Accurate estimates of forest cover and forest fragmentation are critical for developing countries such as Costa Rica, which holds four to five percent of the world's plant and bird species. We estimated forest cover for Costa Rica using Landsat 5 Thematic Mapper satellite scenes acquired between 1986 and 1991. In 1991, 29 percent (ca 14,000 km²) of the land cover of Costa Rica was closed forest cover; of that forested area, ca 30 percent is protected by national conservation policies. Forest loss in a study area representing ca 50 percent of Costa Rica's territory during a five-year period (1986–1991) was 2250 km², and the estimated deforestation rate was ca 450 km²/yr, or ca 4.2 percent/yr, of remaining forest cover. Forests are almost completely eliminated from the Tropical Moist Forest and Premontane Moist Forest life zones, and the level of fragmentation of remaining forests may be more advanced than previously thought.

To study the structure and composition of old-growth forest in the Saracá-Taquera National Forest near Porto Trombetas, Brazil, we established 36 0.25 ha plots and described the vegetation. We collected charcoal from the A2 soil horizon of each plot for radiocarbon dating. Although fires have been very rare in this forest during historic times, the presence of charcoal in these soils indicates fire at some earlier period. The ages (conventional radiocarbon age adjusted to 1997) of the charcoal ranged from 177 to 1547 years. These ages, however, did not correlate significantly with any of several measures of biodiversity or stand characteristics. The relative uniformity of the current old-growth forest indicates that either the prehistoric fires were of such low intensity that they had little long-term effect on the vegetation or that the present stands have progressed to near steady state.

On 22 September 1998, Hurricane Georges passed over the Dominican Republic causing extensive damage to a 4700 ha mangrove forest that has been the site of a detailed study of vegetation and ecosystem dynamics since 1994. We resurveyed the vegetation in permanent plots at 7 and 18 months after the hurricane to document structural damage of the forest and evaluate early recovery patterns. The intensity of damage was patchy across the landscape. Mortality (≥5 cm DBH) ranged from 14 to 100 percent (by density) among the 23 different plots and averaged 47.7 percent across all plots. Reductions in total basal area ranged from 9 to 100 percent, averaging 42.4 percent. Mortality increased by 9 percent between surveys at 7 and 18 months post-hurricane. Interspecific differences in susceptibility to wind damage appeared to be a primary factor contributing to spatial patterns in mortality. Laguncularia racemosa experienced much less mortality (26%) than either Rhizophora mangle (50%) or Avicennia germinans (64%), and plot-level mortality was strongly associated with differences in species composition. There were no clear relationships between canopy height and tree damage at this site. Over 80 percent of the of the surviving R. mangle trees exhibited less than 50 percent crown damage, whereas ca 60 percent of the L. racemosa survivors suffered almost complete (75–100%) crown loss. By 18 months after the hurricane, the percentage of L. racemosa trees in the 75 to 100 percent damage class was reduced to 20 percent; in contrast, the health of many R. mangle individuals appeared to be declining, as the percentage of trees in the 50 to 100 percent damage class increased from 16 to 36 percent. Understory light levels, as measured by the gap light index, increased from an average value of 3 percent in the pre-hurricane forest to 51 percent at 7 months after the hurricane and decreased slightly to 47 percent at 18 months. Few saplings (>1 m tall and <5 cm DBH) survived the hurricane; 72 percent of the tagged individuals in transect-based plots and 66 percent of saplings in pre-hurricane canopy gaps were killed. Seedling and sapling populations of all three species appear to be recovering rapidly although their densities still are lower than in the pre-hurricane forest. It is too early to predict the trajectory of forest recovery, and continued monitoring of the spatial and temporal patterns of forest development is needed to improve our understanding of the role that large-scale disturbance events play on the dynamics of mangrove forest ecosystems.

In the Ninole ridges of Hawai'i, we evaluated changes in plant species diversity and composition and the effect of alien species on the growth and recruitment of native species after disturbance by landsliding. We chose nine landslides representing three age categories (young, 4–17 yr; intermediate, 18–42 yr; and old, ca 130 yr) plus three undisturbed forest sites (325–525 yr) to sample and manipulate the vegetation; the undisturbed forest developed on tephra-derived soils underlain by basalt. The ordination of sites and species using detrended correspondence analysis (DCA) showed that species establishing on landslides were different than those found in the undisturbed forest; moreover, a large fraction of the species establishing on young landslides were represented by aliens, mostly grasses and orchids. The manipulation of alien species (clipping of aboveground parts and removal of above- and belowground parts, including soil) on young landslides (<17 yr) significantly affected seedling growth of the native tree species Metrosideros polymorpha and the recruitment of native species. In addition, the manipulation of grasses and orchids favored the recruitment of a second group of alien species that were uncommon in, or absent from, our study sites. The removal of tephra-derived soils by landslides, in combination with an increased presence of alien species, may greatly alter successional trajectories in the Ninole ridges of Hawai'i.

