AND

Since 1937, studies on the epidemiology of yellow fever have been carried on almost continuously by the staff of the Yellow Fever Research Institute in Bwamba county, western Uganda. The study-area consists essentially of a densely populated agricultural zone and a large belt of uninhabited rain-forest, which is continuous with the main Ituri Forest of the Congo Basin. Investigations were at first confined mainly to the inhabited parts of the county. It was soon established that immunity to yellow fever was present among the African population, that the incidence was greatest in areas adjoining the main forest, and that immunity among children was confined to those localities (Hughes et al., 1941). Subse-quent work (Mahaffy et al., 1942) resuhed in the isolation of yellow fever virus from a sick African and from mosquitoes belonging to a ’ semi-domestic ’ species (A&&s (StegoqGa) simpsoni Theobald), the favourite habitat of which is banana-plantations.-4s a consequence of these discoveries, the entire population of the county was vaccinated against yellows- fever, but 11 months after this mass vaccination virus was again isolated from A. simpsoni (Smithburn and Haddow, 1946). As it seemed almost certain that the virus in this latter case must have been derived from wild animals, attention wa