cent, of those entering diapause) remained until the following spring. In 1954, 99 per cent, of the first-generation pupae from univoltine flies and 97 per cent, of those from bivoltine flies entered the overwintering diapause in July. It then appeared that exposure to low temperatures during the development of the first-generation larvae had induced diapause in the pupae. This was confirmed by the unexpectedly high proportion (49 per cent.) of wild first-generation pupae which entered the overwintering diapause in July in the cool summer of 1954. It was also established that the spring oviposition period for D. antiqua began about the end of May, and it appeared that, in the field, air temperatures of approximately 70°F. were necessary to stimulate egg-laying. First-generation flies emerged in the second and third weeks of July, their numbers depending on the temperature in June. A few second-generation flies, together with late first-generation flies, may emerge in September, but the temperature is then too low for oviposition and there seem to be no third-generation larvae.

Onions are attacked by larvae of Anthomyiid flies from May to August. Two species are concerned, namely the onion fly, Delia antiqua (Mg.), and the bean seed fly, D. cilicrura (Rond.). The onion fly, which in some seasons is mainly univoltine, attacks the crop in June and July. The bean seed fly, which has. 3–4 generations a year, is most injurious to spring-sown onions in May and to summer-sown onions in August. Attack by D. antiqua is the result of the chemotropic response of the species to its host-plants. Attack by D. cilicrura follows the disturbance of the soil during cultural operations in April-May, and occurs again when seed-beds are prepared for August sowings.

The recognition of this complex pattern of attack makes it necessary to reconsider the value of established control measures. The calomel (mercurous chloride) seed-dressing is unlikely to give control of onion fly because its ovicidal efficiency is impaired by exposure to the air during the three months between the sowing of onion seed (end of February) and oviposition (end of May), and by the growth of the plants, which results in the eggs being laid at some distance from the old seed-coat which carries the ovicide. It is also unlikely to have a direct effect on infestation by bean seed fly, because the eggs of this species are not necessarily laid on the seedlings and, in the case of August sowings, the eggs may have hatched before the seed is sown.

A comparison is made of all the members of the genus Thomasiniana Strand at present known from Rosaceous host-plants, namely T. oculiperda (Rübs.), the Red Bud Borer; T. theobaldi Barnes, the Raspberry Cane Midge; T. crataegi Barnes, the Hawthorn Stem Midge and a hitherto undescribed species from wild blackberry. Although morphologically very similar, each can be readily distinguished in trials on host-plant preference. By this means it has also been shown that there are two strains, favouring apple and rose respectively, within the present species T. oculiperda. Other biological and some morphological evidence is given, supporting the above conclusions, and on this basis the midge from blackberry is put forward as a new species for which the names Thomasiniana fruticosi, sp. n. and “Blackberry Cane Midge” are proposed and a formal description given. The biological and economic status of each species or strain is discussed and a list of the known parasites given.

The persistence and toxicity of DDT crystals on thin layers of plant wax were studied. Deposits of crystalline DDT at rates of 2 to 8 microgm. per sq. cm. were applied to glass plates, plain or coated with a 0·5 μ layer of wax obtained from leaves of sisal or cabbage. The plates were stored for various periods at 18°C. and at 43°C. (representing temperature and tropical leaf-temperatures). DDT remaining after storage was estimated biologically by means of its contact action on adult Tribolium castaneum (Hbst.), and chemically by a modification of the Schechter-Haller method. At 18°C, there was no significant change in any of the deposits when estimated chemically. But biological estimation showed that, although the toxicity of the deposits on waxed plates was unchanged, that on glass plates was doubled after storage for eight days. The reason for this change was not discovered. At 43°C., the amount of DDT, estimated both chemically and biologically, diminished rapidly and was negligible after 26 days. The loss of toxicity at 43°C. was shown to be accompanied by loss of weight of the deposits. Loss at 43°C. from plain glass plates could be much reduced by storing them in an atmosphere saturated with DDT vapour. It appeared, therefore, that loss of toxicity at the higher temperature was due to volatilisation of DDT and not to decomposition.

European farms at Abercorn had been suffering from bovine trypanosomiasis carried by Glossina morsitans Westw., and the Tanganyika Department of Tsetse Research were asked in 1935 to advise.

Fly-pickets were established in 1936 to prevent carriage of flies on to the farms, and a belt of country was protected from fire, with the object of thickening the woody vegetation so as to make it unfavourable to G. morsitans. The fire-protected area was extended and continued until 1946; serious escape fires occurred in 1938 and 1941. During these first five years the reduction in catches at the pickets was only about one-third.

