The variations in flowering phenology and pollination success of Sinocalycanthus chinensis were investigated in a field survey, which was carried out on the Daming Mountain, Zhejiang Province, China. Three populations of S. chinensis were chosen. Population A (523 m a.s.l.) was in the edges of broad-leaved forests. Population B (520 m a.s.l.) and population C (840 m a.s.l.) were located in the broad-leaved forest. For these three populations, 26, 21 and 25 individuals were randomly marked in 2006 before flowering, respectively. Flowering and fruiting per plant were monitored weekly. Flowering onset, median flowering date, and ending flowering date were recorded. Then phenology index and synchrony index were calculated. We analyzed the difference in flowering phenology between three populations, the correlation matrix between phenological parameters, the effect of flowering time on the pollination success, the correlation between phenological parameter and pollination success, and the difference in pollination success between three populations. The results were as follows: The flowering duration of S. chinensis populations spanned from early May to late June. Temporal patterns of flowering of S. chinensis were very similar at the population level, with a well-defined peak and marked synchrony. Within each population, the phenological parameters showed no correlation with fruit set and seed set. But among populations, onset and median flowering date had significantly negative correlation with fruit set. Ending flowering date had significantly negative correlation with fruit set and the correlation was extremely significant for seed set. Flowering time conferred a marked impact on fruit set (F=15.960, df=3, P＜0.01) at the individual level of S. chinensis, but not on the seed set (F=2.358, df=3, P=0.073). Although population A and population B appeared at similar altitudes, habitat types had different effects on flowering phenology but not on pollination success. Population C from the higher altitude had significantly lower pollination success, as compared with populations A and B from lower altitude. As an endangered species, S. chinensis exhibited a so-called “Mass-flowering” pattern, thereby increasing the pollination success of individuals and the dispersal of pollens within and/or among individuals. This phenomenon does not seem to facilitate to disperse pollens among populations and would pose, to some extent, self-mating and inbreeding depression, thereby leading to low genetic variability, high genetic differentiation. These traits may, to a larger extent, explain why S. chinensis is readily endangered.