An experimental test of density- and distant-dependent recruitment of mahogany (Swietenia macrophylla) in southeastern Amazonia

According to the Janzen-Connell model, high mortality of seeds and seedlings in proximity to conspecific adults can help maintain species diversity in tropical forests. Using a natural population of big-leaf mahogany (Swietenia macrophylla King), we tested the model's mechanism by examining seed predation and juvenile recruitment in the forest understory and in treefall gaps in the vicinity of both isolated and clumped adults. We used tethered seeds placed in three types of exclosure plots: (1) complete access to seeds, (2) semi-access (access by small-sized seed predators) and (3) no access (all mammals excluded). Exclosure treatments were applied within the understory (both near and far from adults) and in gaps at eight fruiting adults in the late dry season (2001) and scored ten months later. Significantly more seeds were removed in canopy gaps near clumped adults than at isolated adults; otherwise, none of the treatment factors significantly influenced seed predation. In contrast, understory juvenile recruitment was significantly enhanced by distance from adults and was twice as high at isolated than clumped adults, providing novel support for the Janzen-Connell mechanism. No-access exclosures protected significantly more seeds than semi- and full-access exclosures, implicating small mammals in seed losses. Across the eight trees, juvenile recruitment in the no-access exclosures decreased significantly with conspecific adult densities, implicating non-mammalian density-responsive factor(s) in mortality following germination; likely a known specialist invertebrate herbivore. When all treatments were combined, conspecific adult basal area and total DBH explained 72 and 90% of variation in overall juvenile recruitment, respectively. Collectively, these results indicate that Janzen-Connell effects can operate in S. macrophylla, especially during the seed-to-seedling transition, and will likely reduce recruitment in areas of high conspecific densities. They also suggest that further research into the causes of density-dependence in tropical trees should investigate mortality agents following germination.