An intensive survey of the area was made in 1941, as the result of which concentrations of flies were identified in certain shallow valleys forming the headwaters of small tributary streams. Several of these were cleared of trees, with the result that the apparent density of G. morsitans was at once depressed not only in the valleys themselves but over wide areas of untouched woodland round about.

In 1942 and the following years these clearings were extended to new valley sites as they were discovered, both inside and outside the fire-excluded area, and always with striking effects. The last clearing of all was done in 1952.

A list is given of all the species of mosquitos so far recorded from the Seychelles, with their distribution in so far as this is known. A number of new records are included based on collections made recently by one of us (E.S.B.).

One species and one subspecies new to science are here described for the first time. The first full descriptions of the male terminalia and pupae of certain other species are included.

Some notes are given regarding the taxonomy, bionomics and local distribution of the species concerned and the zoogeography of the groups to which they belong.

Appendices are devoted to corrections to the second edition of “The Mosquitoes of the Ethiopian Eegion”, Vol. I by G. H. E. Hopkins, and to the mosquitos of other islands in the western Indian Ocean.

Syntomosphyrum glossinae Wtstn., a Chalcid parasite of the pupae of tsetse flies in East and West Africa, is recorded parasitising the oöthecae of Periplaneta americana (L.) at the West African Institute for Trypanosomiasis Research, Kaduna, Northern Nigeria.

The distribution of termites in north-eastern Tanganyika is discussed in relation to eight zones delimited by contours and vegetation types. Some termites show very definite limits to their distribution; Hodotermes mossambicus (Hagen) and Amitermes lonnbergianus (Sjöstedt) are examples. The significance of this is discussed.

There is a greater variety of Termitinae near the coast, while the subterranean termites are the more abundant in dry bushland. Some species are able to extend their distribution where gallery forest is found along the larger rivers. Two types of mountain forest are included in the region and extremely few termites are to be found within these.

The distribution of termites observed on descending from the mountains to the plains is discussed and illustrated by a diagram. In addition, two methods of indicating the relative abundance of termites in the soil is described.

In this paper the cockroach, a general name used here to include the commoner species of Periplaneta, Blatta and Blattella, etc., is approached from an economic and medical point of view. Special emphasis is placed on the possibilities of its control by biological methods, and to this end the whole field of its inter-relationships with other organisms in various parts of the world has been carefully surveyed. As a result some 73 parasites, predators and symbionts from such widely diverse groups as the Protozoa, Nematoda, Insecta, Arachnida, the Algae and Bacteria have been reviewed and classified.

The more important parasites, some 25 in number, have been placed along with the chief predators in a group (A) by themselves. Most of the former are members of the Hymenoptera-Parasitica, the more prominent being the Eulophid, Tetrastichus hagenowii (Ratz.), and the Evaniid, Evania appendigaster L., with six of its congeners. Amongst the predators of this group, certain Sphecoid wasps of the genus Ampulex, represented by seven distinct species, take pride of place. In the second big division (B) the individual species are mostly members of the Protozoa, with Endamoeba as a prominent genus, and of the Nematoda, with several genera, most of them in the Oxyuroidea.

The present paper, which is intended to serve as the introductory one of a series on the natural enemies of the cockroach, deals in detail with a very important parasite, namely the Eulophid, Tetrastichus hagenowii.

The theoretical objections to the explanation of the persistence of yellow fever virus in areas subject to a prolonged dry season by cycles involving only mosquitos and wild primates are discussed in relation to the results of field surveys in areas of this type in Kenya. The localities studied were Gede, on the coast about 65 miles north of Mombasa, and Taveta, near Mount Kilimanjaro.

The topography, climate, vegetation and human population of the areas are briefly described.

Gede was visited towards the end of the dry season, Taveta during the main wet season. Thus most of the data relating to conditions during the dry season are derived from Gede, the results from Taveta being used mainly for comparative purposes. The work included 24-hour baited catches at two levels, moving catches, and larval collections in all the main typical environments.

All the species of biting arthropod encountered at each locality are recorded; in each case most were mosquitos.

At Gede the only mosquitos which were abundant as biting adults were species which breed in sea water, Aëdes fryeri (Theo.) and A. pembacnsis (Theo.), and they were practically confined to the close vicinity of their breeding places. The vertical distributions and biting-cycles of these species are defined.

