Printer Friendly

FOREST BIRD COMMUNITY STRUCTURE IN CENTRAL PANAMA: INFLUENCE OF SPATIAL SCALE AND BIOGEOGRAPHY.

W. DOUGLAS ROBINSON [1,3]

JEFFREY D. BRAWN [2]

SCOTT K. ROBINSON [2]

Abstract. Historical and biogeographic contexts can play important, yet sometimes overlooked, roles in determining structure of local communities. In particular, few examinations of historical influences on patterns of species richness and relative abundances in tropical communities have been conducted. In part, that gap in our knowledge has been caused by a paucity of data on tropical communities, even for relatively well-studied taxa such as birds. In the Neotropics, only two sites, a 97-ha plot in lowland Peru and a 100-ha plot in French Guiana, have been inventoried on a spatial scale sufficient to estimate population densities for a majority of resident bird species. Results from those studies revealed extremely similar species richness, community biomass, and patterns of relative abundance. A third site in lowland Panama was originally censused in 1968-1969 and has often been compared with many other tropical and temperate sites. Results from Panama suggested an exceptionally different community struct ure from that observed at the Amazonian sites. Informative comparisons among sites have been hampered, however, by differences in sampling protocols. The Panama site was sampled on a much smaller spatial scale (2 ha) than the two Amazonian sites. To improve comparisons, we censused a 104-ha area (the Limbo plot) encompassing the original 2-ha Panama study area and used several census methods, including those used at the Amazonian sites.

As expected, spatial scale had a strong effect on estimates of species richness. We detected 252 species on the Limbo plot, compared with 161 detected on the original 2-ha area. Estimates of total individual birds per 100 ha were similar, but estimates from the original study were based on densities measured for one-third fewer species than we measured on our larger study area. Of the 53 species for which both Panama studies estimated population densities, a significant number of estimates were higher in the original study. Thus, the small spatial scale of the original study apparently led to inflated density estimates. The primary cause of disparities appeared to result from undersampling in the smaller plot of many species with patchy distributions and large territory sizes.

Compared with Amazonian communities, the Panama community had far fewer rare species. Although 33% of species in Amazonian sites had densities of [leq]1 pair/100 ha, only 17% were equally rare in Panama. Furthermore, eight species in Panama were, by tropical standards, "superabundant," attaining densities as high as 212 breeding individuals/l00 ha; the most abundant species in Amazonia barely reached one-third of that number. In total, those eight species accounted for 36% of all individuals at Limbo. The median abundance at Limbo was 7 pairs/l00 ha, vs. 2.5 pairs/100 ha in Amazonia. Consequently, the total number of birds on the Limbo study area was nearly twice that found in Amazonia, despite species richness being only three-fourths as great.

We conclude, first, that spatial scale has indeed had an important effect on the characterization of the Panama bird community. The intrinsically patchy distributions of most forest-dwelling bird species raise the need for large-scale censuses. Second, the Panama community, compared with the two Amazonian sites, has a fundamentally different organization; it hosts nearly twice as many individual birds and is distinctly less dominated by rarity. Similar patterns of community structure appear to be present within tree and mammal communities as well. Therefore, results from the Amazonian studies cannot be generalized to all lowland Neotropical communities. We attribute differences in community structure primarily to differing biogeographic histories. The lower species richness and the greater number of total birds present in Panama appear to derive, at least in part, from two important factors: an area effect linked to the location of Panama on a narrow isthmus, and the repeated history of disturbance on multip le temporal scales in Panama.

Key words: Amazonia; biogeographic history; biomass; birds; community structure; neotropical forest; Panama; spatial scale; species diversity; species richness.

INTRODUCTION

Tropical communities are often broadly categorized as being similar to one another because of high diversity and large numbers of rare species (Wallace 1878, MacArthur et al. 1966, Ricklefs 1990). Recently, however, a growing appreciation for the highly patchy nature of organismal distributions in the tropics has suggested that nearby local communities may differ considerably (Diamond 1980, Campbell 1994, Tuomisto et al. 1995). Small changes in topography, soil characteristics, and productivity appear to be correlated with different assemblages of species (Emmons 1984, Gentry 1990, Terborgh et al. 1996). In addition, a dichotomous view of ecological and evolutionary forces responsible for community structure has arisen among investigators of tropical communities where identities of species co-occurring in a community are thought to be either determined by random events, or to be structured by more deterministic interactions among species (e.g., Hubbell 1979, Hubbell and Foster 1986, Robinson and Terborgh 199 5, Terborgh et al. 1996).

