A high-diversity primate community consisting of 13 species was examined at an entirely undisturbed and remote terra firme forest near the headwaters of the Urucu River, Amazonas, Brazil. Twelve of these species used a 900-ha study plot, located 4 km or more inland from the river, and comprised a total group density of 16 groups km−2, an overall density of 146 individuals km−2, a crude community biomass of 381 kg km−2, and a metabolic biomass of 278 kg0.75 km−2. These species were occasionally found together within small (≥1 km2) areas of forest, so long as they included the full complement of available forest types. More frequently, however, only half the species co-occurred in a single habitat type at any one time, reflecting a relatively consistent pattern of subcommunity spatial organization. Habitat generalist species making extensive, yearround use of the predominant habitat type (high forest), were common, whereas habitat specialists were uncommon to very rare, and confined to minor habitats. Those species coexisting throughout the year in high forest were found to segregate vertically from one another. These and other modal patterns of community structure were disrupted seasonally by long-distance, inland movements by two ‘igapó’ forest species living in large groups – squirrel monkeys and white-face capuchins – to high forest areas, which apparently resulted from a food bottleneck in their habitat. Another species found in large groups – woolly monkeys – also exercised seasonally vagrant movements within the high forest matrix, which coincided with periods of greatest scarcity of mature fleshy fruits.

Floristic and structural comparisons and a numerical classification were performed to identify the priority areas for conservation of genetic resources in central Brazil. The main type of woody vegetation, cerrado sensu stricto (savanna woodland with 10–60% tree cover), was sampled under a uniform methodology in six selected sites (total sample of 6 ha) scattered in a physiographic unit called the Pratinha Plateau located between 15–20° S and 46–49° W. A total of 6971 trees belonging to 139 species and 42 families was measured. Each site had a combination of 50–80 species and approximately 30 families but only 22 species from 14 families were common to all sites. Similarity decreased with distance, indicating the existence of geographical gradients as the physical characteristics were relatively uniform and the soils were dystrophic at all sites. The most dissimilar sites, in the localities of Patrocínio-MG and Paracatu-MG were recommended for the establishment of conservation units. The already existent conservation units arc all clumped in the Federal District and are not enough to protect the genetic heritage of the plateau. Floristics and phytosociological surveys based on land system zoning were recommended as useful tools in planning conservation areas at a regional level to preserve the maximum biodiversity.

Data on litterfall, litter layers accumulated on top of the mineral soil and layer mineralization collected for three years in a tropical rain forest at Pernambuco, Brazil, were implemented on a simulation model. Litterfall was collected biweekly using 11 collectors 1 × 1 m. Every three months, 20 litter mat samples, 0.5 × 0.5 m, were collected, divided into the L, F and H layers and the CO2 evolution from each litter layer was determined in the laboratory. Litterfall, in the three years, averaged 7.8, 8.3 and 8.2 Mg ha-1 y-1, most of it leaves. Litter mat masses varied widely from place to place (15–90 Mg ha-1) and the overall averages were 5.6, 7.6 and 26.1 Mg ha-1 for the L, F and H layers, with CO2 evolution averages of 2.27, 0.507 and 0.123 mgC g litter C-1 day-1. According to the model, the L layer had a high turnover rate, losing 4.7 Mg ha-1 y-1 through mineralization and 3.4 Mg ha-1 y-1 transferred to the F layer. Values for the F and H layers were 1.4 and 1.2 mineralized and 2.0 and 0.8 Mg ha-1 y-1 transferred. Thus, it would take 2.1, 6.7 and 39 years for newly fallen leaves to be mostly fragments, to be transformed to humus and to be incorporated to the soil organic matter, respectively. Variations of litterfall throughout the three years would have little effect on the system which was not very sensitive to litterfall changes, except for the top layer.

The food habits of four hypogeous Macrotermitinae species were studied using stable carbon isotope analysis in several biotopes of a humid savanna of West Africa (Lamto, Côte d'Ivoire). The proportion of woody and herbaceous material in the diet of the different species was determined by measuring the 13C natural abundance in the fungus comb. The diet varied with season and biotope, especially tree density. The results confirm the flexibility of the food habits of the fungus-growing termites as a consequence of their exosymbiosis with the fungus Termitomyces sp.

