This study was designed to introduce a new method of estimating group size and composition of black-and-white snub-nosed monkeys ( Rhinopithecus bieti
) on the basis of faecal amount at sleeping sites at Mt. Baima Nature Reserve. The monkeys spend nights in the form of one-male, multi-female units (OMUs) and all-male units (AMU), and their faecal pellets can be classified into three categories: adult males (the largest), adult females (moderate) and immatures (the smallest) based on their size. Total pellets were counted under sleeping trees used for two nights at Nanren village (99°04'E, 28°34'N, northwest of Yunnan Province, China) in each of four seasons in 2000-2001. Moreover, data on group composition were collected when the monkeys were passing through an open gully in November 2001. Since the number of adults in OMUs shows a positive significant correlation with the amount of pellets amount in each season, the mean number of feces produced per night per individual is the slope of the regression lines. Thus, group size and composition can be relatively reliably and accurately estimated by the faeces under trees compared with the previous methods of estimation, including the use of monkeys' activities and tracks such as broken branches on steep slopes, in deep gorges and under lower visibility. The use of pellets for population estimates displayed 9.4% deviation in regards to population size of adult females. Some causes of the bias were also discussed. The method might be applicable to other monkey groups of this species if their habitats and main foods are similar to those of the study group.