An investigation of seasonal rates of litter production in a mixed forest, Hong Kong Island, was undertaken from October 1982 to January 1984. Total annual litterfall amounted to 1218.96 g m−2 yr−1, comprising 68.6% leaves, 18.9% woody material, and 12.5% fruits, seeds and insect frass. 398.70 g m−2 of litter was recorded immediately after a severe typhoon (9 September 1983). Rates of leaf-fall and organic debris production were correlated with prevailing temperature and rainfall. Leaf-fall peaked in January 1983, March to July 1983, and in September 1983 after the typhoon, while large amounts of insect faeces were collected in May and June. Woody litter production rates were correlated with temperature and rainfall recorded in previous months. Woody litterfall exhibited no obvious periodicity but was slightly increased during March and April. The seasonal distribution of litterfall is discussed in relation to climatic factors and habitat nutrient economy.

The plant vigour hypothesis proposes that herbivores should favour feeding on more vigorously growing plants or plant modules. Similarly, we would expect herbivores to favour plants that regrow vigorously after herbivory. Larger animals, like elephants, may also select plant species relative to their availability and prefer species with larger growth forms in order to meet their intake requirements. The food preferences of the Bornean elephant (Elephas maximus borneensis) in the Lower Kinabatangan Wildlife Sanctuary, Sabah, Malaysia, were investigated along 12 transects in areas where elephants were recently sighted feeding. One hundred and eighty-two plants were eaten and 185 plants were measured for species availability along transects. Species vigour was determined by the monthly regrowth in new shoot length after elephant feeding and the number of new shoots produced on each plant. Measurements were carried out on each plant for 9 mo or until the new shoot was eaten. Plant sizes were determined from their basal diameter. The Bornean elephant did not prefer more vigorous species or species with larger growth forms. New shoots did not grow longer on preferred than avoided species. Additionally, unlike other elephants that live in a forest environment, the Bornean elephant preferred species from the Poaceae (specifically Phragmites karka and Dinochloa scabrida) over other plant types including gingers, palms, lianas and woody trees.

Regeneration processes associated with gap-building-phase sites are described for the tropical montane cloud forest of the El Cielo Biosphere Reserve in Tamaulipas, Mexico. Twelve gap-building-phase sites c. 20 y old were selected within the forest according to their gap size and severity of disturbance (i.e. number of fallen logs). All flowering plants were recorded within the sites and specimens were identified to species level. Species were related to genera geographical affinity and growth form. Abundance, individual height and aerial crown cover were also recorded for all species. The composition for these sites was 121 vascular plant species grouped in 96 genera and 60 families. Results suggest that the advanced regeneration of shrub and tree species of the medium canopy, such as Eugenia capuli, Ternstroemia sylvatica, Turpinia occidentalis, Meliosma oaxacana, Myrsine coriacea and Podocarpus reichei, was favoured by the disturbance regime prevailing in this forest community. All of these woody species belong to tropical genera and are structurally dominant in the gap-building-phase sites. A differential response of the vegetation was obtained in these sites as a function of the severity of disturbance and gap size. This response was revealed by the dominance-diversity relationships. Small gaps (45–95 m2) or gaps created by two treefalls showed a higher diversity and density compared to larger gaps (95–150 m2) or those created by four fallen logs, which had a lower species diversity and significantly lower tree and vine densities. These results imply that gap size and the severity of disturbance have a significant effect on the regeneration of these disturbed sites.

The carabid beetle assemblage found feeding on fig fruit falls at night was studied in a terra firme rain forest near Manaus (Amazonia) from July 1991 to August 1996. A total of 8926 carabid beetles were collected on 64 fruit falls from 10 fig species. The most abundant genus was Notiobia with eight species, N. pseudolimbipennis being the most abundant. The Notiobia species comprised 92% of all specimens collected and all feed on small fig seeds. Their species abundance patterns varied considerably between individual fruit falls and during the course of a single fruit fall. However, the species abundance patterns for all Notiobia at all observed fruit falls for each of the two commonest fig species (Ficus subapiculata, F. guianensis), as well as for fruit falls of the remaining fig species, were very similar. Through feeding and breeding experiments and observations of reproductive success by dissection of females, only two of the eight Notiobia species were found to be specialized fig seed feeders, being able to reproduce only on fig fruit falls. The remaining six species of this genus use fig fruit falls as alternate hosts or ‘stepping stones’ between fruit falls of their host trees, which are widely separated both in time and space.