Seasonal drought in tropical moist forest may be the cue for fine root death and turnover, and it may signal root growth deeper to access subsurface water and (or) nutrients. We examined these predictions by measuring fine root biomass (<2 mm diam.) and the timing of root growth and disappearance in an old growth tropical moist forest on Barro Colorado Island (BCI), Republic of Panama, in the fifth year of a dry season irrigation experiment. Irrigated soil had greater available P concentrations; however, a more pronounced effect was less stable soil aggregates causing higher bulk density. Irrigation did not affect fine root biomass. Mean (±SE) biomass between 0 and 30 cm was 372 ± 63 g/m² within control versus 286 ± 39 g/m² within irrigated plots. Mean biomass between 45 and 75 cm was 74 ± 7 g/m² within control versus 62 ± 7 g/m² within irrigated plots. Dead roots were less than eight percent of the total. We characterized root growth using in-growth screens (1.7 mm mesh) installed between 0 and 15 cm. Root density in the screens peaked soon after the rains began in the control plots but during the dry season in the irrigated plots. Very few dead roots accumulated in the screens, with no differences seasonally or among treatments. We developed a model to estimate birth and death rates of fine roots using root densities in the in-growth screens and the disappearance of roots laced into screens and incubated in situ. Inferred root birth rates were greatest in the early part of the wet season in control plots and in the dry season in irrigated plots. Inferred mortality rates of fine roots less than four months old were reduced by irrigation, but this was offset by much greater disappearance of six- to eight-month-old roots in the irrigated plots. Although irrigation altered the timing of root growth and mortality, roots did not grow year-round in the always-wet soil. Therefore, soil water availability was an important cue for root growth, but an inherent seasonal cycle remained. Fine roots died continuously throughout the year and disappeared quickly.

Little is known about seed bank dynamics in Neotropical savannas. We studied temporal variation in seed density, species richness, species composition, seed mass, and seed number in the topsoil seed bank of a savanna on “La Iguana” Experimental Station, Guárico State, Venezuela. Total soil surface sampled was 57,600 cm², in 400 cm² units, at four times of the year. Seeds from 124 plants from 20 families were encountered. Species composition varied seasonally, but grass seeds dominated almost all the year, both numerically and in species number. Seed density and species richness were highest in the dry season and declined to a minimum in early wet season. Mean density was 1702 seeds/m², with a range of temporal variation from 235 to 4624 seeds/m². Average seed mass across species by date ranged from 0.97 to 2.53 mg, but because tiny seeds were most numerous throughout the year, the abundance-weighted averages only varied from 0.8 to 1.19 mg. We found a high spatial variability in the seed distribution in the soil throughout the year. There were also substantial changes in the topsoil seed bank among seasons, as a consequence of herbaceous plant phenology and germination strategies, which are tightly coupled with hydroperiodism in Venezuelan savannas. Based on the species-area curves, we concluded that the minimal sampling effort needed to adequately estimate the species richness in the topsoil of this savanna is 11 fold higher than that used in most seed bank studies in tropical environments.

Seed germination and seedling establishment patterns have been used to classify species as shade tolerant or intolerant. The main objective of this research was to investigate, under controlled conditions, seed germination of species from different successional positions as well as to follow seed germination and seedling survival under natural shade in the field. The species studied were Solanum granuloso-leprosum, Trema micrantha, Cecropia pachystachya, Croton piptocalyx, Bauhinia forficata subsp. pruinosa, Senna macranthera, Schizolobium parahyba, Piptadenia gonoacantha, Chorisia speciosa, Pseudobombax grandiflorum, Ficus guaranitica,Esenbeckia leiocarpa, Pachystroma longifolium, Myroxylon peruiferum, and Hymenaea courbaril. Field trials were carried out at Santa Genebra Municipal Reserve, Campinas, SP, Brazil, at the forest edge and in the understory. No significant correlations were detected between successional status and seed size or seed water content. Light-regulated germination was present only in small-seeded species. In field experiments, most species, including the light-sensitive ones, were able to germinate under the canopy, where a low red/far-red ratio predominates. Most species, mainly those of early- and intermediate successional positions, presented low seedling survival rates under shade. Myroxylon peruiferum was the most shade tolerant species, while S. granuloso-leprosum,C. speciosa,P. gonoacantha, F. guaranitica, T. micrantha, and S. parahyba were the most shade intolerant. These latter species showed little or no survival under the shade conditions.