Air and soil temperatures were measured during dry season heading fires in three different physiognomic forms of native vegetation common in Central Brazil: cerrado sensu stricto (dense scrub of shrubs and trees), campo cerrado (open scrub), and campo sujo (open grassland with scattered shrubs). The vegetation was protected from fire for 15 y in some areas, had been burned once every two years, and once each year in other areas. The temperatures were measured with type-k thermocouples and recorded at intervals of 22.5 sees. Air temperature measurements were taken at 1, 60 and 160 cm. Maximum air temperatures ranged from 85°C to 840°C, and the duration above 60°C varied from 20 to 270 seconds. In the soil, negligible temperature increases were recorded below 5 cm depth, whereas at 2 cm maximum soil temperatures varied from 29 to 38°C. Possible influences of fuel load and moisture on the recorded data and on the behaviour of fire in the cerrado ecosystems are discussed.

Soil seed banks from four Mexican lower montane forests were studied. These forests are in the transition zone between neartic and Neotropical floristic regions. Soil seed bank characteristics were biased toward the dominant – temperate or tropical – phytogeographical affinity. The northern site (El Cielo Biosphere Reserve, Tamaulipas) had the lowest seed density, 720 seeds m−2. The eastern (Rancho La Mesa, Veracruz), western (Sierra de Manantlán Biosphere Reserve, Jalisco) and southern (El Triunfo Biosphere Reserve, Chiapas) sites all had much higher seed densities (2291, 2901, and 2341 seeds m−2, respectively). Germinated tropical species seeds were more abundant than temperate species seeds at all four sites, and were higher for southern and intermediate sites when compared with the northern site. On the basis of family, El Cielo appears to be different. The tropical family Melastomataceae was absent from this northern site, whereas it represented >30% of total germinated seeds at the other three sites. Trees were the most abundant life form at the southern site but were particularly poorly represented in the eastern and northern sites. Herbs were very abundant in the eastern, western and northern sites but were poorly represented in the seed bank of the southern site. The four seed banks differ between themselves and resemble seed banks of neither temperate deciduous nor tropical forests. In some ways they are unique, at least with respect to floristic composition, seed density, and life form representation.

Descriptions of the fine scale distribution of organisms have frequently been used to investigate various ecological phenomena. Unfortunately, the most widely used spatial analysis techniques are based on single index statistics, which convey only minimal information about the biological processes underlying the studied distributions. Such statistics cannot detect changes in pattern over different scales, and cannot identify some types of distribution. Additionally, both the use of such statistics on the distribution of individuals which have a non-negligible size, and the frequent failure to use an edge correction for points close to the boundaries of a sampled area, have led to the over-reporting of ‘spaced out’ (‘regular’) distributions. Using two spatial distributions recently analysed with a single index statistic (termite mounds, and earthmounds created by termites), I illustrate the benefits gained from using the spatial functions K(t), G(y) and F(x) to analyse both ‘point events’ and events which have a non-negligible size. These functions are considerably more informative about the nature of a spatial pattern and offer wide scope for the fitting of spatial models to biological distributions.

This paper introduces studies of nutrient cycling in disturbed and undisturbed rain forest plots in the upper catchment of Birthday Creek, near Paluma, North Queensland. The catchment is underlain by granite and has soils of comparatively low fertility. Differences between unlogged plots and plots disturbed 25 years previously by selective logging are still apparent. Disturbed plots have soils with higher bulk densities and pH, lower CEC, kjeldahl nitrogen and available phosphorus concentrations, and changed species composition. The data suggest that recovery from selective logging is dependent on soil fertility and intensity of disturbance.

Rates of litterfall over a period of 3 y at one undisturbed site and two sites disturbed by selective harvesting are reported for a tropical rain forest area in North Queensland, Australia. Litterfall over a period of 18 months is also reported for a further undisturbed site in the same area. No significant differences were found in annual litterfall between the sites, with annual litterfall rates ranging from 5.0 to 6.0 t ha−1 y−1. These rates are considerably lower than those reported for other rain forest areas in Australia, and are within the lower ranges recorded for other tropical rain forests. Litterfall was found to be strongly seasonal at all sites with the maximum falls occurring from the end of the dry season to the end of the wet season. The between year variability of the sites was quite low, with the ratio of maximum to minimum production ranging from 1.05 to 1.20. The average percentages of leaves, wood and reproductive material in litterfall were similar at each site. Leaves were the dominant component of litterfall with the average proportion of the total litterfall ranging from 72% to 76% over the study period at each of the four sites. At certain times, however, the fall of wood and reproductive material was quite significant, comprising as much as 71% and 34% of litterfall respectively. A strong negative correlation was found between the fall of leaves and wood at all sites.