MOST of the studies on the genetic structure of wild populations of Drosophila have been on animals trapped by means of a yeasted banana-agar or similar fruit traps1,2,3. The flies may thus have been drawn from a large number of small discrete breeding units, and hence general inferences are difficult. I have found a very marked deficiency in the number of males trapped in D. subobscura Collin and to a lesser extent in D. obscura, although the sex-ratio in cultures approximates unity, thus demonstrating the inadequacy of collections by means of baited traps.