Natural seed deposition patterns and their effects on post-dispersal seed fate are critical in tropical tree recruitment. Previous research showed that the key dispersal agent of the neotropical tree, Virola calophylla, is the spider monkey (Ateles paniscus). Spider monkeys generate a heterogeneous seed deposition pattern because they scatter-disperse seeds diurnally, whereas they clump-disperse seeds at their sleeping sites. The recruitment consequences of this pattern were investigated using manipulative experiments and observations. Scatter-hoarding by spiny rats (Proechimys spp.) caused little rearrangement of the initial seed deposition pattern because they moved seeds only short distances. Seed survival to the seedling stage depended negatively on conspecific seed density and positively on the distance from the nearest adult V. calophylla female. These effects were likely mediated by two important seed predators, spiny rats and beetles (Scolytidae). Furthermore, spider monkeys' seed deposition patterns influenced seed survival. Scatter-dispersed and experimentally dispersed seeds had the highest survival. Conversely, clump-dispersed seeds at sleeping sites, which are far from V. calophylla females, and non-dispersed seeds had equally low survival, suggesting that conspecific density- and distance-dependence acted independently and did not explain all variation in seed survival. Instead, other characteristics of the seed deposition pattern, such as the multi-specific assemblage of seeds at sleeping sites, also affected post-dispersal seed fates.

Resumen: La conexión entre el patrón natural de dispersión de semillas con el destino después de la dispersión es clave para el reclutamiento de árboles tropicales. Mediante experimentos y observaciones se investigó esta conexión utilizando el árbol neotropical Virola calophylla (Myristicaceae). Ateles paniscus (maquisapa), el principal dispersor de sus semillas, genera un patrón de deposición heterogéneo. Durante el día las semillas son depositadas de manera esparcida mientras que en los dormitorios las semillas son depositadas de manera agregada. En este estudio se encontró que los roedores espinosos (Proechimys spp.) almacenaron semillas individuales debajo de la hojarasca, sin embargo estos no alteraron el patrón de dispersión ya que las semillas fueron transportadas distancias cortas y la tasa de predación fue alta. Se encontró que la tasa de sobrevivencia hasta la etapa de plántula tuve una relación negativa con la densidad de las semillas y una relación positiva con la distancia al árbol hembra de V. calophylla mas cercano. Estos efectos sucedieron por medio de roedores espinosos y coleópteros (Scolytidae), predadores importantes de las semillas de V. callophylla. Adicionalemente, el patrón de deposición de los maquisapas influenció la sobrevivencia de las semillas. Tanto las semillas dispersadas por los maquisapas como las dispersadas experimentalmente tuvieron la tasa de sobrevivencia más alta. Por el contrario, tanto las semillas depositadas en los dormitorios, usualmente lejos de hembras de V. calophylla, como las semillas que cayeron debajo del árbol hembra tuvieron bajos niveles de sobrevivencia. Estos reultados sugieren que tanto la densidad como la distancia tuvieron efectos independientes y no explicaron toda la variación observada en la sobrevivencia de semillas. Por el contrario, otras características de la deposición de semillas tales como la riqueza de especies de la comunidad de semillas en los dormitorios también afectaron el destino de las semillas después de dispersadas.

The natural seed shadow created by gibbons (Hylobates mulleri×agilis) in Central Kalimantan, Indonesia, was monitored over 11 mo to discern the role of gibbons and post-dispersal events in the spatial pattern of seed germination. Variability in the content and distribution of 183 scats was used to determine which, if any, scat characteristics influenced seed fate. Nine scat characters were evaluated: (1) seed number; (2) number of seed species per scat; (3) scat weight; (4) seed load; (5) rainfall; (6) scat density; (7) distance to nearest fruiting tree; (8) ripe fig abundance; (9) non-fig fruit abundance. More than 99% of monitored seeds were killed, removed, or had germinated during the monitoring period. Vertebrates killed or removed most seeds (86%) and the probability of them moving seeds was highly dependent on non-fig fruit abundance at the time of deposition; factors (2), (6) and (7) also influenced seed removal/predation by vertebrates, depending on whether seeds were deposited in peak or non-peak times of consumption. Insect predation (2% of seeds) occurred mainly in scats that were deposited in months of high ripe fig abundance, while the actual chance of a seed germinating (11% of seeds) was influenced by non-fig fruit abundance at time of deposition and number of species in the original scat. The gibbon-generated seed shadow was profoundly altered by post-dispersal events and variation in the characteristics of the shadow had little lasting impact on the probability of seeds germinating.