Field data on the roosting ecology of a maternity colony of Pteropus vampyrus natunae in a peat swamp forest in Sarawak were collected for 19 mo. The colony moved between 10 roost sites and the diurnal home range (comprising only day-roost sites) was about 1120 ha. Two movement phases were observed within the diurnal home range. The first phase consisted of continuous movement between roost sites each month (August 1998 to January 1999, and August to November 1999), and this coincided with the months in which late pregnancies and pups dependent on mothers were seen. The second phase had almost zero or restricted movement each month (February to July 1999, and December 1999 to March 2000). Throughout the diurnal home range, the colony roosted in 20 families of trees (27 genera and 31 species). The bats tended to roost in certain species and sizes of trees.

Dung beetles (Scarabaeidae subfamily Scarabaeinae) are ubiquitous and play an important role in the removal of animal dung and the dispersal of seeds embedded therein. They exhibit a range of dung-acquisition and burying behaviours, from burying dung directly beneath the dung deposit to rolling dung balls up to several metres. Dung beetles act as important agents for secondary seed dispersal and seed survival: the burial of seeds is said to be of advantage against predators and desiccation (Andresen & Feer 2005, Culot et al. 2009, Feer 1999). In addition, burial of seeds by dung beetles is considered beneficial as seeds are not only deposited within the range of depths that are favourable for seedling establishment but also among organic fertilizer that is said to increase seedling growth rates (Andresen 1999, 2002; Estrada & Coates-Estrada 1991, Shepherd & Chapman 1998).

Isolated in forest patches, mangrove stands often experience different physical conditions (rainfall, tide, salinity) despite similar floristic composition and structure. These abiotic factors are known to influence mangrove phenology but their potential effect on the local invertebrate and bird fauna have not been investigated. This study contrasts two sites of black mangrove (Avicennia germinans) located on the Caribbean and Pacific coasts of Panama. The two mangrove sites are characterized by different rainfall and tide patterns, and consequently ground inundation and salinity. These mangrove sites also differed in their plant phenology as well as invertebrate and bird composition. Although 95% of the bird species sampled occur on both coasts, the two mangrove sites shared only 34%. Moreover, each bird community was composed of a particular feeding guild assemblage consistent with the abundance distribution of their invertebrate-prey. Because floristically similar mangrove stands can support different invertebrate and bird faunas, conservation efforts should target mangrove patches submitted to different physical factors in order to preserve biodiversity.

Conversion of natural habitats to anthropogenic land uses is a primary cause of amphibian declines in species-rich tropical regions. However, agricultural lands are frequently used by a subset of forest-associated species, and the habitat value of a given land use is likely modified by the presence and characteristics of remnant trees. Here we used mark–recapture methods to examine abundances and movement probability of the poison frog, Oophaga pumilio, at individual trees in forest-fragment edges and adjacent pastures in north-eastern Costa Rica. One hundred and forty-seven trees were surveyed at three replicate sites that each included a forest fragment and adjacent pasture. Trees were sampled at distances of ≤30 m into forest and ≤150 m into pastures for Oophaga pumilio, and local environmental characteristics were measured at each tree. We also measured indices of physical condition (size and endurance) of frogs captured in forest edges and in nearby pastures. Analyses of 167 marked individuals showed no difference in per-tree abundances or sex ratios between pasture and forest edges. We found significant interactions between habitat type and leaf-litter cover, tree dbh and number of logs, indicating greater influence of local variables on abundances in pastures. Movement among trees was infrequent and not predicted by sex, size, habitat type or environmental variables. While results of endurance tests did not differ for individuals from the two habitats, frogs captured in pastures were, on average, larger than frogs captured in forest edges. These data indicate that remnant trees are important habitat features for O. pumilio in pastures and corroborate research in other systems that suggests that large relictual trees should be retained to maximize the potential for altered landscapes to provide habitat for native species.

