Islands can serve as model systems for understanding how biological invasions affect community structure and ecosystem function. Here we show invasion by the alien crazy ant Anoplolepis gracilipes causes a rapid, catastrophic shift in the rain forest ecosystem of a tropical oceanic island, affecting at least three trophic levels. In invaded areas, crazy ants extirpate the red land crab, the dominant endemic consumer on the forest floor. In doing so, crazy ants indirectly release seedling recruitment, enhance species richness of seedlings, and slow litter breakdown. In the forest canopy, new associations between this invasive ant and honeydew‐secreting scale insects accelerate and diversify impacts. Sustained high densities of foraging ants on canopy trees result in high population densities of host‐generalist scale insects and growth of sooty moulds, leading to canopy dieback and even deaths of canopy trees. The indirect fallout from the displacement of a native ‘keystone’ species by an ant invader, itself abetted by introduced/cryptogenic mutualists, produces synergism in impacts to precipitate invasional ‘meltdown’ in this system.

We examined patterns of seedling recruitment and their underlying mechanisms in a population of Chrysophyllum sp. nov. (Sapotaceae), a shade‐tolerant canopy species in mature tropical rainforest in Queensland, Australia. We considered spatial scales ranging from 1 m2 to 1 ha, and temporal scales ranging from 1 to 32 yr. Over the 32‐yr study period there were six episodes of very high seedling recruitment (masts) at intervals ranging from 4 to 10 yr. Less than 2% of new recruits were found in the nine censuses in other years. We found no significant correlations between the numbers of seedlings per census and either seasonal or annual rainfall, number of dry months per year, or El Niño Southern Oscillation (ENSO) events but found two correlations with mean seasonal air temperatures in the years preceding flowering. There were long‐term changes in abundance in both time and space. In 1965 there were 163 seedling‐sized plants in two dense patches, whereas in 1996, there were 4000 in 15 patches. Once a new patch became established, seedlings recruited there in each succeeding mast episode. Adult trees varied in their production of seedlings. Only 25% of the trees in the sample analyzed participated in all mast years; others had few seedlings for up to 28 yr before beginning to produce many seedlings. Seedlings were shade tolerant. They grew extremely slowly in the shaded understory, mean height only doubling in 27 yr. They were also quite long‐lived; 6% of seedlings recruited in 1969 were still alive 27 yr later. There was little effect of natural enemies on seeds or seedlings. We found weak positive and negative effects of density on seed germination, seedling mortality, and growth. Mortality rates of Chrysophyllum seedlings did not show any trends with time, nor with distance from conspecific adults. These results suggest that abundance of older stages is determined by recruitment as well as subsequent growth and mortality. Mortality rates of seedlings of species other than Chrysophyllum decreased with distance from adult Chrysophyllum trees. Under present conditions, the Chrysophyllum population may be increasing in relation to that of other species, perhaps leading to a reduction in tree diversity in this tropical rainforest.

A variety of ecological processes influence diversity and species composition in natural communities. Most of these processes, whether abiotic or biotic, differentially filter individuals from birth to death, thereby altering species' relative abundances. Nonrandom outcomes could accrue throughout ontogeny, or the processes that generate them could be particularly influential at certain stages. One long-standing paradigm in tropical forest ecology holds that patterns of relative abundance among mature trees are largely set by processes operating at the earliest life cycle stages. Several studies confirm filtering processes at some stages, but the longevity of large trees makes a rigorous comparison across size classes impossible without long-term demographic data. Here, we use one of the world's longest-running, plot-based forest dynamics projects to compare nonrandom outcomes across stage classes. We considered a cohort of 7,977 individuals in 186 species that were alive in 1971 and monitored in 13 mortality censuses over 42 y to 2013. Nonrandom mortality with respect to species identity occurred more often in the smaller rather than the larger size classes. Furthermore, observed nonrandom mortality in the smaller size classes had a diversifying influence; species richness of the survivors was up to 30% greater than expected in the two smallest size classes, but not greater than expected in the larger size classes. These results highlight the importance of early life cycle stages in tropical forest community dynamics. More generally, they add to an accumulating body of evidence for the importance of early-stage nonrandom outcomes to community structure in marine and terrestrial environments. diversity | early life-cycle stages | nonrandom | tropical forest P rocesses that operate nonrandomly with respect to species identity contribute to the structure of natural communities (1-3). Evidence from diverse rain forests includes demographic transitions from seeds to seedlings (4, 5), at the seedling (6, 7) and sapling stages (8) and among large trees (9-12). Although the relative contributions of nonrandom processes at each life cycle stage to determining patterns of abundance and diversity in the mature canopy are unknown, one long-standing paradigm is that community assembly is mediated primarily by events occurring from seed dispersal through seedling germination and small-sapling establishment (13-17). However, despite suggestive patterns (6,7,18,19), evidence is lacking for the comparative strength of early-stage dynamics in determining canopy abundance and diversity.Numerous studies demonstrate significant interspecific variation in the susceptibility of tropical tree seedlings to postgermination hazards, including natural enemies (20, 21), adverse climatic or edaphic conditions (22), physical damage (23), and the crowding or shared-enemies effects of con-and heterospecific neighbors (24,25). In other words, the per capita probability of seedling mortality is nonrandom because the probability of death is not t...

Summary 1The seedlings of larger seeded species generally perform better than those of smaller seeded species under a variety of hazards. The reserve effect proposes that larger seeded species retain a greater proportion of their initial seed resources and their seedlings are therefore better provisioned to cope with post-deployment resource deficits. 2 This hypothesis was tested with a suite of 32 Australian rain forest species with storage cotyledons, and seed reserve mass ranging from 36 mg to 25 g. Seedlings germinated in dim light were harvested as their first set of leaves became fully expanded, dissected into shoot, root and left-over cotyledons, dried and weighed. 3 In cross-species, allometric analyses, the mean mass of the shoot-plus-root scaled less than proportionately (slope c . 0.8), and mean cotyledon mass more than proportionately (slope c . 1.1) with either initial seed mass or seedling mass. These slopes were significantly different from each other. Both conditions for a reserve effect were therefore fulfilled. Cotyledons ranged from 45% of total seedling mass in the smallest species, to 92% in the largest. 4 Solid evidence for a reserve effect was detected within two of four families tested (Lauraceae and Myrtaceae), but only in 7 of the 22 species for which there were sufficient data. 5 Even if the reserve effect has present day utility for enhanced seedling performance in larger seeded species, it may have evolved in response to selection for a greater relative retention of initial seed mass during seedling deployment in larger seeded species, combined with selection for greater deployment of seed reserves to the initial shoot and root of smaller seeded species.