Bryocoropsis cotterelli, sp. n.

Colour.—Blackish-brown, with the head, anterior lateral margins of pronotum, sides of the scutellum, sternum, venter, and legs, ferruginous brown. Hemielytra blackish-brown, with a large irregular area at apex of corium, pale yellowish-hyaline, and a small round spot below the middle of the embolium, another at base of corium, one on disc of corium, and some obscure markings on clavus, ochreous; membrane semi-opaque, blackish-grey with a dirty whitish bilobed spot below apex of cuneus between basal cell and the costal margin. Antennae black, the basal segment tinted with ferruginous brown. Rostrum brown with apex brownish-black. Hind tibiae tinted with black along outer side.

Among several parasites received by the Imperial Bureau of Entomology from different countries, the following three parasites and hyperparasites of the pink boll-worm, Platyedra gossypiella, Saund., proved to be new and are described here as they may be of interest in the biological control of this cotton pest.

The theory of phases of locusts put forward as a working hypothesis by the senior author (Uvarov, 1921) has led, as might have been expected, to further researches in that direction, but the results obtained have been in some cases apparently contradictory. A summary of the work done on these phases has been published recently in a book (Uvarov, 1928), and we do not propose to repeat it here, the purpose of the present joint paper being to discuss very briefly certain aspects of the problem in the light of new observations made by Predtechensky (1928) in Russia and by the junior author in Madagascar.

In the few published accounts of insects belonging to the genus Nysius attacking crops in South Africa, the specific name of the Lygaeid concerned has always been given as binotatus, Germ., known locally as the common stinkvlieg, this species having been recorded as attacking not only fruit, potatoes and sunflower seed-heads, but also as being a household pest.

The writer some years ago described a mosquito under the name of Dixomyia elegans, the type of which remained unique until May, 1927, when a few adults, larvae and pupae were collected. The larvae and pupae were found in hoof-holes made by cattle on the edge of a swamp. The water in these holes contained a considerable amount of decaying vegetable matter and was comparatively clear. It was not examined for algae, except that both Nitella and Chara were noted as being absent. The B. D. H. Universal Indicator gave a colour change of greenish-yellow (7·0–7·5) in all samples of water tested for pH.

The pupal period has been shown to be between three and four times as long during the winter months as it is in the summer months.

Breeding continues the whole year round, but the period of greatest density and activity is in September, i.e., towards the end of the dry season. During a period of heavy rains, long grass and full forest foliage, puparia were fairly abundant, while density counts did not indicate any lack of nourishment. The grass in these mopane belts, however, never grows very long.

All types of forest were used for breeding except the heavy river forest. In the mopane a heavy reduction in density occurred after leaf-fall. At the end of the dry season the dry bed of a small well-shaded stream was used extensively as a breeding place. Heavy thorn thickets (Isinanga) and overgrown dongas were only used as breeding places after most of the leaves had fallen.

The puparia are attacked by at least seven species of insect parasites. Of these Mutilla glossinae and possibly two species of Bombyliids are the most important. These parasites show definite seasonal activity. At the end of the dry season, 84 per cent. of all emergences were parasites.

The percentage of collected puparia which failed to yield either flies or parasites increased from 23 per cent. in March to 60 per cent. in November. Experiments showed that half an hour's exposure to the sun (118° F.), even when covered with about one inch of sand, was sufficient to kill the puparia.

The combined effect of parasitism and the hot summer sun on deposited puparia is no doubt a contributary cause of the great fall in density that occurs throughout mopane forest at the end of the late dry season after leaf-fall is complete. This statement does not necessarily apply to other areas not examined.

This fly belt has continued to spread and game has increased, although the area has been thrown open to free shooting since 1905. Between 1924 and 1925 the spread eastward has been rapid, and new settlement areas have been endangered.

During the course of work on coconut pests in the British Solomon Islands Protectorate, from August to October, 1928, I took the opportunity to make a collection of mosquitos. Ten weeks were spent in the country, but most of this time was occupied in travelling round the group on brief visits to many islands, and on only two occasions were more than three days spent in any one locality. It was not possible, therefore, to make anything like an exhaustive collection of the mosquito fauna.

These species, together with A. brunnipes, Theo., have been referred to the Neocellia group of the subgenus Myzomyia on account of the broad mesonotal scales, and (in the case of the first two species mentioned) of the broadly white-tipped hind tarsi. The three species, however, differ from the other members of the Neocellia group in the possession of a distinct propleural hair, whereas the other African species (A. maculipalpis, Giles, and A. pretoriensis, Theo.) agree with all the Oriental species of the Neocellia group in having no propleural hair. It is therefore probable that the first three species are to be regarded as members of the Myzomyia group in which the ornamentation is unusually well developed. This conclusion, arrived at from an examination of the adults, is confirmed by a study of the larvae ; Christophers informs me that the larvae of A. theileri, which have been examined by himself and Puri, have pleural hairs of the Myzomyia type. It is also suggestive that A. rufipes and A. theileri are both variable in wing-markings and thoracic scaling, and as in the case of A. marshalli, some variations appear to be definitely fixed. Some of these variations are described below as new varieties.

The present paper deals with a further 16 species and varieties of Diaspinae of which 9 are believed to be new to science.

In addition to the acknowledgements already made in Part 1 I desire to express my thanks to the Chief Entomologist of the Department of Agriculture, Pretoria, for his kindness in permitting certain authentic material of South African Coccidae from the collections of his Department to be sent to me, and to Mr. H. K. Munro of the same Department for very kindly looking out the material and forwarding it.