In a wild-living, artificially provisioned population of Japanese macaques at Takasakiyama in southern Japan, nine sets of twins were recorded from 12,392 known deliveries over a 56-year study period. Recorded twinning frequency was 0.073%. During the first 28-year period, artificial food was given until macaques were satiated and population size increased rapidly. In the second 28-year period, provisioned food was restricted to about half of the former period’s calorific content. Seven sets of twins were born in the first period and two sets in the second. Twining frequency in the two periods was 0.137 and 0.027%, respectively. In comparing studies of other Catarrhine primate samples, we hypothesize that twinning frequency is influenced by living conditions, and we suggest that living conditions should be carefully evaluated in studies of twinning frequency.