It's All in the Timing: Birth Seasonality and Infant Survival in Eulemur rubriventer

作者：Tecot Stacey R^*

来源：International Journal of Primatology, 2010, 31(5): 715-735.

DOI：10.1007/s10764-010-9423-5

• Eulemur rubriventer
• infant mortality
• relaxed breeding
• reproductive strategies
• seasonal reproduction

Highly seasonal breeding has been considered one of the keys to understanding Malagasy primate socioecology. Strict seasonal breeding may be particularly critical for Malagasy primates because they live in such energetically challenging seasonal environments. Lemurs also live in highly unpredictable environments, and there is growing evidence that reproductive timing may be mediated by additional factors, suggesting that more relaxed breeding seasonality is adaptive in some cases. I tested the adaptive breadth of the birth peak in Eulemur rubriventer, which breed in several different months. I describe reproduction in the species by determining the timing and extent of the birth season (period in which all births occur) and birth peak (period in which the majority of births occur); test whether relaxed reproductive seasonality might increase reproductive success by comparing infant mortality within and outside the birth peak; and model the extent to which fruit availability has an influence on the timing of reproduction. I collected birth data on 5 groups in 2003-2005, which I combined with demographic data that D. Overdorff collected from 5 focal groups and additional censused groups between 1988 and 1996. Thirty births occurred in 8 different months. Births were significantly seasonal, with a unimodal birth peak in late August/September/October, and a mean birth date of October 11. Twenty-three births (76.7%) occurred within 54 d (14.79%) of the year. No births occurred May-July, indicating that conceptions did not occur from late December through late February, and cycling (estimated using gestation length) did not occur until ca. 101 d after the austral summer solstice (December 21). Of 22 infants followed regularly, 18 were born in the birth peak, of which 2 died (11%). All 4 infants born out of season died. Based on fruit availability, I calculated a Theoretical Overlap index (T), which indicated a 3-mo window with optimal food conditions for reproduction. This window corresponded to the timing and breadth of the birth peak in Eulemur rubriventer. These results indicate that a breeding season > 3 mo within a given year is not adaptive in the species, likely due in large part to the availability of fruit during key reproductive stages, particularly before breeding.

• 出版日期2010-10