Earth mounds are important determinants of vegetation pattern in savannas, acting as foci for establishment of woody plants. In the Kenya highlands, mounds formerly attributed to termites have recently been found to be Mima mounds produced by rhizomyid mole rats. We investigated Mima mounds on black cotton soils near Kenyatta International Airport and on brown clay soils between Nairobi and Thika. At Kenyatta Airport, mounds had loamier, more granular soils than intermounds, and contained small rocks that mole rats can move. At the site between Nairobi and Thika, mound soils were more friable, higher in pH, and lower in carbon than intermound soils. Data from this and other studies indicate that Mima mound soils are more favourable for plant growth than those of true termite mounds. Mound vegetation exhibited lower coverage of grasses and greater coverage by forbs, shrubs, and bare ground, a pattern indicative of intense grazing and soil disturbance. Acacias, absent from most mounds, were abundant in the zone bordering the mound. This pattern may reflect the favourability of mound soils, combined with the effects of high grazing pressure. Future studies of savanna vegetation should distinguish between termite mounds and Mima mounds.

Several studies have shown that qualitative (binary) ant-plant networks are highly nested in tropical environments, in which specialist species (with fewer interactions) are connected with generalists (with the most interactions) in cohesive subgroups. Interactions occur in both qualitative and quantitative networks, however, how their frequency may structure the nestedness in ecological networks involving these organisms is, we believe, unknown. Based on this perspective, we used nestedness analysis to address the effect of interaction frequency on ant-plant networks (n = 14 networks). Unlike binary networks, quantitative networks are often significantly non-nested. In addition, species with a higher interaction frequency have a higher number of links, indicating that these species are possibly more abundant and/or competitive. Moreover, different biological parameters can change the nature of ant-plant interactions, as a plant can be a good resource for one ant and a ‘bad’ resource for another. Thus, this suggests a new perspective for the study of interaction networks in the tropics, since species with lower interaction frequency are not necessarily subsets of species with higher frequency, and consequently generate the non-nested pattern in quantitative networks.

Rainfall seasonality can strongly influence biotic interactions by affecting host plant quality, and thus potentially regulating herbivore exposure to natural enemies. Plant defences are predicted to increase from dry to wet forests, rendering wet-forest caterpillars more vulnerable to parasitoids due to the slow-growth-high-mortality hypothesis. We collected and reared caterpillars from the understorey and trail edges of a wet forest and a seasonally dry forest to determine whether wet-forest caterpillars suffered a higher prevalence of parasitism and were less abundant than dry-forest caterpillars. In the two forests, caterpillar abundances (on average 8 h−1) and prevalence of parasitism (18%) were very similar regardless of feeding niche for both parasitism (27% versus 29% in shelter builders, and 16% versus 11% in external feeders) and caterpillar abundances (shelter builders: 1.42 versus 2.39, and external feeders: 8.27 versus 5.49 caterpillars h−1) in the dry and wet forests, respectively. A similar comparative analysis conducted in the canopy and understorey of the dry forest revealed a higher prevalence of parasitism in the canopy (43%) despite caterpillar densities similar to those in the understorey. Overall, shelter builders suffered higher parasitism than external feeders (32% versus 14.9%), and were attacked primarily by flies, whereas external feeders were more vulnerable to attack by parasitoid wasps.

Effects of tree species heterogeneity on leaf fall were studied in a primary (4 ha) and in a selectively logged forest plot (2.5 ha) in the Ulu Segama Forest Reserve, Sabah, Malaysia, from April 1988 to December 1989. Leaf fall was collected at 30 sampling points in each plot, and identified to species.

Dipterocarpaceae, Euphorbiaceae, Lauraceae, Fagaceae and Meliaceae are important tree families in both plots with regard to their contribution to total basal area, tree density and annual leaf fall. The total number of tree species was higher in the primary forest plot (267) than in the logged forest plot (218), although the number of climber species was higher in the logged forest (44) than in the primary forest plot (33). The overlap in species composition between the two forest plots was relatively small (49%) compared with that in family composition (88%).

