In a mature lowland ‘terra firme’ forest near Araracuara in Colombia, a study was conducted to determine the above-ground biomass by means of regression analysis. Dry weight, DBH (i.e. stem diameter at 1.3 m above ground level, or just above buttresses if these surpassed 1.3 m in height), total height and specific wood density were measured on 54 harvested trees, chosen in a ‘selected random’ manner. Nine different regression models were evaluated for statistical correctness, accuracy of the estimates and for practical use. The logarithmically transformed models with DBH2, and DBH2 × height as independent variables appeared to be the only models meeting the above criteria, the latter being the most accurate.

The exclusion of big trees (DBH >45 cm) from the regression did not result in significant changes of the regression coefficients.

Five forest plots of differing disturbance levels were sampled qualitatively for termites in the Mbalmayo Forest Reserve, southern Cameroon, using 100 m belt transects. Results showed a large reduction in species richness in two severely disturbed plots compared with a Near Primary plot, but little difference in two less disturbed regenerating plots; there is some evidence for a slight increase in species richness in the regenerating plots. Soil-feeders predominate in the primary and regenerating plots, but are greatly reduced in the severely disturbed plots. Wood-feeders appear to be more resilient to disturbance than soil-feeders, although their species richness is low in the most disturbed plots. The Termitinae are the dominant taxonomic group in the Near Primary plot, the Apicotermitinae in the less disturbed regenerating plots and the Macrotermitinae in the clear felled areas. There appears to be no secondary invasion of plots by savanna-associated species, and the small apparent increase in species richness in the less disturbed regenerating plots appears to be due to the influx of forest species usually associated with natural gaps.

As forest fragmentation and intentional burning of grasslands increase, the frequency of fires penetrating the dry and subhumid tropical forests of Bolivia is also likely to increase. To expand our understanding of the role of fire in tropical dry forest, the physical and thermal properties of barks of tree species were studied to determine their relative resistances to cambial damage by fire. For 16 tree species found in the dry forest of the Lomerío region of eastern Bolivia, bark thickness, moisture content, and specific gravity were measured. Insulating capabilities of bark were measured by obtaining cambial and surface temperatures during experimental wick fires. Bark thickness on trees 5-100 cm dbh (diameter at 1.4 m) ranged from 2–51 mm and both thick- and thin-barked species were represented. For all species, bark thickness increased as stem diameter increased. Bark thickness explained more (63%) of the variation in peak cambial temperatures during fires than did bark moisture content (4%) or specific gravity (1%). A threshold bark thickness of 18 mm was associated with the ability to withstand lethal cambial temperatures during the experimental, low intensity fires. For 13 of the 16 species included in this study, trees ≤20 cm dbh have bark thickness below the threshold 18 mm and, therefore, are likely to experience cambial injury from low intensity fires. Our results suggest that the forest presently characteristic of the Lomerío region did not develop with frequent fires and that species composition is likely to be substantially affected by an increase in fire frequency.

Relative size at onset of maturity (RSOM) is defined as size at first reproduction divided by asymptotic maximal size. RSOM is remarkably constant among species within many higher clades of animals, but varies widely among tree species from the Pasoh Forest Reserve, Malaysia according to the work of S. C. Thomas. RSOM was examined for 16 mid-storey and canopy tree species from a second tropical forest at Barro Colorado Island (BCI), Panama. Interspecific variation in RSOM was equally large for BCI and Pasoh and was unrelated to gap dependence or life form for BCI species. The shape of the relationship between size and the proportion of individuals that were reproductive differed between forests, with an abrupt increase over a narrow range of sizes at Pasoh and a more gradual increase over a wider range of sizes at BCI. Both overtopping trees and heavy liana infestation reduced the probability that BCI trees were reproductive. This presumably reflects reduced availability of carbon for reproduction. We speculate that greater liana loads and a greater abundance of large, shade-casting trees at BCI may increase variation among individuals and contribute to the relatively wide range of sizes characterized by a mixture of sterile and fertile individuals observed for most BCI species.

The vertical stratification of bat communities in primary forests of the Central Amazon (80 km north of Manaus, Brazil) was investigated using capture nets in the canopy (17 to 30 m high) and in the understorey (from 0-2.5 m). Seventeen sites were sampled during one year (3398.5 mistnet-hours) and 936 individuals captured, belonging to 6 families, 29 genera and 51 species. Utilizing Non-Metric Multidimensional Scaling (NMMDS), a well-marked vertical stratification between the communities was verified, the canopy being the more utilized region. Fifteen species were exclusively captured in the canopy, 10 were predominantly captured in the canopy, and 12 species were exclus ively captured in ground nets. Species recorded and the communities they form were analysed using a matrix of guilds. The matrix obtained had 24 cells. A guild composed by background cluttered/gleaning frugivores was the richest in species (19), followed by background cluttered/gleaning insectivores (12 species). The results illustrate that when studying tropical forests it is highly desirable to involve both the lower and the upper part of the forests; otherwise the fauna would be merely subsampled, thus under-estimating the status and abundance of some species.