At Gede, mosquito larvae could readily be obtained by filling tree holes artificially with water but no naturally water-filled tree holes were discovered by intensive searches in all the natural environments present in the area. Nor were any water-containing leaf axils discovered.

The duplication of a clypeal spine and the bending of some of the hairs of an antennal tuft in a fourth-instar larva of Culex (Culex) annulirostris Skuse, and the gross malformation of the antennae and siphon of an apparently younger larva from the same pool, are described and discussed.

Percentage mortality by itself is not a good measure of the practical value of an insecticidal treatment. In a relatively simple case, such as the fumigation of stored products, the control value can be expressed in terms of the time required for the reduced population to return to its former level of density, provided the mode of increase of the population is known.

Assuming that increase is at a regular exponential rate, the control values of different mortality percentages can be compared in terms of the calculated recovery times; thus, each of the values in the series 99·6, 93·75, 75, 50 and 29 per cent. mortality is twice as good as the one that follows it. Such relationships can be deduced from the graph provided, which is essentially a curve of constant exponential growth, made linear by the use of the logarithmic scale.

The above comparisons do not depend on the value of the constant rate of increase. If the rate of increase varies, control values can still be compared so long as it is possible to calculate the relative times needed for recovery from different levels of density.

The paper describes the layout and analysis of a test of two types of insect light-trap, the Rothamsted and the Robinson, and two types of illumination. The layout consisted of four variations of the sequence of six nights X A B O D D (X being a night of no trapping and A-D the four traps) in each of four sites in a small woodland. At the end of a complete cycle of 24 nights there had been four nights without trapping; each trap had been an equal number of times in each site; and each trap had followed all four trap-types, and also no-trapping, an equal number of times. The complete cycle was repeated four times at intervals between May and September. All insects captured were sorted into orders and counted and each order was analysed separately. By analysis of variance (after transforming the numbers caught to a logarithmic scale) it was possible to show separately the effect of differences, (1) between nights, (2) between sites, (3) between types of trap, (4) between types of light, and, (5) between cycles, and also the effect (if any) of the previous night's trapping. It was found that ultra-violet light was more efficient for all orders than ordinary electric light, that the Robinson type of trap was more efficient for the larger Lepidoptera but much less efficient for the small Diptera, but that there was no regular effect of the previous night's trapping. The largest source of variation was the difference between nights: difference between locations was small.

The action of certain drugs and insecticides on the frequency of the beat of the isolated cockroach heart, immersed in an aerated physiological solution at 29°C., have been studied. The frequency of the heart-beat is counted by eye and the drugs applied in the physiological solution. Acetylcholine at a concentration of 5 × 10-8 to 10-5 causes a sustained acceleration of the heart-beat which is blocked by atropine (5 × 10-5) and is unaltered by hexamethonium (5 × 10-5), prolonged exposure to nicotine (5 × 1O-5) or ergotamine (3 × 10-6). In this range of concentration ACH has a “muscarine-like” action.

Adrenaline (3 × 10-6 to 5 × 10-8) also causes an acceleration of the heart-beat, which soon passes off. Nor-adrenaline has the same action. Nicotine, hexamethonium or atropine do not modify the action of adrenaline, but ergotamine antagonises it.

Nicotine (5 × 10-5 produces an acceleration of the heart-beat which declines with prolonged exposure until ultimately no effect is produced on reapplieation, but recovery occurs with sufficient washing. Hexamethonium antagonises the effects of low concentration of nicotine. It is concluded that nicotine acts on the cardiac ganglia. Since atropine substantially reduces the action of nicotine, it is suggested that the ganglia excited may be chiefly cholinergic.

ACH in high concentration (10-3) in the presence of atropine (5 × 10-5) has a stimulant effect which is abolished by nicotine. It is inferred that under these conditions, ACH is stimulating ganglia which may be adrenergic.

A new genus, Pseudotheraptus, is described for a Coreid bug, Ps. wayi, sp. n., Causing early nutfall of coconuts in East Africa. Two other species, Ps. devastans (Dist.) and Ps. ugandae, sp. n., are placed in this genus, with descriptions and a key for determination of the three species.

The taxonomic relation of Pseudotheraptus to other genera is discussed, with particular reference to Theraptus, in which the species now described as Ps. wayi was formerly placed provisionally, and to Amblypelta, of which one species (A. cocophaga China) is the cause of identical damage to coconuts in the British Solomon Islands Protectorate.

A key is given for the determination of the two species now remaining in the genus Theraptus.