Increasingly, ecologists are adopting a balanced view that both stochastic and deterministic processes are important in determining community structure (Ricklefs 1987). In particular, investigations of the historical and biogeographic contexts in which communities lie are especially of interest (Ricklefs and Schluter 1993). However, the general rarity of detailed data on tropical communities has slowed progress in addressing the importance of historical factors influencing community structure. A comparative approach in which communities at multiple sites with differing biogeographic contexts are examined is likely to provide the perspective necessary to identify important biogeographic differences (Schluter and Ricklefs 1993). Few detailed investigations of tropical community organization at multiple sites have been conducted. Here, we evaluate differences in avian community organization between two Amazonian sites and a site in lowland Panama. To improve comparability of the data, we first examine the influ ence of spatial scale on interpretation of community characteristics at the Panama site, because spatial scale can have a pervasive effect on our ability to identify and interpret patterns in community ecology (Wiens 1989, Levin 1992, May 1994). We then investigate historical and biogeographical differences among the communities that may contribute to observed differences in community organization.

TROPICAL BIRD STUDIES

Although birds are perhaps the best known organismal group in the tropics (e.g., Stotz et al. 1996), the difficulty of identifying them has limited the effectiveness of community investigations. Only recently have field guides and catalogs of voice recordings of tropical birds become available for many areas. Previously, because of unfamiliarity with vocalizations in species-rich sites, many authors relied heavily on the use of mist nets to sample birds (Karr 1971, 1980, Terborgh 1971, Bierregaard 1990). Few efforts were made to estimate abundances of the large proportion of species occupying forest strata undersampled by ground-level nets. Recent studies, however, have demonstrated that it is possible to estimate abundances of most species in a community when multiple census methods are used (Terborgh et al. 1990, Thiollay 1994).

Debate continues over basic ecological issues such as population densities, community biomass distribution, territory sizes, and patterns of seasonal movements in tropical bird communities (Karr 1971, Karr and Freemark 1983, Terborgh et al. 1990). The debate over these issues arises primarily from disagreements in results from studies of perhaps the two best known neotropical bird communities: the Cocha Cashu site in Manu National Park in southeastern Peru (Terborgh et al. 1984, 1990, Robinson et al. 1990, 1995) and the Limbo site in Soberania National Park in central Panama (Karr 1971, 1990, Karr et al. 1990, Brawn et al. 1995). In a 97-ha area of Peruvian floodplain forest, Terborgh et al. (1990) used an array of different census methods, but primarily relied on spot mapping of vocalizing birds. They found 245 resident species, comprising a total of 955 pairs/100 ha and representing nearly 200 kg/100 ha of biomass. Many species were rare, being present in densities of [leq]1 pair/100 ha, and average home r ange size was [sim]4-5 ha. In contrast, Karr (1971) relied more heavily on mist netting and following individually color-marked birds on his 2-ha study plot in lowland central Panama. There, he found that 140 resident species were present at a combined population density of 1800 pairs/l00 ha, nearly twice the density found at Cocha Cashu. Yet, estimated biomass of birds was nearly 50% less than in Peru (131 kg/100 ha). That curious difference led Terborgh et al. (1990) to challenge the methods of Karr. They argued, based on evaluations of their data from Peru, that a 2-ha plot is of insufficient size to sample adequately a tropical bird community, primarily because average home range size is larger than the study area; that discrepancy would make estimates of population densities very difficult because, on average, the small plot would not contain a single entire territory of most species.

Furthermore, results from a recent study on a 100-ha study area at Nouragues, French Guiana, were almost identical to those from Cocha Cashu (Thiollay 1994). Despite the fact that the French Guianan study area was located on the upland Guianan plateau, whereas the Peruvian study was located several thousand kilometers away in the floodplain habitat of the Amazon basin, measures of species richness, community biomass, and relative abundance patterns were extremely similar, suggesting convergence in community characteristics. The similarity of the Amazonian studies then provides additional reason to question why results from Panama are so different. Is organization of the Panama community fundamentally different and, if so, why? Or, are results from the Panama plot simply affected by the small spatial scale of the original study?

Efforts to make a critical comparison of the data from Panama and the Amazonian sites have proven problematic. Not only do plot size differences complicate comparisons, but also a lack of standardized methodology for measuring bird populations limits the effectiveness of comparisons. Even in temperate areas, where species richness is generally lower and population densities tend to be much higher, methodological differences can make comparisons among studies problematic. Tropical birds, however, present a much more diverse array of social systems, frequency (or infrequency) of vocalizations, and degrees of detectability (Karr 1971, 1981). Consequently, no single method will provide accurate measurements of population densities for most species in a community (Karr 1971, Terborgh et al. 1990, Remsen and Good 1996). Thus, the need to use a wide array of techniques arises.