In Gabon, we investigated the floral morphology and biology of Napoleonaea vogelii (Lecythidaceae), a cauliflorous understory tree of lowland humid rain forest, and made observations on the behavior of insect visitors. The major insects visiting flowers were: a thrips (Ceratothripoides brunneus), a small moth (Glyphipterix sp.), four species of leaf beetles (Gabonia colae, Gabonia sp. undescribed, Macrocoma sp., and Monolepta sp.), three species of weevils (Endaeus sp. 1–3), and two species of ants (Technomyrmex sp. and Monomorium sp.). In N. vogelii, the flower is quite complex, and we propose a floral morphological definition for complexity. Insect visitors are spatially segregated on the flower. The evolution of the morphology and biology of the flower of N. vogelii is influenced by insect visitors, both pollinating and non-pollinating. The thrips are the likely pollinators of the flower.

At the Beni Biological Station, Bolivia, we compared the fate of seeds of the palm Attalea phalerata dispersed by tapirs to a latrine with that of non-dispersed seeds. Survival of seeds that had passed through tapir guts did not differ significantly from that of unconsumed seeds, demonstrating that tapirs are legitimate seed dispersers of A. phalerata. Experimental and observational studies showed that tapir dispersal can contribute to escape from seed predation by bruchid beetles such as Pachymerus cardo. Nevertheless, the major contribution of the dispersal process may simply be the mechanical barrier to bruchid attack represented by the fecal material covering seeds. Furthermore, total absence of seedlings and adults from the tapir latrine studied suggests that seasonally flooded tapir latrines are not safe sites for A. phalerata.

I describe the fruit characteristics of species closely associated with black lemur seed dispersal (i.e., species that are often dispersed by the black lemur and only dispersed by the black lemur in the Lokobe Forest). A black lemur group was habituated and observed during the day and night for all months of the year (total 1272 h). When fruits were eaten, the plant species was identified, the maturity of the fruit and treatment of the seeds noted, and the fruit described in terms of ripe fruit color, husk thickness, fruit length, and seed length and width. Black lemur feces were searched for seeds; these were identified and signs of damage noted. Other potential dispersers feeding on the fruits of species eaten by the black lemur were also noted. The black lemurs were seen eating the fruits of 70 species. Of these, 51 species were defined as closely associated with black lemur dispersal and 16 species had seeds that were either often preyed upon or wasted by the black lemur, or were seen being eaten by other potential seed dispersers. Fruits of species in the former group were often dull colored (94% of species); rarely less than 1 cm long (6%); rarely had seeds less than 0.1 cm long or less than 0.1 cm wide (2 and 4%, respectively); never had seeds more than 4 cm long or greater than 2 cm wide; and quite often had either thick husks (49%) or thin husks (51%). In contrast, the fruits of the latter group were often brightly colored (75% of species) and nearly always had a thin husk (94%). Also, this latter group included more small and very large fruits and seeds than the former group. Fruit characteristics significantly associated with the former group were: dull color, thick husk, fruit length greater than 2 cm, seed length 1–4 cm, and seed width 1–2 cm. The extent to which these traits are the result of coevolution between plants and the guild of lemur dispersers that includes the black lemur is not clear, but a coevolved lemur–fruit syndrome remains a possibility.

Mistletoes are hemiparasitic plants that frequently depend on frugivores for seed dispersal; yet their seed dispersal ecology is still poorly known. The mistletoe Psittacanthus schiedeanus (Loranthaceae) was studied for a five-month period in cloud forest remnants in central Veracruz, Mexico, to determine its seed dispersal ecology. We hypothesized that dispersal ecology of this mistletoe species is directly affected by the fruit-eating bird abundance, and asked if this species is following the high-investment strategy as suggested by Godschalk (1983). To evaluate this, we determined the morphological and nutritional characteristics of mistletoe fruit, the availability of ripe fruits along the mistletoe fruiting phenology, and the abundance of fruit-eating birds. Because P. schiedeanus is dispersed by generalist rather than specialist birds, we asked if this mistletoe species is ripening fruits during the fruit scarcity time (with highest possibilities of being consumed and dispersed by birds). We compared P. schiedeanus ripe fruit production to the cloud forest tree leaf fall and fruit production. Psittacanthus schiedeanus matures its nutritious fruits from November to April, with a peak of abundance in January and February. We found a synchrony between the abundance of P. schiedeanus ripe fruits and the abundance of Gray Silky-flycatchers (Ptilogonys cinereus) and Social Flycatchers (Myiozetetes similis); however, this was not the case with Cedar Waxwings (Bombycilla cedrorum), although these birds play an important role in the mistletoe dispersal. The fruiting pattern of P. schiedeanus, together with the fact that it is an important food resource for generalist rather than specialist frugivorous birds, suggest that this mistletoe is using the nutrient investment strategy to attract dispersers during a time of fruit scarcity.

