Co-evolutionary processes have moulded interactions between hummingbirds (Trochilidae) and ornithophilous flowers. Generally, long corollas exclude short-billed trochilids, whereas short corollas allow access for both long- and short-billed species. At La Mucuy, Sierra Nevada National Park, Mérida, Venezuela, we studied interactions between Aphelandra runcinata (Acanthaceae) and its avian pollinators. Of 13 trochilid species reported from this site, only two, Coeligena torquata and C. eos, have bills long enough to allow legitimate access to the nectar of A. runcinata flowers. However, we observed that hummingbirds with much shorter bills were also able to extract nectar and pollinate the flowers. This is due to a reversible compression mechanism in A. runcinata, caused by the pressure exerted by the bird in flight and the physical opposition of the inflorescence. The compression mechanism observed in A. runcinata flowers is a morphological characteristic that provides mutualistic advantages for both the plant and its avian visitors. The spectrum of potential pollinators is increased by giving access to nectar even to species with short or medium-long bills. Moreover, nectar-robbing, a foraging strategy that does not result in pollen transfer common among short-billed trochilids, seems to be reduced.

Cerberiopsis candelabra is a long-lived monocarpic rain-forest tree endemic to New Caledonia that shows mass flowering across a substantial proportion of a population, and across a substantial number of populations. We investigated the relationship between tree size and flowering (and subsequent dying) across 18 populations from the flowering event of 2003 in order to understand the role of possible size thresholds for flowering in the life history and regeneration ecology of this monocarpic species. There was a strong positive correlation between trunk diameter and the incidence of flowering when population data were combined. However, the relationship between size and flowering was complex in that flowering occurred across a wide range of tree sizes, with almost complete overlap in size between flowering (5–79 cm dbh) and non-flowering trees (5–64 cm dbh), and with large trees in both the flowering and non-flowering state in the same population. In about half the populations studied there was no significant difference in mean trunk diameter of flowering and non-flowering trees. Nonetheless, we suggest that tree size may play a fundamental role in the life history and regeneration ecology of this species. The seedlings appear to be relatively shade-intolerant and dependent on large canopy gaps for recruitment. A significant effect of mass flowering and subsequent death of multiple large trees is the potential to form large canopy gaps and enhance seedling survival, as the gap is temporally linked with seed germination. However, it is unclear why there is such a large size range of flowering trees, i.e. whether this is just a consequence of the proximate cue, or whether the optimal size for flowering does indeed vary among individuals.

Fig pollinators (Agaonidae, Chalcioidea) lay their eggs in fig inflorescences (Ficus, Moraceae). Reproductive success for both partners is thus largely dependent on the dispersal of these tiny wasps. Some are known to cover substantial distances (> 10 km) using wind above the canopy. However, fig ecology is extremely varied, and hence one might also expect a diversity of pollinator dispersal strategies. We studied fig pollinator dispersal in Sarawak (2001 and 2004) and Peninsular Malaysia (2003). The results indicate substantial differences in dispersal ecology between the pollinators of monoecious and dioecious figs. Monoecious-fig pollinators were common, and species composition and rank abundances were similar between years despite short sampling periods. Substantial temporal and spatial variation in their production is thus smoothed out by long-distance dispersal. Some species whose hosts do not occur at our Sarawak site and are rare throughout Borneo were caught, suggesting exceptionally long-distance dispersal in these species. Conversely, few dioecious-fig pollinators were caught and species overlap between years was low. Dispersal range in many dioecious-fig pollinators may be more restricted. At a finer scale, among genera pollinating monoecious figs we found marked differences in flight behaviour (height and time-of-dispersal). We relate these findings to the ecology of their hosts, and discuss the implications for fig–fig-pollinator coevolution.

The diets of the acouchy Myoprocta exilis, agouti Dasyprocta leporina and paca Agouti paca, the three largest terrestrial rodents of French Guiana forests, were investigated from stomachs collected over several years. These species are primarily frugivores. The percentage of fruit in their diet decreased, and seasonal variation increased from the smallest species, the acouchy, to the largest, the paca (contents in dry mass: respectively, 98.5% ± 0.6%, 87.4% ± 3.5%, and 83.9% ± 7.6%). They differed from each other in the proportions of pulp and secondary foods (leaves, fibre, insects) consumed annually as well as seasonally. Dietary diversity in the acouchy was half that of the two other species. The seasonal variation in consumption of different fruit parts and of secondary foods by these species corresponded well with the fruit production in the forest, and was linked to their reproductive events. Overall, the diets of these three large rodents appeared intermediate between the small rodent species and the artiodactyls (peccaries and brocket deer) inhabiting the same areas, which also concords with their body mass and their place within the guild of terrestrial mammals.

