Seasonality of mammalian reproduction is ultimately related to dietary and climatic factors, and the timing of birth is scheduled to optimize offspring survival (Bronson 1985). Reproductive synchrony also represents a strategy that individuals adopt to maximize reproductive success in the course of sociobiological and ecological processes such as sexual selection and predation (Ims 1990). Pinnipeds (seals, walruses and sea lions; Carnivora, Pinnipedia) are distinct from other mammals in having two different phases in their life cycles: reproduction on land (or on ice) and foraging in water. Reproduction in most pinniped species is characterized by tight synchrony of births which ensures that pups are born at the optimal time of year (Atkinson 1997). Seasonality and synchrony of births in pinnipeds then lead to temporal aggregation of postpartum estrus and copulation (Stirling 1975; Boness 1991). Reproductive cycles of female pinnipeds are characterized by embryonic diapause and delayed implantation, a preadaptive feature shared with ursid and mustelid carnivores (Mead 1989; Boyd 1991). Lengths of postimplantation gestation (active gestation) are relatively constant in most pinnipeds (ca. 8 months), and the period of delay in embryo implantation adjusts the length of the whole gestation period and determines the timing of births (Boyd 1991). The northern fur seal (Callorhinus ursinus), a highly polygynous pinniped species, inhabits the North Pacific Ocean, Bering Sea, Okhotsk Sea, and Japan Sea, and breeds on subpolar and temperate islands during summer (Gentry 1998). Their reproduction is characterized by highly synchronized annual cycles (Trites 1992). Synchronization of estrus is linked to monopolizability of receptive females by adult males in terrestrial breeding areas and thus provides a basis for the development of polygyny in this species (Boness 1991). Temte (1985) compared the mean birth dates of two wild populations in Alaska (57°N) and in California (33°N), and postulated that the mean photoperiod of 12.5 h/day in autumn might act as a cue for the implantation and control the synchronization of annual breeding cycles of female northern fur seals. On the contrary, Spotte and Adams (1981) disagreed with the hypothesis of photoperiodic control of reproduction based on the fact that breeding activity of captive northern fur seals captured in Alaska (57°N) and translocated to Connecticut (41°N) was observed in July and August being unaffected by the difference in photoperiod between the two sites. However, their results were inconclusive because they only examined breeding activity by month and their birth records consisted of seven stillbirths that occurred over a period ranging from February to July. Temte (1993) examined birth data of pinnipeds kept in North America, Korea and New Zealand, and found a significant latitudinal gradient of birth timing in California sea lion (Zalophus californianus) and harbor seal (Phoca vitulina). Temte and Temte (1993) further explained that the latitudinal variation of births dates in California sea lions were a response to a common photoperiod (11.5 h/day) before implantation. But Temte (1993) failed to find a significant latitudinal variation in birth timing for northern fur seals due to the small sample size. In Japan, northern fur seals were kept in five aquaria under special permit for domestic law that regulates the capture and possession of this species. In four of the five aquaria, northern fur seals were kept outdoor and exposed to natural photoperiodic conditions (at least during the period from estrus to implantation). These four aquaria were located in two geographical areas with different latitudes: Honshu area (35°N), and Hokkaido area (42–43°N). In this study, we analyze the birth dates of northern fur seals kept in these aquaria to ascertain