Here, we present the first comparison of bird communities from Panama and South America using similar sampling methods and complete community censuses. We established at the Panama site a 104-ha plot, which included the 2 ha originally censused by Karr (1971). Furthermore, we used the same suite of methods used by Terborgh et al. (1990) for estimating population densities. Previous attempts to compare these and other communities resorted to mist-net samples in an effort to standardize the data being compared (Karr et al. 1990). The sacrifice made, of course, was that only a subsample of species from each community could be compared. Perhaps as a result, the mist-net samples showed differences in species richness, but few other important differences (Karr et al. 1990). We show, however, based on multiple methods and data from all species in the community, that the Panama community differs from the Amazonian ones in many respects, including lower species richness, greater abundance of many species, and fewer t ruly rare species; we attribute many of those differences to the influence of biogeographic history and history of disturbance at the Panama site. We also demonstrate that spatial scale has indeed affected estimates of species richness and population density at the Panama site, and that large study areas are necessary to adequately sample the intrinsically patchy distributions of most tropical species.

STUDY AREA

We censused birds in a 104-ha study area (the Limbo plot) of lowland forest in Soberania National Park in the Republic of Panama (9[degrees]9'35" N, 79[degrees]44'36" W; Fig. 1). We placed the study area to include the 2-ha plot originally investigated in 1968-1969 (Karr 1971). Limbo lies in a relatively flat, circular basin of [sim]600 ha, surrounded on all sides by steep terrain deeply dissected by ravines. Elevation within the plot varies from 35 to 80 m above sea level, but it reaches 225 m immediately to the north and east. Limbo is [geq]3.5 km from the nearest nonforest edge, and thus contains a bird community composed almost exclusively of forest-interior species.

The vegetation is classifed as tropical moist forest (Holdridge 1967). It receives [sim]2600 mm of rain annually, with 90% falling during the late April to mid-December wet season (Windsor 1990). Temperatures during the wet season vary from mean daily lows of 23[degrees]C to highs of 29[degrees]C. The dry season is characterized by little rain and strong northeasterly trade winds, with mean daily low temperatures of 23[degrees]C and highs of 32[degrees]C. Although some tree species drop their leaves during the dry season, the canopy remains mostly closed throughout the year. Frequent treefalls, however, have created many openings that allow light penetration to the forest floor; gaps range in size from a few to several thousand square meters. A 2.5-ha gap in the north-central portion of the study area was created in the 1992 wet season when a tornado-like storm raged through the area (Fig. 2). Partly as a function of the frequent treefall gaps and blowdowns, the forest is a heterogeneous mixture of patches o f trees of various ages. In general, age varies from a few years old, in the case of the large blowdown, to nearly 400 yr old in remnant patches of tall forest in the north and northwest sections of the plot (Karr 1971). The forest within 10-30 m of the single-lane, gravel Pipeline Road, is [sim]30 yr old, and has now grown tall enough that it has virtually closed the canopy above the stretch of road through the study area. South of the road, the forest is 60-120 yr old, apparently having been disturbed by farmers, as well as the United States military, earlier in the century (e.g., Foster and Brokaw 1982).

The most abundant tree species in the upper understory and canopy, based on identification of all stems in four 1-ha samples on the study area (R. Condit, unpublished data), are Oenocarpus mapora, Welfia georgii, Tapirira guianensis, Socratea exorrhiza, Poulsenia armata, Virola sebifera, Lindackeria laurina, and Terminalia amazonia. The lower understory is dominated by Oxandra longipetala, Quassia ainara, Sorocea affinis, Alseis blackiana, Cupania sylvatica, Protium panamense, Poulsenia armata, and Oenocarpus mapora. In the older forest, canopy height averages 30-40 m; in the younger forest along the road, it averages [sim]10 m less (W. D. Robinson, unpublished data). Ground cover tends to be minimal except near treefall gaps, where tree seedlings and lianas frequently form impenetrable masses of vegetation.

Numerous small, gravel-bottomed streams flow southward through the area. Many are fed primarily by run-off from rainfall, and thus have little flow during the depth of the dry season. Larger streams, however, retain a moderate flow throughout the year. Rio Limbo, the largest stream, averages 7 m in width, and is not large enough to attract some bird species, such as Basileuterus fulvicauda and Chloroceryle americana, that occur along larger watercourses elsewhere in the park.

