Records of stainer infestation in cotton have been taken at the Cotton Experiment Station, Barberton, South Africa, since 1931, and in 1933 a uniform system of recording stainer populations in cotton by sampling 10-acre blocks was extended to four farms in the Barberton district, three farms in Swaziland and the Cotton Experiment Station at Magut, Natal.

The records so obtained show that stainers normally appear in plant cotton in late February or early March. D. intermedium, Dist., is present in comparatively small numbers throughout the season; D. nigrofasciatus, Stål, and D.fasciatus, Sign., are very variable in relative abundance. In 1933 the latter species was practically absent from cotton at all points save those in the vicinity of ratoon or standover cotton, but in 1931 and 1932 it appeared in numbers equal to those of D. nigrofasciatus and bred up a very much larger population in the crop.

Where normal migration occurs, all three species of stainers pass through two and a partial third generation in the crop; at the end of the season part of the adults migrate from the crop and the remainder, together with the bulk of the nymphs, may be destroyed by appropriate clean-up measures.

Extensive surveys of the Transvaal Low Veld and rapid tours of portions of Swaziland, Zululand and Portuguese East Africa, indicate that the principal wild food-plants of stainers in these regions comprise the genera Abutilon, Gossypium, Hibiscus and Sida in the Malvaceae, Melhania and Sterculia in the Sterculiaceae and Adansonia in the Bombaceae.

The Malvaceous host plants and Melhania are all herbaceous or shrubby species and upon these D. nigrofasciatus is frequently found. It is possible that colonies of these plants existing in sheltered situations may provide overwintering grounds for this species.

Two species of Sterculia are known, S. rogersii and S. murex. The former is wide-spread and abundant throughout the Low Veld and probably constitutes the main breeding-ground of all species in the early summer. The latter species is rarer, and its status as a food-plant is not yet fully investigated.

The Baobab occurs in large numbers in the Northern Transvaal, but it has not yet been proved to be a winter food-plant. The latest information shows that during the summer it may commonly be infested with D. fasciatus.

In conjunction with stainer population records in the crop, weekly systematic records of damage to the crop have been obtained from samples of bolls which have been examined for puncturing and graded for degree of staining. The number of punctures per boll and the percentage staining are strongly correlated, but it has proved difficult to correlate these with stainer population, except where young bolls are examined.

Internal boll disease, particularly early in the season, may be due to bacterial organisms transmitted by species of Hemiptera other than stainers. Later in the season the infection of the crop is more definitely due to Nematospora spp., of which N. gossypii is commoner than N. coryli.

All species of stainers collected on cotton have been found to transmit Nematospora, though they are not efficient vectors until the fourth instar is reached. Adult stainers collected on wild food-plants (Gossypium herbaceum var. africanum, Hibiscus spp., and Sterculia rogersii) have been shown to be infected with N. gossypii.

The etiology of the disease produced by both species of Nematospora has been followed in inoculation experiments, using pure cultures. The rate of spread of the disease varies with the age of the boll at the time of inoculation, being slower when the boll has passed middle age. In neither species does staining extend beyond the foculus in which infection starts, nor does the fungus occur within the seed except lollowing direct puncturing of the seed.

The fact that the staining is not co-extensive with the region occupied by the fungus, but goes far beyond it, and that a pathological condition indistinguishable from that due to the living organism may be produced by injecting a sterilised suspension of the fungus, suggests that the death of the lint hairs, producing staining, is due to a toxic substance liberated by the developing fungus.