We documented one of the most species-rich assemblages of tropical rain-forest Auchenorrhyncha, comprising 402 phloem- and xylem-feeding species, by sampling adults from forest vegetation. Further, we reared 106 species from larvae sampled on 14 plant species. Both xylem- and phloem-feeding guilds exhibited wide host-plant ranges, as 74% of species fed on more than one plant family. In comparison, using data extracted from the temperate-zone literature, phloem-feeders exhibited lower host specificity in Papua New Guinea than in Germany, because in Papua New Guinea they were dominated by generalist Fulgoroidea while in Germany by specialist Membracoidea. The similarity of Auchenorrhyncha assemblages from different plant species was unrelated to the phylogenetic distance between their hosts. Host specificity, abundance and species composition of Auchenorrhyncha assemblages were unrelated to the optimum of their host plant species on succession gradient from secondary to primary forest. Higher host specificity did not lead to greater species richness in Auchenorrhyncha assemblages feeding on different plant species, but the number of species feeding on a particular plant species was a strong predictor of the Auchenorrhyncha abundance on that plant. These patterns suggest that Auchenorrhyncha assemblages on these plant species are not saturated with species and determined by division of limited resources among competitors, but instead are dependent on the number of colonizers from the regional species pool.

Major tropical rivers have been suggested to be important dispersal barriers that increase the beta diversity of animal communities in lowland rain forests. We tested this hypothesis using assemblages of frogs in the floodplains of the Sepik River, a major river system in Papua New Guinea. We surveyed frogs at five sites within a continuous 150 × 500-km area of lowland rain forest bisected by the Sepik, using standardized visual and auditory survey techniques. We documented 769 frogs from 44 species. The similarity in species composition decreased with logarithm of geographical distance between the sites, which ranged from 82 to 465 km. The similarity decay did not depend on whether or not the compared sites were separated by the Sepik River or whether the species were aquatic or terrestrial breeders. Likewise, a DCA ordination of frog assemblages did not show separation of sites by the river as a significant factor explaining their composition. Our results suggest that even major rivers, such as the Sepik, may not act as dispersal barriers. Rivers may not limit the distribution of frogs and therefore have a limited effect on determining frog species abundance and assemblage structure in rain forests.

In New Caledonia, rain forests with an upper canopy dominated by single species of Nothofagus occur next to mixed-canopy forests, without discernible environmental cause. A potential explanation is that they are different successional stages. To test this hypothesis and predict long-term change in canopy dominance, population size structures of 61 canopy species were analysed in six Nothofagus-dominated forests and three adjacent mixed rain forests. Weibull analysis suggests that these Nothofagus forests are secondary forests, with recruitment insufficient to maintain monodominance, except at a high-altitude site. At low- to mid-altitudes the Nothofagus canopy is predicted to develop into a mixed canopy, unless moderate to severe disturbance occurs within its reproductive lifespan. However, adjacent mixed rain forests are also secondary, with 85% of analysed species showing no evidence of continuous regeneration. Fifteen species from both forest types showed reverse-J curves suggesting continuous regeneration, but only Calophyllum caledonicum did so consistently. Since few canopy species showed evidence of high shade tolerance and persistence, a small number of shade-tolerant species is predicted to dominate both forests in the long term, in the hypothetical absence of disturbance. Hence, temporal factors associated with disturbances play a key role in determining dominance in these forests.

Distyly is a floral polymorphism that presumably evolved to facilitate cross-pollination and to prevent sexual interference. However, pollen transfer is often asymmetric, with one floral morph acting as a pollen donor and the other as a pollen recipient. We evaluated the association between floral morphology, pollinator visitation and effectiveness on patterns of pollen transfer in distylous Palicourea tetragona at La Selva Biological Station in Costa Rica. To assess floral variation we measured corolla, pistil and stamen traits from 66 plants. We quantified pollinator visitation and efficiency on 56 individuals and counted pollen loads on stigmas of flowers observed for 1 h. We determined fruit set 2 mo later and assessed between-morph variation in pollen transfer and female reproductive success. Floral variation was mostly consistent with a typical distylous system; however, there was overlap in the stigma heights of pin and thrum individuals in the study population. Primary pollinators were two species of hawkmoths that visited both morphs at a frequency of 2 visits per flower h−1. The mean number of pollen grains deposited on stigmas was 89 for pin and 153 for thrum individuals. However, loads of illegitimate pollen were higher on stigmas of thrum individuals, while loads of legitimate pollen were higher on stigmas of pin individuals. Consistently, fruit set was higher in pin (31%) than in thrum individuals (22%). High deposition of illegitimate pollen, in addition to the lower female reproductive success in the thrum morph reveal that distyly in P. tetragona does not always prevent sexual interference. We suggest that in long and narrow tubular flowers, like those of P. tetragona, stigma clogging by deposition of self- or same-morph pollen may reduce legitimate fertilization of ovules causing the observed asymmetric fruit set.