METHODS

Counting methods

To generate population density estimates for as many species as possible, we needed to use several methods, but our primary method was spot mapping (Kendeigh 1944, Terborgh et al. 1990). During our surveys, we conducted the majority of censuses from just prior to dawn until about 1200. That time period encompassed the principal period of song activity for most species, as well as the midmorning period when many raptors and other aerial species are particularly active and conspicuous. We also conducted some censuses in the two hours centered at dusk, when several species of wood-creepers and antbirds sing as they head to their roosts, and when nightjars and owls are most vocal. Unlike some Amazonian sites (e.g., Gocha Cashu; Terborgh et al, 1990), however, the dusk chorus was not very active, so we concentrated the bulk of our efforts at dawn. More than 90% of the censuses were conducted by W. D. Robinson. We censused on a weekly basis from January to July, 1994 and 1995, which encompassed the majority of the breeding season for most species, and biweekly from August through December 1995, when song activity levels were generally substantially lower (W. D. Robinson, unpublished data).

We spot-mapped locations of all individuals seen or heard in and near the Limbo plot (see Figs. 3, 4). To provide even coverage of the area, we walked parallel transects spaced 100 m apart. We estimated the distance and direction from permanently marked coordinates (at 25-m intervals) of all birds detected. Particular emphasis was placed on mapping locations of counter-singing individuals, which facilitated estimates of population densities, especially for the most common species. Because both sexes of many tropical species sing regularly, we were careful to map pairs and singing males whenever possible. For many species (especially antbirds), female song is distinguishable in the field from male song and both members of a pair typically forage together (Skutch 1996). In species in which both sexes sing similarly and forage separately, such as several woodcreepers (e.g., Pierpont 1986), we used mapped observations of color-marked birds to improve interpretation of our data.

We also incorporated point counts into our censusing methods. Although used frequently in studies of temperate bird communities (e.g., Verner 1985), point counts have been little-used in the tropics until recently (Blake 1992, Lynch 1995). Point counting provides two important advantages over spot mapping: it forces even coverage of the entire study area, and it provides a methodological framework that can be exactly repeated by future investigators. We conducted 8-mm, unlimited radius counts at each of 126 points spread uniformly in a 100-m grid across the study area. During a visit to a point, we recorded direction and estimated distance ([pm]10 m) away from the point for each bird detected aurally or visually. We made a total of eight visits to each point, four in the dry season, and four in the early wet season (May-July). We used a 100-m grid because preliminary efforts in the study area convinced us that voices of some species would not carry [greater than]50 m through the liana-crowded forest. Consequ ently, voices of some louder species could be heard from more than one point. We found that to be an advantage, however, rather than a drawback, because we were then able to use mapped observations from point counts to supplement spot-mapping data. We do not present detailed analyses of point counts here, however, but simply use mapped observations from those counts to supplement data from spot mapping.

Another important component of the spot mapping was observations of color-marked birds. We captured understory-dwelling birds with 12-m ATX (36-mm mesh) mist nets at nearly 150 sites in the study area (Fig. 2). Each bird was given a numbered aluminum band and a unique combination of colored leg bands to facilitate mapping locations of individual sightings. Combining those data with the spot-map data, we were able to generate "territory" maps for a majority of the resident species, as well as for two of the wintering migrant species.

Spot mapping is ineffective for some species, however, especially many highly mobile frugivores and nectarivores (Karr 1981, Terborgh et al. 1990). In those cases, the netting efforts were again useful. From the capture data, we generated estimates of the numbers of understory hummingbirds, manakins, Mionectes fly-catchers, and other highly mobile and wide-ranging species, following methods outlined in Terborgh et al. (1990). Recent computer simulations have indicated that mist-netting is not a particularly reliable method for estimating abundances because minor differences in vegetation structure and flight behavior may influence capture rates (Remsen and Good 1996). Despite such concerns, we feel that it is still valuable to compare results from the studies here (see also Methods: Cautionary note), especially considering that no workable alternative has yet been proposed for estimating abundances of those common and important members of tropical bird communites. For several lekking species (see Fig. 3), we corroborated estimates derived from netting by counting the number of birds displaying at and visiting leks.

To improve detection of some quieter species such as Oncostoma olivaceum, Platyrinchus coronatus, and Terenotriccus erythrurus, we also walked transects between the 100-m grid trails. The voices of these species do not carry more than [sim]30 m in this forest (W. D. Robinson, personal observation), so the additional transect walks were useful for filling in gaps between primary routes.

