The distribution of sexually efficient males has been analysed in two populations made up of 5 times more females than males. When observation stops as soon as the number of inseminated females equals the number of males, 25 per cent of males in both strains had not mated, 50 per cent of the matings were performed by 50 per cent of the males who mated only once and the other 50 per cent of matings were performed by 25 per cent of the males who mated 2 or 3 times. Productivity of the successive matings falls drastically after the third mating. The relative role of sexual selection and multiple matings as components of fitness are discussed, as well as their implications for the effective population size.