We assessed the differences and similarities in diel activities among five phytophagous bat species at two habitats over two seasons within the Sierra del Rosario Biosphere Reserve in Cuba. We characterized temporal patterns of activity and overlap of temporal activity for frugivore and nectarivore bat species (Artibeus jamaicensis, Monophyllus redmani, Phyllonycteris poeyi, Phyllops falcatus and Brachyphylla nana) that occur in tropical evergreen forest sites with distinct altitude and vegetation structure during wet and dry seasons. Capture frequencies using mist-nets of 1180 capture events were the empirical basis for analyses. For each species we compared activity patterns between habitats, between seasons, between males and females, as well as between reproductive and non-reproductive females. We also assessed temporal overlap among each possible pair of species at each habitat and used Monte Carlo simulations to evaluate assemblage-wide temporal niche overlap using a new algorithm, termed Rosario, designed specifically for temporal data. The two habitats had the same species composition and bat diel rhythms tended to be consistent across habitats, seasons and sexes for most bat species. In general bat species pairwise temporal niche overlap was high, and the ensemble-wide temporal overlap was consistently high across habitats and seasons indicating a common constraint for bat activities. Activity peaks of most bat species coincided at 4–5 h after sunset, this being in sharp contrast to other Neotropical bat assemblages at continental sites where activity peaks usually overlap within the first 2 h after sunset. This strong disparity in timing of activity peaks between continental and insular bat assemblages can provide the framework for the generation of hypotheses that explain the potential role of time as a mediator of ecological interactions in bat assemblages.

Vast areas of tropical forests have been deforested by human activities, resulting in landscapes comprising forest fragments in matrices of deforested habitats. Soil seed banks (SSB) are essential sources for the regeneration of tropical forests after disturbance. In a fragmented montane landscape in the Bolivian Andes, we investigated SSB in three different habitat types that were associated with different degrees of disturbance, i.e. in forest interior, at forest edges and in deforested habitats. Sampling of habitats was replicated at six sites ranging in altitude from 1950 to 2450 m asl. We extracted seeds from dried soil samples by sieving, classified seeds into morphospecies and size classes, and characterized SSB in terms of density, species richness and composition. We tested effects of disturbance (i.e. habitat type) and altitude on SSB characteristics. Overall, small seeds (<1 mm) dominated SSB (81% of sampled seeds). Seed density and species richness were lowest in deforested habitats, especially in large seeds and distant from adjacent forests (≥20 m), while small-seeded species were most numerous near forest margins. Species turnover between habitats was high. Altitude altered the composition of SSB, but had no effects on seed density and species richness. We conclude that the potential of SSB for natural regeneration of deforested habitats is low and decreases with increasing distance from forest remnants and that forest edges may be eventually invaded by small-seeded species from deforested habitats.

Patterns of social organization and mating systems have been shown to be functions of ecological factors such as resource allocation and breeding density. In some species, particularly birds, social organization and genetic mating systems differ with molecular studies providing evidence of extra-pair young frequently occurring within broods of socially monogamous species. Here we examine the social and genetic mating system of an ecologically little-known forest raptor endemic to the island of Hispaniola in the Caribbean. From 2005–2009, our field observations of over 60 breeding pairs verified a social mating system of monogamy for the species. During the same time period, we collected blood samples (n = 146 birds, 48 nests) and used microsatellite profiles from 10 loci to estimate genetic relatedness among nestlings in a brood and assign putative fathers. We found no evidence of extra-pair paternity in 41 broods. We had one instance where a social male was not assigned as the putative father, however, the confidence level of this assignment was not significant since the genotypes of the social and assigned males were very similar. Our results support our hypothesis that genetic monogamy would be exhibited by Ridgway's hawk, an island-endemic tropical raptor.