The distribution of the forest tree Pisonia grandis (Nyctaginaceae) coincides with seabird colonies on small tropical islands. Its seeds are enclosed in a calyx exuding extremely sticky resin which adheres strongly to feathers. Birds are obviously seed-dispersal vectors, but the multi-seeded infructescences frequently entangle birds, often fatally. On Cousin Island, Seychelles, I investigated the production, germination, survival, and tolerance to seawater of Pisonia seeds, and the occurrence of entanglement with birds. Fresh seeds had high germination success (62–87% in experimental trays), but seedling survival in Cousin's forests was low (0.1% of 6020 seeds survived as seedlings after 2–8 mo). Some seeds tolerated 30 min daily immersions in seawater (8–15% germination after 14–28 d treatment), but not continuous immersion in seawater (reduced germination after 5 d and none after 12 d). Inter-island dispersal is likely via living unencumbered seabirds carrying a few seeds, but not on floating carcasses. Seeds attached to carcasses did not have improved germination or survival. Intra-island propagation is almost entirely vegetative, with negligible local seed dispersal. Pisonia plants therefore do not benefit from fatal entanglements. The extreme stickiness of the seeds evidently evolved to resist removal by seabirds and so facilitate long-distance dispersal. The mortality of some potential vectors is an unfortunate consequence, but has little impact on the large populations of tree- and ground-nesting seabirds which nest in or under these trees.

Along eight forest–savanna transects at seven semi-deciduous forest islands in the southern Comoé National Park data on spatial distribution of tree-size classes and environmental parameters (fuel load, shading by upper tree layers, and soil depth) were collected. For dominant tree species, a sequential series was observed from the forest border into the forest interior. At the forest border, Anogeissus leiocarpus was the most abundant tree with juveniles (<1 cm dbh) reaching highest density values (mean of 502 individuals ha−1) at the outer periphery of the forests. Regression analysis of juveniles of dominant tree species and environmental parameters resulted in a separation of forest and savanna species. Forest tree species regenerated well at forest sites, but also in the shade of A. leiocarpus stands. We conclude that (1) the studied forest islands advance against savanna by sequential succession, and (2) A. leiocarpus has a high potential to regenerate at savanna–forest boundaries under moderate fire impact and on rather shallow soils. The potential of A. leiocarpus to act as an important pioneer in the replacement of savanna by forest due to its effective regeneration at savanna sites and subsequent modification of site conditions, especially fire intensity by shading out savanna grasses, is discussed.

Major components of the flux density of global photosynthetically active radiation (PAR) were measured above and within canopies in a tropical deciduous forest on the Pacific coast of Mexico. At each of 69 locations grouped along a topographic sequence the PAR reflected from the top of the canopy, the vertical profile of transmittance, and the reflectance from the ground, were measured as many as four times in the year, including the extremes of the wet and dry seasons. With these observations an annual balance of the portion of PAR radiation reflected and absorbed by the canopy and ground was assembled and the detailed spatial and temporal dynamics of PAR within canopy layers were estimated. Canopy stature declined along the topographic sequence and the shape of the transmittance profiles reflected this. In locations of declining moisture availability the fraction of PAR absorbed by the ground increased and the fraction absorbed by non-foliar tissues decreased. Seasonal variation in canopy structure was the dominant influence on the partitioning of radiation – spatial variation was less important. Of a total annual PAR input of 15 200 mol m−2, about 95% of incident PAR was absorbed, 50% by leaves, 25% by non-foliar tissues and 20% by the ground. The remaining 5% was reflected by the top of the canopy.

Plant productivity, distribution and diversity in tropical rain forests correlate with water availability. Water availability is determined by rainfall and also by the available water capacity of the soil. However, while rainfall is recognized as important, linkages between plant distribution and differences among soils in available water capacity have not been demonstrated. One reason for this may be that measurements of soil moisture, such as gravimetric water content, may be overly simplistic. To investigate this, we compared two sites in Panama, Allee and Rio Paja, which have similar rainfall but different plant communities. Soil water release curves were obtained from about −0.1 MPa to −9 MPa, permitting us to calculate available water capacity. The Rio Paja site had 17% greater available water capacity (between −0.1 MPa to −3 MPa), whereas the gravimetric water content at Rio Paja was lower by 16% in rainy season and by 41% at the end of the dry season. Hence soil gravimetric water content and soil available water capacity did not correspond. The results suggest that available water capacity may better predict plant distributions. Hence, whenever possible, available water capacity should be determined in addition to gravimetric water content.

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.