Parasitoid insects face considerable trade-offs in locating suitable hosts within complex environments. Apocephalus paraponerae (Diptera: Phoridae) locates its host ant Paraponera clavata (Formicidae: Ponerinae) using olfactory cues. Here, comparing two populations of A. paraponerae, I describe differences in host location cues between two sites, Barro Colorado Island (BCI) in Panama and La Selva Biological Research Station in Costa Rica. At La Selva, A. paraponerae uses the ant mandibular gland products 4-methyl-3-heptanone and 4- methyl-3-heptanol in host location, but does not do so on BCI. I propose that higher colony density of P. clavata causes A. paraponerae to use more species-specific cues on BCI. I also discuss how geographic variation in host location cues could lead to allopatric speciation.

I compared species richness and habitat correlates of leaf-litter herpetofaunal abundance in undisturbed and selectively logged forests, and an abandoned pine plantation in Kibale National Park, Uganda. I sampled 50 randomly located 25 m² litter plots in each area during the wet and dry seasons in 1997. Ten anuran, five lizard, and three snake species were captured in plots over the study. Assemblage composition was most similar at logged and unlogged sites. The logged forest herpetofauna had higher species richness and abundance than the unlogged forest, but diversity was greater in the unlogged forest due to greater evenness. In contrast, the pine plantation site had the highest richness, abundance, and evenness of the three study sites, but species composition was distinct from the other areas. Herpetofaunal densities were significantly lower in all three areas during the dry season than in the wet season. During the dry season, soil moisture, litter mass, topography, shrub cover, and number of fallen logs were significant positive predictors of herpetofaunal presence in litter plots, but only soil moisture was significant in the wet season. The interaction of moisture and topography appears to be important in determining seasonal patterns of litter herpetofaunal distribution. Comparison of litter herpetofaunal studies across the tropics have shown that mid-elevation faunas generally support fewer species than lowland faunas. Compared with other tropical mid-elevation litter faunas, Kibale supports an intermediate number of species, but at lower densities than observed at any other mid-elevation site reported in the literature.

I studied the relationship between diurnal raptor diversity, density, and richness, and landscape heterogeneity in continuous primary forests and forests farmed by native Amerindians in the Río Plátano Biosphere Reserve of northeastern Honduras from January to June 1996 and 1997. I estimated landscape heterogeneity—the variability in naturally occurring and/or anthropogenic habitat mosaics—by mapping the extent and distribution of five human-modified and natural habitats in 24 1 km² survey plots. I used the Shannon index to calculate landscape heterogeneity values for the respective plots based on the proportion of total area of each habitat within each plot. Diurnal raptor surveys from canopy-emergent viewpoints in these plots resulted in 137 observations of 18 species of raptors. Four species (Coragyps atratus, Ictinia plumbea, Leucopternis albicollis, and Buteo magnirostris) differed significantly in abundance among heterogeneity classes. Raptor diversity, density, and richness all increased directly with increasing landscape heterogeneity. Landscape heterogeneity was more important in explaining differences in raptor species diversity than the presence or extent of any single habitat or combination of habitats. In contrast to previous studies, my results indicate the importance of indigenous shifting cultivation in altering the naturally occurring patterns of habitat mosaics in lowland rain forest and its effect on bird species abundance and diversity in a rain forest ecosystem.

We estimated the relative contribution of fruits and insects as sources of dietary protein in two species of Neotropical frugivorous bats (Artibeus jamaicensis and Sturnira lilium) using stable carbon and nitrogen isotope analyses. An insectivorous species (Pteronotus parnellii) was also included for comparison. We found constant patterns in stable carbon and nitrogen isotope composition in blood that separated the two species of frugivorous bats from the insectivorous bat. When we used these isotopic values (combined with those of dietary fruits and insects) to estimate the percent contribution of fruits and insects to the diet of the bats, we obtained different results, depending on assumptions and model adopted. We tested models using both δ¹⁵N and δ¹³C results simultaneously and separately and further used diet-tissue fractionation factors of 3‰ for nitrogen and 1 and 3.5‰ for carbon. We found that a carbon-based model with a diet-blood enrichment factor of 3.5‰ produced the most parsimonious results. The model estimated that A. jamaicensis and S. lilium obtained most of their protein requirements from fruits, whereas P. parnellii fed mostly on insects. No sexual or seasonal variations in the diet of the two frugivorous species were detected. We found no evidence that the diet of sexually active females differed from that of nonsexually active females in the two species of frugivorous bats. We suggest that future studies better define isotopic fractionation between diet and tissues of bats using captive rearing and controlled diets.