A demographic study was carried out on Bertholletia excelsa, the Brazil nut tree, in two primary forest sites in Northern Bolivia where Brazil nuts have been harvested for several decades. In spite of the large proportion (93%) of seeds that are harvested, reasonable densities of recently emerged seedlings were found. Seeds of Bertholletia are contained in woody fruits that are primarily opened by agoutis. Most fruits are left untouched on the forest floor for 1–2 y before they are opened, possibly due to high energetic costs of fruit opening just after fruit fall. However, the proportion of viable seeds is strongly reduced in older fruits.

Growth in diameter at breast height (dbh) was low for pole-sized trees (< 15 cm dbh) and adult trees (> 100 cm dbh) and peaked for intermediate-sized trees (30–60 cm). These trees often attained a growth rate of > 1.5 cm y−1, which is high compared with other non-pioneer tropical trees. This, and the strong growth response to increased light availability found for seedlings and saplings, suggest that Bertholletia excelsa can be classified as a gap-dependent species. Matrix population models were constructed for both study populations. Population growth rates (λ) were close to one, and were most sensitive to persistence in one size category. Age estimates revealed that age at first reproduction (at dbh > 60 cm) amounts to over 120 y, and age in the last category (dbh > 160 cm) to almost 300 y. Given the continuous rejuvenation of the population, the stable population size, the high age at maturity and the long reproductive period, it is concluded that current levels of Brazil nut extraction may be sustained at least for several decades and perhaps for even longer periods.

The dynamics of tropical forest woody plants was studied at the Nouragues Field Station, central French Guiana. Stem density, basal area, above-ground biomass and above-ground net primary productivity, including the contribution of litterfall, were estimated from two large permanent census plots of 12 and 10 ha, established on contrasting soil types, and censused twice, first in 1992–1994, then again in 2000–2002. Mean stem density was 512 stems ha−1 and basal area, 30 m2 ha−1. Stem mortality rate ranged between 1.51% and 2.06% y−1. In both plots, stem density decreased over the study period. Using a correlation between wood density and wood hardness directly measured by a Pilodyn wood tester, we found that the mean wood density was 0.63 g cm−3, 12% smaller than the mean of wood density estimated from the literature values for the species occurring in our plot. Above-ground biomass ranged from 356 to 398 Mg ha−1 (oven-dry mass), and it increased over the census period. Leaf biomass was 6.47 Mg ha−1. Our total estimate of aboveground net primary productivity was 8.81 MgC ha−1 y−1 (in carbon units), not accounting for loss to herbivory, branchfalls, or biogenic volatile organic compounds, which may altogether account for an additional 1 MgC ha−1 y−1. Coarse wood productivity (stem growth plus recruitment) contributed to 4.16 MgC ha−1 y−1. Litterfall contributed to 4.65 MgC ha−1 y−1 with 3.16 MgC ha−1 y−1 due to leaves, 1.10 MgC ha−1 y−1 to twigs, and 0.39 MgC ha−1 y−1 to fruits and flowers. The increase in above-ground biomass for both trees and lianas is consistent with the hypothesis of a shift in the functioning of Amazonian rain forests driven by environmental changes, although alternative hypotheses such as a recovery from past disturbances cannot be ruled out at our site, as suggested by the observed decrease in stem density.

Species composition and turnover that have occurred in a series of permanent sample plots established during the 1930s and 1940s in Budongo, a semi-deciduous Ugandan forest, are reported. The plots were established as part of a sequence first used to describe forest succession, five of which have been maintained and which were last measured in 1992-1993. One plot (plot 7) provides 53 y of data from old-growth pristine forest. Plot 15 was established in wooded grassland at the forest edge and is now closed high forest. Evaluation of the remaining three plots is complicated by silvicultural interventions carried out in the 1950s. Forty species have been added since the first evaluations and a total of 188 tree species (over 80% of Budongo's forest tree flora, and including two exotics) has now been recorded from within the plots. The pattern of shade-tolerance in the original plot series conforms to patterns expected for succession with an increasing proportion of shade-tolerant species with development, and large stems appearing to ‘lag behind’ smaller stems in this respect. The time series data are less consistent, and while plot 7 increased in the proportion of shade-tolerant stems through time, the proportion of shade-tolerant species actually declines. Stem-turnover (the mean of mortality and recruitment) slowed with implied successional stage. Most species have a higher recruitment than mortality rate and stem numbers have thus increased in all plots. This is most pronounced in the putatively ‘early successional’ plot. Stem size structure has changed within the plots, with an increased proportion of smaller stems. Species show different rates of turnover and these vary from plot to plot and period to period. In plot 7, the overall mortality rate decreased with initial stem size. Estimates imply that some tree species may easily live longer than 500 y after reaching 10 cm DBH, and that 1000 y is possible. The importance of large trees in determining forest dynamics is illustrated by the finding that death of only seven stems in plot 7 contributed over 60% of net basal area losses recorded over the 53-y observation period. Many of the observed patterns were not predicted and could only have been found by long-term studies.

