A 15-year perspective on the social organization and life history of sifaka in Kirindy Forest

Abstract

In this chapter, we summarize some fundamental demographic and morphometric data from the first 15 years of a long-term study of Verreaux's sifaka (Propithecus verreauxi) at Kirindy Forest in Western Madagascar. We first describe this research site, its history, and infrastructure, as well as the methods employed to study a local sifaka population. Regular censuses, behavioral observations, and systematic captures of members of up to 11 groups began in 1995 and yielded a data set on demography and life history that can contribute comparative insights about sifaka life history. Our analyses revealed that average group size fluctuated very little around a mean of six individuals across years. Group composition was modified by dispersal (mostly male transfers) or disappearances, births, and deaths. Predation and female transfer were the main mechanisms triggering group extinctions and foundation of new groups (N=5 cases in 149 group years). These exceptional cases of female transfer were most likely motivated by female competition or inbreeding avoidance. One female was a member of at least four different groups. Median age at first birth was 5 years. All females gave birth to single infants, but the proportion of adult females reproducing varied between 25 and 85% across years. The mean interval between 112 births was 15.1 months. Loss of an infant before weaning reduced the subsequent inter-birth interval only by about 1 month. The probability that individual females reproduced successfully decreased as the number of adult females per group increased, implying that subtle forms of female competition limit group size. Mortality is especially high (62%) in the first 2 years of life. Predation by the fossa (Cryptoprocta ferox) is the main cause of death. Maximum female reproductive lifespan is at least 15 years, but longevity is still impossible to estimate. These analyses revealed new insights into female reproductive strategies and their interaction with social organization that were only possible because of the long-term nature of the study, but problems of small sample size still limit the analysis of many vital statistics.