Negative density-dependent recruitment of seedlings, that is, seeds of a given species are less likely to become established seedlings if the density of that species is high, has been proposed to be an important mechanism contributing to the extraordinary diversity of tropical tree communities because it can potentially prevent any particular species from usurping all available space, either in close proximity to seed sources or at relatively larger spatial scales. However, density-dependent recruitment does not necessarily enhance community diversity. Furthermore, although density-dependent effects have been found at some life stages in some species, no study has shown that density-dependent recruitment affects community diversity. Here we report the results of observations in a lowland, moist forest in the Republic of Panamá in which the species identities of 386,027 seeds that arrived at 200 seed traps were compared with the species identities of 13,068 seedlings that recruited into adjacent plots over a 4-year period. Across the 200 sites, recruit seedling diversity was significantly higher than seed diversity. Part of this difference was explained by interspecies differences in average recruitment success. Even after accounting for these differences, however, negative density-dependent recruitment contributes significantly to the increase in diversity from seeds to seedling recruits.

Summary 1.We investigated the relationships of seed size, dispersal mode and other species characteristics to interspecific variation in mean primary seed dispersal distances, mean annual seed production per unit basal area, and clumping of seed deposition among 41 tropical tree species on Barro Colorado Island, Panama. 2. A hierarchical Bayesian model incorporating interannual variation in seed production was used to estimate seed dispersal, seed production, and clumping of seed rain for each species from 19 years of data for 188 seed traps on a 50-ha plot in which all adult trees were censused every 5 years. 3. Seed dispersal was modelled as a two-dimensional Student's T distribution with the degrees of freedom parameter fixed at 3, interannual variation in seed production per basal area was modelled as a lognormal, and the clumping of seed rain around its expected value was modelled as a negative binomial distribution. 4. There was wide variation in seed dispersal distances among species sharing the same mode of seed dispersal. Seed dispersal mode did not explain significant variation in seed dispersal distances, but did explain significant variation in clumping: animal-dispersed species showed higher clumping of seed deposition. 5. Among nine wind-dispersed species, the combination of diaspore terminal velocity, tree height and wind speed in the season of peak dispersal explained 40% of variation in dispersal distances. Among 31 animal-dispersed species, 20% of interspecific variation in dispersal distances was explained by seed mass (a negative effect) and tree height (a positive effect). 6. Among all species, seed mass, tree height and dispersal syndrome explained 28% of the variation in mean dispersal distance and seed mass alone explained 45% of the variation in estimated seed production per basal area. 7. Synthesis . There is wide variation in patterns of primary seed rain among tropical tree species. Substantial proportions of interspecific variation in seed production, seed dispersal distances, and clumping of seed deposition are explained by relatively easily measured plant traits, especially dispersal mode, seed mass, and tree height. This provides hope for trait-based generalization and modelling of seed dispersal in tropical forests.