The Smithsonian Venezuela Project (SVP) conducted extensive surveys of mammals and ectoparasites in the 1960s. The 25 238 individuals and 130 species of bat collected by SVP hosted 36 663 streblid bat flies, representing 116 species of these ectoparasitic dipterans. Roosts of bat species differ in durability and protection, and bat flies separate from the host to pupate in the roost. We predicted higher levels of parasitism and more parasitic associates for bats roosting in more permanent structures (e.g. caves, tunnels) that would facilitate their association with hosts. We also predicted wing development of flies should correlate inversely with roost duration, restricting flightless forms to bats in permanent roosts. Ranking roosting structures by durability and protection, we sought correlations among bat species in prevalence, mean intensity and number of associated fly species. All three measures of parasitism were positively and significantly related to roosting habits: bats roosting in more permanent, enclosed structures were more likely to be infested, to carry heavier parasite loads, and to harbour more species of ectoparasitic flies. However, roosting habits were not correlated with the average wing development of bat flies. Although other factors affect parasitism rates in bats, the study provides a compelling example of both ecological and evolutionary responses of parasites to features of the host's environment.

The role of chimpanzees (Pan troglodytes) as seed dispersers in the Kibale Forest Reserve of western Uganda was assessed by analysing 1849 dung samples from two chimpanzee communities and by conducting germination trials on dispersed and non-dispersed seeds. Of the chimpanzee dung samples, 98.5% contained seeds, and fig seeds were the most common. The number of large seeds (>2 mm) per dung sample was often high, even for species with a relatively large seed. For example, Mimusops bagshawei has a 15-mm diameter seed and had an average of 26.5 seeds per sample. In addition, many seed species were found in the dung repeatedly over several months. Germination trials demonstrated that the species of seed commonly collected from dung were viable. Comparisons of both the rate and success of germination of chimpanzee-dispersed seeds with seeds collected directly from the parent trees, showed that gut passage increased the speed and probability of germination. The number of large seeds dispersed by the chimpanzee population was estimated at 369 large seeds km-1 day-1. We suggest that in Kibale, chimpanzees may play a more significant role in primary seed dispersal than implied by their low numbers and biomass.

Forest structure and species distribution patterns were examined among eight topographically defined habitats for the 205 species with stems ≥ 1 cm dbh inhabiting a 25-ha plot in the Sinharaja rain forest, Sri Lanka. The habitats were steep spurs, less-steep spurs, steep gullies and less-steep gullies, all at either lower or upper elevations. Mean stem density was significantly greater on the upper spurs than in the lower, less-steep gullies. Stem density was also higher on spurs than in gullies within each elevation category and in each upper-elevation habitat than in its corresponding lower-elevation habitat. Basal area varied less among habitats, but followed similar trends to stem density. Species richness and Fisher's alpha were lower in the upper-elevation habitats than in the lower-elevation habitats. These differences appeared to be related to the abundances of the dominant species. Of the 125 species subjected to torus-translation tests, 99 species (abundant and less abundant and those in different strata) showed at least one positive or negative association to one or more of the habitats. Species associations were relatively more frequent with the lower-elevation gullies. These and the previous findings on seedling ecophysiology, morphology and anatomy of some of the habitat specialists suggest that edaphic and hydrological variation related to topography, accompanied by canopy disturbances of varying intensity, type and extent along the catenal landscape, plays a major role in habitat partitioning in this forest.

In a mangrove forest in southern Thailand, the root biomass was estimated for the Sonneratia zone, the Sonneratia-Bruguiera ecotone, the Bruguiera zone and the Rhizophora zone. A 0.2 m (width) × 15.5 m (length) × 1.0 m (depth) trench was dug between two trees of Rhizophora apiculata and the roots in the trench harvested, rinsed, sorted by diameter and weighed. The dry weight of roots collected from the trench was 135.6 kg, equivalent to a total root biomass of 437.5 t ha−1 including 220.5 t ha−1 of fine roots.

