The extent to which environmental heterogeneity can account for tree species coexistence in diverse ecosystems, such as tropical rainforests, is hotly debated, although the importance of spatial variability in contributing to species co-existence is well recognized. Termites contribute to the micro-topographical and nutrient spatial heterogeneity of tropical forests. We therefore investigated whether epigeal termite mounds could contribute to the coexistence of plant species within a 50 ha plot at Pasoh Forest Reserve, Malaysia. Overall, stem density was significantly higher on mounds than in their immediate surroundings, but tree species diversity was significantly lower. Canonical correspondence analysis showed that location on or off mounds significantly influenced species distribution when stems were characterized by basal area. Like studies of termite mounds in other ecosystems, our results suggest that epigeal termite mounds provide a specific microhabitat for the enhanced growth and survival of certain species in these species-rich tropical forests. However, the extent to which epigeal termite mounds facilitate species coexistence warrants further investigation.

Questions: All else being equal, populations of dioecious species with a 50:50 sex ratio have only half the effective reproductive population size of bisexual species of equal abundance. Consequently, there is a need to explain how dioecious and bisexual species coexist. Increased mean individual seed mass, fecundity, and population density have all been proposed as attributes of unisexual individuals or populations that may contribute to the persistence or resilience of dioecious species. To date, no studies have compared sympatric dioecious and cosexual species with respect to all three components of fitness. In this study, we sought evidence for these compensatory advantages (higher seed mass, greater seed production per unit basal area, and higher population density) in dioecious species. Location: Five 20–25 ha forest dynamic plots spanning a latitudinal gradient in China, including two temperate, two subtropical, and one tropical forest. Methods: We used a phylogenetically corrected generalized linear modelling approach to assess the phylogenetic dependence and joint evolution of sexual system, seed mass and production, and ecological abundances among 48–333 species and 32,568–136,237 individuals per forest. Results: Across all five forests, we detected no consistent advantage for dioecious relative to sympatric cosexual species with respect to mean individual seed mass, seed production or the density of stems in any size class. Conclusions: Our study suggests that seed traits may provide compensatory mechanisms in some forests, but most often the coexistence of sexual systems cannot be explained by advantages of dioecy related to seed quality and demographic parameters. Future investigations of the factors that promote coexistence may increase our understanding by expanding the search to include attributes such as lifespan and tolerance or resistance to herbivores.