Strong distance-dependent effects for a spatially aggregated tropical species

A fundamental aspect of the Janzen-Connell Hypothesis (JCH) is that distance- and density-dependent mortality reduce the local dominance of species and cause regular rather than random or aggregated spatial patterns. Despite this explicit linkage between process and pattern, very few studies have explored how JCH processes translate into the spatial distributions of adult populations. In field experiments, we assessed germination, mortality and growth of conspecific and heterospecific seedlings beneath and away from Esenbeckia leiocarpa, a highly aggregated tropical tree species. We also investigated the effects of vertebrates using exclosures in the field, and the effects of pathogens in a soil sterilization shadehouse experiment. Germination of conspecifics underneath Esenbeckia was reduced by 64 % and mortality increased 28-123 % when compared to seedlings growing under other tree species; 99-100 % of Esenbeckia seedlings died under conspecifics. Heterospecifics were much less affected by Esenbeckia canopies. However, we found no evidence that either vertebrate herbivores or soil pathogens affected seed germination and seedling performance. Although many tropical tree species are aggregated, our results are the first to demonstrate strong negative distance-dependence for an aggregated species and one of the few to explore germination in the context of the JCH and show differences between conspecifics and heterospecifics, thus suggesting a broader role for Janzen-Connell processes as determinants of the distribution and abundance of tree species in tropical forests.