Bird species were surveyed in a variety of habitat types in a lightly populated area of western Brazilian Amazonia. Habitats surveyed were untagged forest, selectively logged forest, ‘capoeira’ (scrub regenerating on cleared land), cropfields, and an isolated 35 ha forest ‘island’. All habitats were within a few kilometres of each other. Many bird species were found to occur in all or most habitats, although the similarlity of the species assemblages dropped with increasing levels of disturbance of the natural vegetation. Certain understorey insectivores common in untagged forest were rarely observed in other habitats; disturbed areas contained higher numbers of flycalching insectivores or birds feeding on both insects and fruit, some of which were open-country rather than forest species. The avifauna of the forest ‘island’ was more similar to that of regenerating scrub than to that of tall forest, and the persistence of forest species was attributed to their ability to move to and from nearby continuous forest areas.

Using data from forest sites in Sarawak, East Malaysia, a demographic comparison is made of pioneer species in the genus Macaranga with understorey trees of the primary forest in the closely related genus Mallotus. In primary forest, saplings and trees of these genera arc uncommon and have low growth rates. For the six years following disturbance by logging, a Macaranga population was characterized by high rates of diameter growth and recruitment. Eleven years after disturbance, seedling and sapling abundance had declined sharply. In contrast, the Mallotus population was stable in size class distribution and increased slightly in total population size during the 11 years after logging. Growth rates of Mallotus trees increased after logging but were still much less than Macaranga trees.

A morphological comparison shows that most Macaranga species, in contrast with Mallotus species, have large hollow twigs that are occupied by ants and have larger, hairier, thicker, more toothed leaves with longer petioles, axillary inflorescences with more flowers, infructescences with more fruit, fruits that are smaller in size and smaller seeds.

The dispersal of seeds by howler monkeys (Alouatta palliata) and the activity of dung beetles in modulating the fate of the dispersed seed were studied at Los Tuxtlas, Mexico. Howlers consumed the fruits of 35 species of plants. The seeds of 28 of these were dispersed by the monkeys. The majority (≥90%) of the seeds dispersed by monkeys were destroyed by rodents. Rapid relocation and burial of dung by dung beetles resulted in accidental relocation and burial of large numbers of seeds shortly after deposition. Faecal clumps (20 mg) remained on the ground for an average of only 2.5 h (range 1–3 h). Ball rolling beetles transported balls up to 5.0 m from the site of deposition (range 1–5 m). Burrowing and ball-rolling dung beetles buried seeds at depths ranging from 2.5 to > 12.0cm. The deeper a seed is buried, the less likely it is to be found and eaten by rodents. Eighty percent of the species used by Alouatta as sources of fruit at Los Tuxtlas benefited by the dispersal and post dispersal service provided by howlers and dung beetles respectively. Seasonality in dung beetles abundance in the forest may influence the number of seeds per species escaping post-dispersal predation during the year. Dung beetles play not only an important ecological role in the recycling of matter and energy in the ecosystem, but also in the process of rain forest regeneration.

Population energy budgets of four sympatric isopods, Burmoniscus ocellatus, Burmoniscus sp. (Philosciidae), Formosillo raffaelei and Orodillo maculatus (Armadillidac), were calculated at two sites in a Hong Kong mixed forest. Mean population standing stock of the four species ranged from 1.25 to 13.61 kj m-2. Mean total isopod biomass at the two sites was 0.85 and 1.85g m-2. Annual production of the four species varied from 4.24 to 30.43 kj m-2 with 5.2–26.4% (energy content) allocated to reproduction. Annual production of the four species combined at the two sites was 23.22 and 50.15kJ m-2. ratios varied from 2.24 to 4.27. Annual population consumption on mixed litter by the four isopod species ranged from 28.4 to 183.0 kj m-2. Total annual consumption by all species at the two sites was 233.7 and 360.2 kj m-2 (1.96 and 2.53% of annual litter fall). Isopod respiratory rales were proportional to body weight, with an exponent ranging from 0.68–0.78 at 23°C. The annual population metabolism of the four species varied from 14.62 to 86.11 kj m-2. Total annual isopod metabolism at the two sites was 116.33 and 185.95 kj m-2. These results are discussed with reference to the role of isopods in litter breakdown and energy flow in the Hong Kong forest.

All woody growth was enumerated in three transects, each 960 m x 20 m, in derived savanna in the Olokemeji Forest Reserve in south-west Nigeria. The data for species density, species basal area and stem girth classes were analysed by nested-block analysis of variance and covariance. The use of ‘total covariance’, the sum of all covariances at a block size, contributes to understanding of the pattern present.

Three scales of pattern were evident. Patchiness at 160–320 m is interpreted as a response to soil differences or to the pattern of previous farming, which may itself have been determined by soil differences. Patchiness at 20–40 m is attributable to varying intensity of burning. At the smallest scale, of 10 m, there is evidence of regularity of distribution resulting from interference between individuals, possibly due to competition for water.

Abundance and distribution of frogs inhabiting the litter layer of an area of primary lowland rain forest in Central Amazonia were studied over a period of 15 months by sampling 498 plots each 5m × 5m. The litter frog fauna of the area consists of 23 species, but only 12 of these were encountered in the plots, and 84% of the frogs encountered belonged to only six species. Total abundance and diversity within the plot data are strongly seasonal and peak in the late wet season. Both are positively correlated with litter volume and moisture. Most of this variation is due to seasonality of reproduction, as indicated by patterns of occurrence of juveniles of the most abundant species.

These results indicate that the plot sampling method docs not sample the entire fauna adequately. Since this technique has been used to study other tropical forest litter herpetofaunas, however, comparison with other studies may be useful. Species diversity of litter frogs appears to be approximately the same in lowland primary forest sites studied, averaging around 20 ‘regular’ species. Abundances, however, vary widely. Central American communities contain 14–15 frogs (100m)-2, African 9–10, South American 4–6, and South-east Asian 1–2. These differences may be due to differential nutrient availability in forests of different ages and/or on different soils.

The distributional patterns and food preferences of some cerrado vegetation termites were studied in an area of 2500 m2 in south-east Brazil. All the termitaria were mapped and opened for identification of the species present. Their spatial distribution was studied using the nearest neighbour method (Clark & Evans 1954). We found 46 termite species in 450 nests. The majority of them were wood feeders. The general distribution of termitaria was regular. The distributional patterns of grass and litter feeders were not significantly different from a random distribution. Soil and organic matter feeders were contagiously distributed. The distribution of wood feeders was also contagious. Their spatial distribution was largely influenced by the distribution of the species Nasutitermes sp. 1 and Armitermes euamignathus. These two species occurred in two distinct patches, suggesting resource partitioning.