Field experiments and survival analysis were used to test whether the Janzen-Connell model operated for two common Amazonian tree species in Peru: the midstorey palm, Astrocaryum murumuru, and the canopy-emergent legume, Dipteryx micrantha. Seed and seedling survival patterns of these species partially supported the model, depending on tree species, type of predator, spatial scale and the particular year. At a small scale of a 2.5-ha plot, Astrocaryum seed predation by insects and mammals was density-dependent. At a larger scale of 100 ha, Dipteryx seed survival increased with distance from conspecific adults, but for Astrocaryum seeds there was no distance dependence. At the scale of 200–400 ha Dipteryx seed survival was negatively related to the number of individuals present in groups of conspecific adult trees. In 1992 a higher proportion of Astrocaryum seedlings survived far from, compared with close to, conspecific adult trees, whilst in 1993 more Dipteryx seedlings survived beneath conspecific adult trees than farther away.

Seasonal changes in the densities of dormant seeds in the soil around eight pioneer trees in the 50-ha Forest Dynamics Plot, on Barro Colorado Island, Panamá were studied, and how seed dispersal and seed dormancy influenced patterns of seed abundance and distribution were examined. Twenty-four, 3-cm-deep soil samples were collected on 30 m transects radiating out from each of the trees in each of four time-intervals through the year, and four 21-cm-deep samples were collected beneath the focal tree crowns. In the surface 0–3 cm of soil, germinable seed densities of all species combined declined from a peak of 1090 seeds m−2 in the mid-wet season in August, to 330 seeds m−2 by the end of the wet season in November. In contrast, at soil depths >3 cm, there was little variation in soil seed bank density through the year. Some variation in soil seed bank density for individual species could be accounted for by distance to reproductive conspecifics. Among species, abundance in the soil was negatively correlated with seed size. Seed persistence varied greatly among species at this site; after 1 y of burial in mesh bags, seed germinability of four species was near zero, while four other species showed no consistent decline in seed germinability after >2 y of burial. For at least one species, Trema micrantha, prolonged seed dormancy was also possible under natural conditions. Twenty-five percent of Trema seeds extracted from the soil at a site occupied by an isolated Trema tree that died between 1982 and 1985 were still germinable in 1994.

The importance of figs as keystone resources depends partly on their year-round fruiting pattern. However, most non-equatorial fig species show seasonality in reproductive phenology. If fig and other species' fruiting peaks coincide, the importance of figs as key resources during lean fruiting periods may be reduced. Fig and other forest tree phenology was studied in a deciduous and an evergreen site in the Western Ghats, southern India, from January 1991 to May 1992. In the deciduous site, most non-fig flowering and leaf flushing occurred during the two dry seasons, while most fruiting occurred during the dry and early wet seasons. In the evergreen site, all non-fig tree phenological stages – flowering, fruiting and leaf flushing – were concentrated during the two dry seasons annually (dry and post-monsoon). Individual fig trees produced crops one to three times annually, at different times of the year for different trees. Fig community fruiting peaks in the evergreen site coincided with fruiting peaks of the other forest trees, but in the deciduous site occurred slightly offset from non-fig fruiting peaks. The role of figs as key resources is likely to differ among forests, and to depend on factors such as diversity of fruiting pattern among fig species, fig tree density and territoriality of frugivores.

One of the major uncertainties concerning the role of tropical forests in the global carbon cycle is the lack of adequate data on the carbon content of all their components. The goal of this study was to contribute to filling this data gap by estimating the quantity of carbon in the biomass, soil and necromass for 23 long-term permanent forest plots in five life zones of Venezuela to determine how C was partitioned among these components across a range of environments. Aboveground biomass C ranged from 70 to 179 Mg ha−1 and soil C from 125 to 257 Mg ha−1, and they represented the two largest C components in all plots. The C in fine litter (2.4 to 5.2 Mg ha−1), dead wood (2.4 to 21.2 Mg ha−1) and roots (23.6 to 38.0 Mg ha−1) accounted for less than 13% of the total C. The total amount of C among life zones ranged from 302 to 488 Mg ha−1, and showed no clear trend with life zone. In three of the five life zones, more C was found in the dead (soil, litter, dead wood) than in the live (biomass) components (dead to live ratios of 1.3 to 2.3); the lowland moist and moist transition to dry life zones had dead to live ratios of less than one. Results from this research suggest that for most life zones, an amount equivalent to between 20 and 58% of the aboveground biomass is located in necromass and roots. These percentages coupled with reliable estimates of aboveground biomass from forest inventories enable a more complete estimation of the C content of tropical forests to be made.

The amount of leaf litter fall produced by different tree species in tropical forests varies in space and time. Falling litter may cover seeds and thereby enhance their survival by making their detection by seed predators more difficult. Tests were made to determine whether Astrocaryum murumuru and Dipteryx micrantha seeds survive better in microsites covered by leaf litter. Seed numbers and litter cover on the forest floor were experimentally manipulated. How natural variation in leaf litter cover and thickness affects seed and seedling survival of these two plant species was also examined. Seed survivorship was significantly higher for both plant species in microsites with leaf litter than in those with bare soil. Results from an experiment in which the litter was not disturbed showed that Astrocaryum seed survival was positively correlated with litter thickness (defined as the number of overlying dead leaves). Astrocaryum seedling survival was also significantly affected by leaf litter; more seedlings survived in shallow litter. Leaf litter had no effect on Dipteryx seedling survival. The results of the study suggest that early recruitment of both Astrocaryum and Dipteryx is influenced by the spatial distribution and amount of forest leaf litter.

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.

In a 1-y study of vegetation harvested by the leaf-cutting ants, Atta colombica Guérin, daily harvesting activity of two nests was observed for 24 h at c. 1-wk intervals (colony I: June 1993-June 1994; colony II: February-June 1994) on Barro Colorado Island. The average daily quantity of green leaves harvested by colony I was higher during the wet season (11.4 m2 d−1) than during the dry season (9.0 m2 d−1), but was highly variable between survey days. Total annual herbivory of green leaves was estimated to be 3,855 m2 foliage area for colony I and 1,707 m2 for colony II. Total dry weight of biomass harvested was higher in the dry season because most material collected during the wet season consisted of green leaves, while during the dry season, more than 50% of the total collected biomass was non-green plant material (stipules of Ficus sp., fruits, seeds, and flower parts of a variety of other species) which represented c. one third (111 kg y−1) of the total annual intake (370 kg y−1) of plant material. Total daily biomass intake was negatively correlated with daytime rainfall. The peak of daily foraging was affected by timing and duration of rainfall events. Highest input rates normally occurred between 15:00 and 16:00 h (colony I).Dry weight and surface area of harvested leaf fragments differed between plant species, with thicker leaves generally being cut into smaller pieces. Significant linear correlations were found between total daily harvest of fragments and the respective harvesting rate at the maximum of daily activity. High correspondence was found between estimates using this relationship and the measured daily leaf harvest of four other Atta colonies and of two colonies reported in published literature. The use of this relationship as a research tool is discussed.