In the primary forest plot, the Dipterocarpaceae contributed 29% of the total basal area and 34% of the annual leaf fall. In the logged forest plot these contributions were much lower, 11% and 15%. The contribution to annual leaf fall made by climbers and pioneer trees was higher in the logged forest plot (34%) than in the primary forest plot (8%).

In the primary forest plot, leaf fall was dominated My large emergent and main canopy trees, mainly dipterocarps, and occurred as regular large peaks. In the logged forest leaf fall was dominated by climbers and many, relatively small trees of pioneer species, such as Macaranga hypoleuca, and was more evenly distributed in time.

We studied diet choice by a generalist herbivore, the red howler monkey (Alouatta seniculus) under conditions of high and normal population density. Densities equivalent to 800–1000 km−2 (roughly 20–40 times normal) occurred in populations trapped on small, predator-free islands in Lago Guri, Venezuela. For three successive years, we studied one such population, a group of six animals living on a 0.6-ha island and compared its feeding ecology to that of two groups living at normal densities on a 190-ha island. The 0.6-ha island supported a total of 351 trees >10 cm dbh of 46 species, whereas >100 species probably occurred within the 16- and 23-ha home ranges of the two large-island howler troops. Small-island howlers were thus predicted to consume fewer resources, in particular less fruit, and to be less selective in diet choice than large-island howlers. As predicted, small-island howlers consumed fewer resources and obtained a smaller fraction of their intake from rare tree species (those contributing <1% of basal area). Small-island howlers consumed less fruit (2% of feeding time vs. 22%) and more foliage (73% vs. 55%) than the large-island groups. Diet breadth of small-island howlers was markedly less than that of their large-island counterparts. Tree species not present on the small island contributed ≥60% of leaf consumption by large-island howlers. Foliage sources preferred by large-island howlers were different in each of 3 years, whereas foliage of the same species of tree consistently ranked first on the small island. Long-term persistence (17 y) of self-perpetuating howler groups on Lago Guri islets at >20 times normal density strongly suggests that food availability does not limit mainland populations.

The effect of the change in vegetation structure from closed rain forest to tall open forest on the small mammal assemblage was studied by live trapping at three sites where the ecotone was very narrow (> 20 m) near the southern end of the Wet Tropics World Heritage Area of Australia. Habitat heterogeneity was significantly higher in the mixed open forest/ecotone area than in the adjacent rain forest. There was a large change in the struture of the small mammal assemblage coincident with the vegetation discontinuity. Although the species richness of small mammals was relatively constant across the gradient, the evenness and diversity of the assemblage declined across the transition from open forest into rain forest and biomass increased, largely due to the high abundance of Rattus fuscipes in the rain forest. The results suggest that the species richness of the small mammal assemblage was not determined by the spatial heterogeneity of the vegetation struture. The species composition of the rain forest is probably related to the historical biogeography of the area whereas the species richness of the wet sclerophyll forest is probably due to a mass-area effect from the adjcant large areas of rain forest and dry sclerophyll forest. However, the evenness, and therefore the diversity of the assemblage, was strongly affected by habitat heterogeneity.

The integrity of ecological interactions, such as pollination and seed dispersal, has been recognized as important for long-term conservation of biodiversity (Asquith et al. 1999, Buchmann & Nabhan 1997, Howe 1984, Pizo 1997). This is illustrated by the fact that the loss of large-bodied animals like birds and mammals after habitat fragmentation can result in population changes or even extinction of many plant species (Silva & Tabarelli 2000). In the absence of primary seed dispersers (vertebrates), the importance of invertebrates such as ants, that can carry or clean seeds, probably increases.