Seedling growth under three shade levels was studied at Varanasi, India, for five tree species of tropical dry forest, which differed in life-history traits. Three of these were small-seeded pioneer (Albizia procera, Acacia nilotica and Phyllanthus emblica) and the other two were large-seeded non-pioneer (Terminalia arjuna and Terminalia chebula) species. Seedlings of all the species were subjected to three light levels (80–100%, 20–30% and 3–7% of full sunlight) comparable to sunlit gaps and shaded microsites in the dry forest. After 4 mo of shade treatment, height, basal area, biomass and other growth traits, viz. RGR (relative growth rate), NAR (net assimilation rate), and SLA (specific leaf area) were determined. Etiolation and plasticity indices were calculated. Reduction in seedling height, biomass and relative growth rates and enhancement in SLA due to shade was greater for small-seeded pioneer species. Seedlings from large-seeded non-pioneer species exhibited a stronger etiolation response to shade than seedlings from small-seeded species. Phenotypic plasticity indices for basal area, plant biomass and relative growth rate were greater for the three small-seeded early successional species (A. procera, A. nilotica and P. emblica), indicating their specialization in a more favourable light environment such as large gaps and forest peripheries. The non-pioneer and pioneer species differed only in the degree of shade tolerance, and we suggest that dry forest species cannot be strictly categorized into two distinct groups (shade tolerant vs. intolerant), rather shade preference and gap preference would be the more expressive terms. Marked environmental heterogeneity in terms of irradiance and the phenology of dry tropical trees permits coexistence of species of varying ecological traits, contributing to the maintenance of diversity in the dry forest.

Stand structure dynamics during early secondary forest succession were related to mortality, growth and recruitment rates, and the dependence of these demographic processes on fallow age and initial stand structure attributes was evaluated. In 11 secondary tropical rain-forest sites (1.5–19 y) in Chiapas, Mexico, one plot of 10 × 50 m was established. Diameter and height were measured for all trees ≥ 1 cm dbh, and their survival, growth and recruitment was monitored over a 2-y period. Changes in stand structure were especially fast in the first 5 y of succession, and decreased rapidly afterwards, which resulted from similar stand-level changes in relative mortality, growth and recruitment rates. Demographic processes were negatively related with initial stand basal area, but independent of initial tree density. Basal area was a better explanatory variable of the among-stand variability in these rates than fallow age. Results suggest that asymmetric competition and resulting patterns of tree-thinning are major driving forces determining secondary forest successional pathways. Fallow age per se is a compound variable reflecting community organization at a certain point along the successional axis, while community structure drives succession. Sudden mass mortality among dominant species in some stands showed that early secondary forest succession is not always a gradual and unidirectional process.

The small size of many African protected areas makes adjacent rangelands potentially important in the local survival of wild animals. In order to assess the importance of pastoral areas to wild ungulates, we studied density and habitat choice of wild ungulates and cattle in Lake Mburo National Park, Uganda, the adjacent exclusively pastoral Nshara Dairy Ranch and on private land consisting of a mixture of ranching and subsistence farms. Transects, in the three land-use zones, were walked during the wet season and the data were analysed by DISTANCE sampling technique. We found significantly higher total density of wild ungulates on the dairy ranch compared with the National Park and private land. There was no significant difference in total wild animal density between the National Park and private land. Impala (Aepyceros melampus), zebra (Equus quagga), bushbuck (Tragelaphus scriptus) and waterbuck (Kobus ellipsiprymnus) had significantly higher densities on the dairy ranch compared to the National Park. Only eland (Taurotragus oryx) density was higher in the National Park compared to private land. Wild ungulates and cattle showed a high degree of habitat overlap, generally preferring open grassland. Our study shows that high densities of wild ungulates are not necessarily associated with protected areas. Pastoral areas may be important for populations of wild herbivores during the growing season despite a pronounced presence of livestock.

‘Size matching’ of forager size and prey size is an ecological mechanism that should partition diets both within ant colonies and ensembles. We studied the relationship between ant body size and food size in tropical leaf-litter ant ensembles at different levels. In three plots of primary rain forest in Sabah, Malaysia, we observed altogether 50 species of 18 genera of ants (e.g. Pheidole, Recurvidris, Lophomyrmex, Paratrechina, Odontoponera) that harvested seed particles from baits of milled rice of various particle size. At colony level ‘size matching’ of individual foragers with their load was found only in one of 12 tested species, viz. Pheidole lucioccipitalis. However, ant species differed considerably with respect to ‘seed’ size used: on average, foragers of larger species transported significantly larger ‘seed’ fragments. This was highly significant in a subset of ant species with at least 12 records of size choice per species. But when we investigated resource partitioning within each of the three ant ensembles with a null model, analysis proved that there was a significant overlap in seed size selection of species at two of the sites, thus indicating that food size choice of foragers had little influence on niche patterns and community structure of ants. Ant species with workers that were not matched with their seed resources (viz. Oligomyrmex or Pheidologeton) engaged in cooperative ‘seed’ transport, mass recruitment to and in situ mastication of rice fragments. Mean ‘seed’ transport distance was 38.4 cm.

