An investigation of the breeding habits of the yellow fever mosquito, Aëdes aegypti, Linn., in West Africa was begun in December 1925. This was undertaken for the purpose of ascertaining if the selection of breeding-places of this mosquito in Africa differs from that in the Western Hemisphere, and, if so, to learn if this difference is sufficient to be of material importance in yellow fever control work.

Since it was believed that breeding in tree-holes constituted a problem worthy of investigation, it was decided to make a survey of the rot-holes and fork-hollows in trees in the vicinity of Lagos, in the Southern Province of Nigeria. This survey was begun on 8th December 1925, shortly after the rainy season had passed and the dry season had set in. However, after inspecting a number of trees, the fact became evident that this was not a propitious time of the year for finding larvae in such places, since practically all of the holes were without water, and their contents, consisting of leaves and débris, were dry.

This early drying of the material in the holes drew my attention to the probability of mosquito eggs remaining in the holes, surviving the drought of the dry season and producing larvae with the beginning of the rains. In furtherance of this question, an examination of the dry holes was then began, and material collected from all those that were suitable. The term “ suitable ” in this sense signifies that the hole was in a living tree—of a formation to contain water, for a short time at least—that the aperture was of sufficient size to allow the débris to be removed and the inside of the hole scraped, and that the contents were dry. The distance of the trees from habitations, their location in open areas or thick bush, the height of the holes from the ground, the exposure of the holes to the sun, or the amount of débris contained caused no discrimination to be made.