To investigate whether seedlings of ten dipterocarp species differed significantly in terms of growth and mortality or whether species were not significantly different and could be considered ecologically similar, seedlings were grown, two per pot, in two experiments: (1) where the two seedlings were of equal height (30 cm); and (2) where one seedling was 10 cm shorter than the other. Seedlings were grown in a shade house with 15% above-canopy light in a 50:50 forest soil–sand mixture and were watered frequently; pots were placed so that seedling density was 130 seedlings m−2 of ground. In the first experiment there were 45 pairwise combinations of species when seedlings were 30 cm tall (AB, AC, AD,…. BC, BD…IJ; where A, B, C…J signify different species); each combination was replicated 10 times so there were 450 pots with 900 seedlings. In the second experiment there were 100 pairwise combinations of species and size e.g. Aa (30 cm A with 20 cm a), Ab (30 cm A with 20 cm b), each combination was replicated 10 times hence there were 1000 pots with 2000 seedlings. After 22 mo 79% of the initial 2900 seedlings survived; on average they had grown 42 cm (i.e. to 72 cm tall from their initial 30 cm). The most frequent outcome of competition-trials between different sized individuals (784 of 1000 trials) was that the initially taller seedling of each pair ‘won’ (it was the taller or surviving seedling). When 900 of these trials (setting aside, Aa, Bb, Cc etc.) were analysed as 45 comparisons between species with different sized individuals (Ab and aB are one interspecific comparison for these purposes), initial height determined the outcome in 23 cases (even in some competitions between light hardwood species and heavy hardwood species); in 6 cases a species (mostly light hardwoods) behaved as a ‘dominant’ – they usually won even if they were smaller initially. We found few significant differences between species in: initial seedling heights; leaf nitrogen concentrations; and specific leaf areas when they were grown in similar conditions, and these attributes were not correlated with growth rates. The similarity of seedlings of different species meant that often a height difference of just 10 cm was enough to determine the outcome of a pairwise competition-trial in high seedling densities and light equivalent to that in forest gaps.

To study vertical gradients of arthropod species richness in tropical forests, adult chrysomelids were surveyed with similar sampling effort by beating in four plots of 0.8 ha, representative of the canopy and understorey of one wet and one dry forest in Panama. Samples included in total 4615 individuals representing 253 species, and were of similar species richness at the two study sites. At both sites, chrysomelids were significantly more species-rich in the canopy than in the understorey. The proportion of species shared between the two study sites was 24%, whereas 16% and 28% of species were shared between the canopy and understorey of the wet and dry sites, respectively. Mature trees supported more and different chrysomelid species than conspecific saplings. A higher proportion of liana feeders vs. tree feeders occurred at the dry site than at the wet site. Multivariate analyses confirmed the faunal differences between the wet and dry sites and that stratification was more marked at the wet site than at the dry site. The latter observation may relate to differences in forest physiognomy (a tall and closed canopy at the wet site) and to the high interconnectivity via lianas between the understorey and canopy at the dry site.

Habitat selection can be detected at different scales. For small mammals, habitat studies normally evaluate micro- or meso-habitat selection, but rarely are multiple scales analysed simultaneously. These studies use quantitative data of structural characteristics around the trap stations to evaluate selection. We test an alternative method to evaluate habitat selection in three didelphid marsupials (Didelphis aurita, Philander frenatus and Metachirus nudicaudatus), using a spool-and-line device, and measure habitat selection at two spatial scales. Habitat was characterized by seven habitat variables measured at ten points of direction change along the path traversed by each individual, and at 25 trap stations distributed in a 1-ha area. Micro-habitat selection was estimated at each point by comparing habitat measures between the direction selected against other directions. Meso-habitat selection was estimated by comparing habitat measurements on the whole path of an individual against habitat measures on the 25 trap stations. Patterns of selection for each species could only be detected at the meso-habitat scale, although a few individuals in each species were selective at the micro-habitat scale. Studies of habitat selection need to address the scale of study quantitatively, if possible comparing two or more scales to determine the scale of selection.

Competition and predation have often been referred to as key forces regulating community structure. However, in the last decade great attention has been paid to the structuring role of positive interactions between plants, which occur when the presence of one species enhances fitness, survival or growth of another (Bruno et al. 2003, Callaway et al. 2002). Positive and negative interactions are unlikely to occur separately in nature (Brooker & Callaghan 1998, Holmgren et al. 1997), and the overall importance of positive interactions on community structure tends to be higher in resource-poor environments (Callaway et al. 2002). Moreover, this balance can be subjected to spatial and temporal shifts (Callaway 1998, Morris & Wood 1989, Tielbörger & Kadmon 1997) linked with plant ontogenetic development and/or changes on resource availability (see Callaway & Walker 1997).

In comparison with the Neotropical big cats, jaguar (Panthera onca L.) and puma (Felis concolor L.), medium and small felids are poorly studied. Furthermore, studying wild felids in forest habitats is extremely difficult using direct methods given that most species are principally nocturnal and secretive (Gittleman 1996). Indirect methods are therefore particularly important, e.g. radio-telemetry (Emmons 1987, 1988; Konecny 1989, Ludlow & Sunquist 1987) or camera trapping (Maffei et al. 2002, Trolle & Kéry 2003). Using systematic camera trap surveys, we compare the population density of ocelots (Felis pardalis L.) across five Bolivian dry-forest sites with different habitat types and/or annual rainfall regimes (Table 1). We hypothesize that ocelot densities will decline as rainfall declines. In addition, we estimate the population of ocelots in the 34 400-km2 Kaa-Iya del Gran Chaco National Park. Finally, we describe and evaluate additional ecological information provided by camera trapping: activity patterns relative to seasonality and moon phase, sex ratios, ranging patterns and relative abundance compared with sympatric felids.