Comparisons of sympatric reptile species were used to assess the variation in niche overlap for food between potential competitors at different trophic levels. Omnivorous tortoises and carnivorous vipers inhabiting the rain-forest region of West Africa were used as study models. Food niche overlap between species increased with habitat alteration in both the independent study systems: tortoises (Kinixys homeana and Kinixys erosa) and vipers (Bitis gabonica and Bitis nasicornis) showed lower values of Pianka's niche overlap index in the pristine habitat than in the altered habitat, and these differences in overlap values did not depend on chance after Monte Carlo simulations. There were higher inter-habitat food niche overlaps within-species than between-species. Permutation tests (assessed after 5000 iterations) revealed that, for both study systems, the P-values became significantly smaller with fewer resource states, thus showing the niche overlap between species really increases after habitat alteration. The observed increases in food niche overlap between species accomplished with rain-forest habitat degradation in turn may be predicted to have cumulative effects on reducing the level of forest biodiversity.

The species richness, number of plants, biomass, and species composition of holo- and hemi-epiphytes were recorded in fifty-six 0.04-ha plots, distributed over forest fallows of 2–30 y old and mature forests in lowland Amazonia (Amacayacu National Park, Colombia). A total of 9190 epiphytic plants representing 162 species were recorded on 4277 phorophytes. Seventy species were classified as holo-epiphyte and 85 as hemi-epiphyte. Aroids were most diverse (58 species) and represented 76% of the total recorded biomass. Anemochory was more dominant among holo-epiphytes and zoochory among hemi-epiphytes. The species richness, density and biomass of both holo- and hemi-epiphytes increased significantly from young fallows to old fallows and mature forests. Hemi-epiphytes had greater density and biomass than holo-epiphytes. In canonical ordination, forest age did not relate to the species composition of holo-epiphytes. However, for hemi-epiphytes, the age effect was significant, suggesting that species turnover takes place in the ageing fallows.

All plant species face a fundamental reproductive trade-off: for a given investment in seed mass, they can produce either many small seeds or few large seeds. Whereas small seeds favour the germination of numerous seedlings, large seeds favour the survival of seedlings in the face of common stresses such as herbivory, drought or shade (Leishman et al. 2000). One mechanism explaining the better survival of large-seeded species is the seedling size effect (SSE) (Westoby et al. 1996): because seeds with large reserves result in bigger seedlings, seedlings from large-seeded species would have better access to light and/or to reliable water supply than seedlings from small-seeded species.

Over the last few decades a rapid and extensive conversion of tropical forests to agricultural land has taken place resulting in mosaics of fragmented forest patches, pastures and farmland. While the effects of forest fragmentation on biodiversity have been intensively studied within the remaining forests, relatively little is known about the biodiversity in tropical farmland (Daily et al. 2001, Pimentel et al. 1992). Frugivorous birds are an important group of species in tropical farmland ecosystems. Frugivorous birds are significant seed dispersers and can play a prominent role in transporting seeds into disturbed areas and setting the stage for the regeneration of these systems. Isolated fleshy-fruited trees in agricultural landscapes have been shown to attract birds, leading to an increased seed rain and seedling establishment under their canopies (Carrière et al. 2002, Duncan & Chapman 1999, Guevara et al. 1986, 2004; Slocum & Horvitz 2000).

Sessile organisms interact locally on the scale of their body sizes, and one of the great advances in population and community ecology is the use of individual-based models to examine species interactions (Biondini 2001, Bolker & Pacala 1999, Pacala & Deutschman 1995, Pacala & Silander 1985, 1987; Silander & Pacala 1985). Canopies are often taken as a proxy for body size in the plant literature, even though roots can make up substantial amounts of a plant's biomass, have productivity that equals or exceeds the above-ground parts, and are critical in both competition and mutualisms involving nutrient capture and water balance (Biondini 2001, Casper & Jackson 1997, Casper et al. 2003, Rajaniemi & Reynolds 2004, Robinson 2004). Root zones, however, are seldom incorporated in plant ecological studies because they are exceptionally difficult to measure, and the importance of intra- and interspecific root interactions is little known (Schenk & Jackson 2002).

The population biology and ecology of most members of the neotropical family Centrolenidae, or glass frogs, are unknown. Glass frogs deposit their eggs in a gelatinous mass on vegetation overhanging streams, the eggs hatch, and the tadpoles drop into the water, where they complete development (Savage 2002). This study will contribute to our limited understanding of centrolenid reproductive ecology by quantifying variation in clutch size, embryonic development and embryonic mortality in a population of Hyalinobatrachium pulveratum.