The root density distribution of a single tree was assumed to be predictable exponentially. Two variables of the equation were determined by iteration using the data from a trench excavation and tree distribution study. This equation was applied to 1.3 ha and the root biomass distribution was analysed. The root biomass in each of the above four zones was estimated at 171.8, 84.8, 242.6 and 509.5 t ha−1, respectively. The proportion of fine roots (<2 mm) ranged from 46.4% to 66.4%.

Data on the laxonomic composition of the arboreal arthropod fauna of 10 Bornean lowland rain forest trees are presented, based on samples obtained using insecticide fogging. Combined samples from all trees comprised 23,874 individuals of at least 3000 species. The mean number of species on each tree was 616.7 with one tree sample containing more than 1007 species. The relative rank of the major orders of arthropods in terms of both species and individuals was remarkably constant across the trees. Hymenoptera, Coleoptera, Diptera and Hemiptcra were the orders with the most species and individuals. Formicidae, though not particularly species rich, was the most abundant family in terms of individuals and the commonest species in six of the 10 trees was an ant. Refogging of one of the trees 10 days after initial sampling showed that the arthropod fauna had not completely recovered. What many arthropods are doing in the canopy is discussed.

Lianas (woody vines) contribute substantially to the diversity of woody plants in Yasuní National Park, Eastern Ecuador. In total 606 individuals, belonging to 138 species, were found in two 20-m × 100-m plots. The liana diversity was higher than in any comparable study, but the density was relatively low. Sapindaceae and Leguminosae were the most species-rich families, whereas Leguminosae and Celastraceae were the most abundant families. The number of liana individuals as well as the number of liana species was partially explained by forest structure, but 92% of the variation in number of liana species depended on the number of liana individuals. Areas with high density of small trees had high liana density, and areas with a high number of tree saplings had a relatively high diversity of climbing lianas. The probability of trees being colonized by lianas increased with tree diameter. The presence of one liana on a tree increased its risk of being colonized by additional lianas.

We studied the response of vegetation and vertebrate assemblages to fire and grazing, and their interacting effects, in Eucalyptus woodland in north-eastern Australia. In this vegetation type, many pastures remain free of cattle grazing due to the occurrence of a native shrub poisonous to livestock. Vegetation (floristic data and 22 habitat variables) and vertebrate fauna (birds, mammals, reptiles) were sampled in 29 standardized 50 × 50-m quadrats in the 2001 wet season, representing four treatments: sites burnt recently (within 2 y) and grazed by cattle (4–8 ha per livestock unit); sites unburnt (last burnt >2 y ago) and grazed; sites burnt recently and ungrazed; and unburnt and ungrazed sites. Fire and grazing had a significant influence on vegetation: both grazing and fire reduced ground cover (fire in grazed sites 51–23%, fire in ungrazed sites 68–39%) and increased the cover of forbs (8% in burnt and grazed sites, 3% if ungrazed) and tussock grasses (20% in grazed and unburnt sites and 5% when ungrazed). Grazing caused a shift in floristic composition from the perennial hummock grass Trioda pungens to tussock grasses (e.g. Aristida spp., Enneapogon spp.), forbs (e.g. Phyllanthus spp.) and shrubs (e.g. Acacia spp.). Of the vertebrate groups, birds responded more to fire effects (9 species), reptiles to grazing effects (6 species) and mammals to the interaction (2 species). Species reacted to increases in bare ground (e.g. crested pigeon Ocyphaps lophotes, hooded robin Melanodryas cucullatus, Ctenophorus nuchalis) and to the dominant ground cover (e.g. Ctenotus pantherinus) or change in vegetation architecture (e.g. singing honeyeater Lichenostomus virescens, variegated fairy-wren Malurus lamberti). The clearest example of an interacting effect was the cycle of complementary dominance between the rodents Pseudomys delicatulus and P. desertor, the latter's post-fire recovery becoming more muted in sites where cattle grazed (modelled time for population recovery twice as long as in ungrazed sites).