Nutrient subsidies from one ecosystem to another serve a critical link among ecosystems. The transfer of materials across the terrestrial-to-marine boundary is considered to be driven by hydrological connectivity, but animal movement can provide another pathway for nutrient transfers. In two separate studies we assessed the role mammals (bats and kangaroos) play in alleviating nutrient limitation in mangrove forests in Australia. At Lizard Island, we measured tree growth and foliar elemental and isotopic composition of trees growing within and outside a large flying fox roost. In Western Australia, we measured foliar elemental and isotopic composition of trees within two forests frequented by kangaroos that feed in spinifex grasslands and shelter in the shade of the mangroves. We compared those with mangroves from adjacent forests that are not frequented by kangaroos. We show that at both locations, the mangrove forest receives terrestrial nutrient subsidies through animal movement. At Lizard Island dominant mangrove species were significantly enriched in nitrogen within the bat roost, as evidenced by higher foliar N concentrations (by up to 150%), N:P and N:C ratios in trees within the roost compared with trees outside the roost. The isotopic signature of foliar N was significantly enriched in 15N by 1–3‰ within the roost, further suggesting that the source of the N enrichment was the bat roost. Growth rates of mangroves within the roost were nearly six times higher than trees outside the roost. In the arid coast of Western Australia, we show elevated foliar 15N abundance of up to 3‰ in mangroves where kangaroos shelter relative to trees where they do not. Thus, this study presents two examples for mammalian herbivore mediated localized transport of nutrients from terrestrial to marine ecosystems, consequently affecting mangrove tree growth, productivity and forest structure.

Decomposition of leaf litter is an important process that releases energy and nutrients in both terrestrial and aquatic environments (Moore et al. 2004, Wallace et al. 1997); therefore, the physical, chemical and biological processes controlling leaf-litter decomposition rates can affect nutrient cycling and productivity in these systems (Cross et al. 2007, Wood et al. 2009). Several studies have shown that leaf decomposition is faster in aquatic than in terrestrial habitats due to relatively constant temperatures, continuous leaching and the physical breakdown of leaves by flowing water (Hutchens & Wallace 2002, Langhans & Tockner 2006, Langhans et al. 2008). Yet, comparatively few studies have examined these relationships in tropical systems with flooded forests. Flooding is a predominant feature of the upper Amazon Basin, but its occurrence and magnitude is complex and not strictly seasonal (Junk et al. 1989). To identify the dominant energy pathways and understand the nutrient dynamics of upper Amazon rain forests, it is imperative to investigate organic matter processing in the aquatic/terrestrial transition zones of these ecosystems.

Fires can significantly affect the structure, floristic composition and biomass content of tropical forests, which are not adapted to this disturbance. To assess the impact of understorey fires on above-ground biomass, this study was conducted in the northern Brazilian Amazon (Roraima state), where uncontrolled forest fires are recurrent. Fifty plots (0.25 ha each) distributed across five fire disturbance classes were inventoried. Losses in biomass stocks were significant (−57% and −63%) for forests that suffered from recurrent fires and progressively occurred until some point between 3 and 7 y after the last fire, as deduced from previous studies. Twelve years after a fire event, biomass stocks were reconstituted, although differences in the floristic composition were associated with greater fire severity, particularly driven by the dominance of the Cecropia spp. In thrice-burned forests, live biomass reached levels of secondary forests of the same region. For large trees (> 50 cm dbh), the reduction (−54%) in biomass was significant in thrice-burned areas. These findings highlight the threat that fires represent for tropical rain forests and emphasise the need for long-term surveys in order to distinguish between the effects of fire severity, fire frequency and time-since-fire artefact.

Bolomys lasiurus is the only species of rodent commonly found in the savanna of Alter do Chão, central Amazonia. Peak densities during 42 months were less than have been reported from the cerrado biome of central Brazil. However, the population at Alter do Chão appears to be more stable than those in other areas and we never recorded zero densities. Many animals were captured over periods greater than 7 months, indicating that either mortality or dispersal was less than in previous studies. Mean home-range size of females was inversely related to the proportion of females that were gravid or lactating. Female home ranges were mutually exclusive and averaged 35% (SD = 8.2%) of male home ranges. Male home ranges overlapped extensively. Indices of home-range size varied seasonally and among years.