The Janzen–Connell hypothesis suggests that highly specific pathogens decrease seedling survival close to the parent plant; however, the underlying mechanism remains unclear. Here, we tested the host specificity of soil pathogens to germinating seeds of Cyclobalanopsis fleuryi and Cryptocarya chinensis in tropical montane rain forest and cloud forest on Hainan, south China. Rhizospheric soils surrounding eight adult trees per species were collected in each forest type and divided into five soil treatments: parent, fungicide-sterilized, autoclave-sterilized, Fusarium-added and Pythium-added soils. Surface-sterilized seeds were sown in each of the five soil treatments and grown in two forest types. The seed germination percentages were significantly higher in sterilized soils (C. fleuryi, 41.5%; C. chinensis, 29.4%) than in non-sterilized soils (C. fleuryi, 28.3%; C. chinensis, 17.1%) in montane rain forest. The seed germination percentages in rhizospheric soil of conspecific parent trees were significantly lower in montane rain forest (C. fleuryi, 17.3%; C. chinensis, 10.5%) than in cloud forest (C. fleuryi, 37.1%; C. chinensis, 21.1%). Our results also suggest that the level of pathogen activity in each tree species varies depending on the environment. Our results support the hypothesis that host-specific pathogens shape tree species composition by differentially affecting seed germination under different environmental conditions.

Our study investigated the influence of termitaria on vegetation utilization by elephants in Chewore North, Zimbabwe. Ten termite mounds and ten similar sized control plots were surveyed for soil nutrients, tree species diversity and plant biomass removal by elephants. Termite mounds had higher mean (± SE) concentrations of P, Ca, Mg and Na (0.15 ± 0.01, 48.8 ± 9.32, 5.78 ± 1.59 and 0.47 ± 0.12 meq per 100 g respectively) than control plots (0.05 ± 0.01, 3.33 ± 0.56, 1.53 ± 0.22 and 0.19 ± 0.02 meq per 100 g). However, Shannon Wiener index of tree diversity did not vary significantly between termite mounds (1.13 ± 0.280) and their control plots (0.827 ± 0.469). Contrary to most studies investigating patterns of vegetation utilization by large mammalian herbivores on termitaria, biomass removal was five times more on control plots than termite mounds. No difference in biomass removal was noted for Colophospermum mopane which had enough replicates for statistical analysis both on termite mounds and control plots. Our study negates the hypothesis that nutrient-rich soil patches will act as feeding hotspots for large mammalian herbivores. We concluded that vegetation utilization by elephants may be tree species specific, particularly in nutrient-rich environments.

In savanna, termite mounds support more diverse vegetation than off-mound areas, but little is known of the patterns in plant assemblages on mounds. To explain vegetation differentiation between (1) component structures of termite mounds (conical centre vs. pediment), (2) active and inactive mounds (termites present vs. termites absent), and (3) sites on and off mounds (on mounds vs. surrounding savanna), species composition, richness and abundances of woody plants were recorded on 70 mounds and in 13 savanna plots (each 20 × 20 m) in north-eastern Namibia, focusing on soil hardness, mound status (active or not) and mound micro-topography as explanatory factors. Woody plants were absent from 33% of active mounds (54% of active cones) but were absent from only 5% of inactive mounds. Species richness and abundance per mound (mean ± SD) were lower on active mounds with (2.0 ± 1.8 and 4.6 ± 6.6, respectively) and without pediments (0.6 ± 0.6 and 0.9 ± 1.1, respectively) than on inactive mounds (4.4 ± 2.7, 19.4 ± 18.8, respectively). Despite the lower woody plant cover, some characteristic species, such as Salvadora persica, occurred preferentially on active mounds; this species occurred on 42% of active mounds. Mean soil hardness (± SD) was higher on conical parts of active mounds (4300 ± 2620 kPa) than on adjacent pediments (583 ± 328 kPa) and inactive mounds (725 ± 619 kPa). This study suggested that mound status, mound micro-topography, and soil hardness promote variability in the vegetation on mounds.

Plants can develop novel adaptations for nutrient acquisition in nutrient-limited ecosystems. These adaptations include colonization by roots of tree trunks and logs that can act as nutrient reservoirs. Termites may facilitate this root colonization by digging tunnels and accelerating decomposition in logs and tree trunks. We measured the frequency with which above-ground tree root colonization co-occurs with the presence of termites or their tunnels inside living trees above 20 cm dbh (n = 178) and dead tree trunks and logs at least 15 cm in diameter (n = 146) in a Bornean tropical forest. Roots above the soil surface co-occur with termite tunnels 39% more frequently than expected by chance in trunks of living trees and 17% more frequently than expected by chance in logs. By categorizing logs according to hardness through ease of penetration, we found that softer logs at a late stage of decay did not show co-occurrence of termite activity and roots to the same extent as harder logs. This suggests that trees forage where termites have removed physical barriers to colonization. In this fashion, termites may accelerate nutrient cycling in tropical rain forests.