We tested the hypothesis that the El Niñ o Southern Oscillation influences forest-wide fruit production, which, in turn, limits frugivorous and granivorous mammals on Barro Colorado Island (BCI), Panama. Observations of BCI mammals have been compiled for 49 years. Frugivorous mammals experienced famine . The most recent famine is evident from an 11-yr record of natural deaths of mammals and a 2-yr record of population densities. Famine occurred every time a mild dry season followed an El Niñ o event in the 49-yr record. This coincidence is statistically improbable, as demonstrated by a randomization test.A 2-yr cycle of high, then low community-level fruit production has been observed twice for BCI when a mild dry season followed an El Niñ o event. We used 260 litter traps to monitor community-and species-level fruit production from 1 January 1987 through 30 June 1996. Community-level fruit production was greatest during the 1992 El Niñ o event and lowest one year later, after the mild 1993 dry season. We also reinterpret an earlier 2-yr record of fruit production in light of our 9.5-yr record of fruit production. Communitylevel fruit production was elevated during the 1969 El Niñ o event and was very low one year later, after the mild 1970 dry season. We hypothesize that (1) El Niñ o conditions enhance fruit production; (2) high fruit production consumes stored reserves, limiting the next reproductive event; and (3) mild dry seasons reduce fruit production. Each plant species may respond to any combination of the three components of this hypothesis. Communitylevel fruit production is extremely low when species sensitive to components 1 and 2 are entrained with species sensitive to component 3, or when a mild dry season follows one year after an El Niñ o event.El Niñ o events bring dry, sunny conditions to BCI and a large portion of the wet tropics. Drought and sun may both favor fruit production in wet tropical forests. Drought is known to synchronize flowering, and sunny conditions may relieve light limitation. The El Niño Southern Oscillation has a strong 24-mo periodicity. This creates a strong tendency for dry, sunny years to alternate with wet, cloudy years in Central America and elsewhere in the tropics. We present evidence that this alternately enhances and reduces fruit production on BCI.Terborgh hypothesized that predators regulate frugivorous mammals in tropical forests. As a corollary, he further hypothesized that famines occur on BCI because several large predators are absent and frugivores escape predation. We extended censuses to the nearby mainland to evaluate this hypothesis. Poachers, who are active on the mainland, reduce the abundances of game species below levels maintained in the presence of large felids and raptors. There was evidence for famine in poached, mainland populations, and we rejected the hypothesis that reduced predation pressure is a prerequisite for famine.

It has recently been reported that humid tropical forests are changing rapidly in response to global anthropogenic change and that these forests experience greater tree mortality and even fire during droughts associated with El Niño events. The former reports are controversial largely because a single method has been used -repeated censuses of tree plots. The latter reports focus on recent extreme El Niño events. Here, we show that flower and seed production both increase during El Niño events in an old-growth tropical forest in Panama. Flower production, but not seed production, has also increased over the past 18 years. The sustained increase in flower production was greater for 33 liana species than for 48 tree species. These results indicate that moderate El Niño events favour seed production, document long-term increases in flower production for the first time, and corroborate long-term increases in the importance of lianas using independent methods. Changes in levels of solar irradiance might contribute to all three patterns.

Abstract. An economy of scale may lead to selection to increase interannual variation in seed production when the per seed probability of seedling establishment increases with seed production. Variable annual seedfall will, however, reduce this probability when postdispersal seed fate is negatively density dependent on the local density of seeds, and seed dispersal and density dependence act identically across years. Intuitively, more variable annual seedfall causes the representative seed to experience a greater density of conspecific seeds and suffer greater density-dependent effects. This handicap must be overcome for the per seed probability of recruitment to be greater in years with greater seed production.We quantified spatial and annual variation in seedfall and seedling recruitment, evaluated density dependence and economies of scale during the seed-to-seedling transition, and investigated the synergistic consequences of density dependence and variable annual seedfall for seedling recruitment on Barro Colorado Island (BCI), Panama. Weekly censuses of 200 0.5-m 2 seed traps documented seedfall for 15 years and 108 plant species. Annual censuses of 600 1-m 2 seedling plots documented recruitment for nine years and 32 species. The density of seedling recruits tended to increase with the density of seeds; however, the per seed probability of recruitment invariably decreased with seedfall density. Negative density dependence characterized the seed-to-seedling transition. Observed levels of spatial and interannual variation in seedfall density would reduce long-term recruitment by up to 28% if negatively density-dependent survival acted identically across years; however, the strength of negative density dependence varied significantly among years for 12 of 32 species. Negative density dependence occurred in all years for these species but was significantly weaker during the one or two years of greatest seedfall than during the remaining years of lower seedfall. The per seed probability of recruitment increased significantly with annual seedfall for eight of these species. These eight species realized postdispersal economies of scale despite the reduction in long-term recruitment expected from the synergism between variable annual seed production and negatively density-dependent seed fate.