Seeds of the Brazilnut tree (Bertholletia excelsa Humb. & Bonpl., Lecythidaceae) sustain one of the most important extractive industries in neotropical forests. Yet little is known about the demography and seed dispersal ecology of Bertholletia, particularly in natural stands which have not been previously harvested. This study presents data on the population density, spatial distribution, and seed dispersal ecology of Brazilnut trees at a pristine stand located within the Kayapó Indian Area of southeastern Amazonia, Pará, Brazil. Brazilnut trees were primarily found within groves (castanhais) of 75 to 149 trees, with a few isolated trees in between. Although the density of trees ≥ 10 cm in diameter at breast height (hereafter, dbh) at two groves was 4.8 to 5.1 trees ha–1, the overall density for the entire study area of c. 950 ha was estimated at 1.3 tree ha–1. Within-grove nearest neighbour distances averaged 21 m and were markedly skewed towards even shorter distances. Seed dispersal experiments using 709 marked seeds indicated that this pattern can be largely explained by the highly restricted seed shadows imparted by the main seed dispersal agents of Bertholletia at this site, the red-rumped agouti (Dasyprocta leporina). Agoutis on average scatterhoard Bertholletia seeds to an average distance of 5 m, and rarely beyond 20 m, from seed stations. We suggest that, once edaphic and climatic conditions are suitable, the highly contagious spatial distribution of Bertholletia trees at the landscape level can be largely accounted for by the quantitatively dominant effect of short-distance dispersal by caviomorph rodents, and rare events of long-distance dispersal provided by other vectors. This mechanism of grove formation need not resort to untested conjectures of human dispersal and intentional planting in prehistoric and historic times as it has often been suggested in the literature.

In central Amazonian white-water floodplains (várzea), different forest types become established in relation to the flood-level gradient. The formations are characterized by typical patterns of species composition, and their architecture results in different light reflectance patterns, which can be detected by Landsat TM image data. Ground checking comprised a detailed forest inventory of 4 ha, with Digital Elevation Models (DEM) being generated for all sites. The results indicate that, at the average flood level of 3 m, species diversity and architecture of the forests changes, thus justifying the classification into the categories of low várzea (várzea baixa) and high várzea (várzea alta). In a first step to scale up, the study sites were observed by aerial photography. Tree heights, crown sizes, the projected crown area coverage and the gap frequencies provide information, which confirms a remotely sensed classification into three different forest types. The structure of low várzea depends on the successional stage, and species diversity increases with increasing age of the formations. In high várzea, only one successional stage was found and species diversity is higher than in all low-várzea formations. The more complex architecture of the high-várzea forest results in a more diffuse behaviour pattern in pixel distribution, when scanned by TM image data.

Phenological observations were made in a Sumatran rain forest during three years (1980–1982). Phenological changes followed a consistent seasonal pattern. The abundance of young leaves and the fall of leaf litter peaked between December and February (first dry season); flowers were most abundant between January and April (first dry and first wet sea son), and ripe fruits in July-August (the second dry season). The fruit of strangling fig trees showed peaks in April and October, both wet season months. Within the study area there was variation in both the phase and the amplitude of the phenological cycles. One year, 1981, displayed mast flowering and fruiting. The observations indicate that the conditions for production were better during the mast year, a finding that facilitates our understanding of the evolution of mast fruiting.

A 100-ha quadrat of primary rainforest in French Guiana was ccnsused over two consecutive years by the mapping method, and a 24-ha core area was intensively mistnetted at the same time. More extensive surveys were conducted in successively larger areas including up to a large part of the forested interior of the country. From a total of 441 resident species found in the rainforest zone, 248 were regularly recorded within the 100-ha plot, 157 of which had mean densities of more than 1 pair km-2. The total estimated density of birds was at least 829 pairs km-2. Only two dominant species exceeded 20 pairs km-2, while 52% of all the species were represented by less than 2 pairs km-2. The distribution of body masses and niche characteristics among the bird community is given as well as the sample bias of the mistnet captures.

Rare species were a key component of this primary rainforest community. Several patterns of rarity were described and no marked and constant relationships were found between rarity and body size, habitat, diet, foraging behaviour or social system. Among a wide set of environmental constraints, the population of each species may be limited by a different combination of factors. Habitat heterogeneity, low, irregular and seasonal food availability, interspecific competition or heavy predation pressure were prominent and likely causes of the low density of most species which in turn facilitated the high species richness.

The effects of tropical forest fragmentation on frugivorous and nectarivorous bats were studied on recent islands created by the flooding of a dam in French Guiana. Study sites include forest islands isolated by water, and control plots in nearby continuous forest. Studies began 1y before the onset of forest fragmentation and encompassed the first 3y after flooding. Forest fragmentation greatly modified the diversity and abundance of bats. Changes occurred more rapidly in the smallest fragments than in the largest one, and trends were remarkably similar among all the studied islands. Bat captures in islands were characterised by the scarcity of understorey frugivores. Differences in home range size and foraging strategies may explain why understorey fruit bat species are more sensitive to fragmentation than canopy ones. Changes in the frugivorous bat community may have indirect consequences on both the demographic and the genetic structures of plant populations inside forest fragments.