Some species could not be censused well with spot mapping or netting because they either ranged in mono-or multispecific groups over large areas of the plot, or they were relatively nonvocal and spent most of their time in the canopy. In the case of the colonial icterids, Cacicus cela and Psarocolius wagleri, we simply counted the number of active nests at the only colony on the plot, assumed one female per nest, and added the number of males visiting and courting females at the colony. Then, because these two species range widely during foraging, over an area of [sim]400 ha (S. K. Robinson, unpublished data), we divided the number by four to estimate the number per 100 ha. The more solitary icterid, Cacicus uropygialis, frequently foraged in pairs or small flocks with groups of the cotinga, Querula purpurata; these two species had well-defined areas of high use within their home ranges, presumably around nest sites, which facilitated estimates.

For other canopy group-living species, such as toucans and parrots, we combined our spot-mapping data of the foraging and calling locations of groups with counts of group size to derive minimum density estimates. The two Ramphastos toucans frequently sang from exposed snags reaching above the canopy and foraged in large fruiting trees where they could be counted visually. Especially in early morning and at dusk, groups were more cohesive and could be counted more easily than during the remainder of the day, when they often dispersed and foraged as subgroups. Pteroglossus torquatus formed bigger groups, on average, than the larger-bodied toucans, and concentrated activity around their canopy nest sites. Parrots roosted in large congregations some distance outside the plot, and then flew into the area to forage during the day. To estimate numbers of parrots on the plot, we used a method similar to that employed by Terborgh et al. (1990). We counted the number of individuals in each flock during their early mor ning flights and calculated an average number of birds in groups of each species. We then spot-mapped foraging locations of each species and calculated an average number of groups foraging on the plot for monthly time intervals. By multiplying the number of groups derived from the spot-mapping information by average group size, we estimated the number of individuals using the study area.

Unlike the Cocha Cashu site, where Munn's (1985) intensive studies of canopy flocks allowed estimates of population densities of flocking species, we had few color-marked canopy birds. Our site differed by having very poorly developed canopy flocks. Typically, insectivorous flocks were composed of only a few, highly vocal species (e.g., Vireolanius pulchellus and Hylophilus decurtatus) that could be detected easily and mapped. Frugivore flocks, however, were less vocal and distinctly less well developed than those at many other tropical sites. An average canopy tanager flock was composed of 2-4 individuals of Dacnis cayana and 1-3 individuals each of 1-3 species of honeycreepers or euphonias. We counted canopy flock members by observing them at flowering and fruiting trees near treefall gaps and along transects.

Canopy-dwelling hummingbirds were difficult to census. They were infrequently vocal and highly transient, being conspicuously present only during brief flowering periods of some canopy tree species. We inventoried species at flowering trees. Estimates of abundance were made for only a subset of species, and we view those numbers as crude approximations at best.

For species with home range sizes [greater than]100 ha, such as raptors, we estimated population densities for some species, based on our experiences in the areas surrounding the plot. We periodically counted birds in a 10-[km.sup.2] area around Limbo, and used our knowledge of the distribution of raptors in that area to derive the Limbo estimates.

Cautionary note

As discussed at length by Terborgh et al. (1990), the great variety of social systems of tropical birds requires a variety of methods for estimation of population densities (see also Karr 1981). Because a primary focus of our study was a comparison with the results of Terborgh et al. (1990), we used the same methods to derive estimates of population density. We emphasize, however, that the relationship between the true population density of a species and estimates derived from some methods used here remains unclear. Spot mapping and observations of color-marked birds appear to work well for estimating abundances of species that are strongly territorial. However, further efforts to develop methods for determining population densities of tropical birds, especially nonterritorial species such as manakins and hummingbirds, are needed. Currently, little is known about the heterogeneity of detection probabilities arising from a wide array of sources among tropical birds, including variation among species, habitats , and observers (Beehler et al. 1995). Advances in methodology will certainly improve the precision with which both ecological and theoretical issues can be addressed by data from tropical communities. At present, we restrict our comparisons of the data on population densities to broad analyses of general patterns.

Species status and presentation of population density estimates

We classified species as residents, migrants, or vagrants. Residents were forest-dwelling species that bred on the study area or in nearby forested areas in central Panama. Migrants were transient species that were present for only part of the year; that category includes migratory species that breed in North America and spend the north temperate winter at Limbo (e.g., Oporornis formosus), as well as intratropical migrant species that breed at Limbo and winter elsewhere (e.g., Legatus leucophaius). Vagrant species were those that occurred infrequently at Limbo (fewer than four records), and that were not expected to occur regularly because of a lack of appropriate habitat on the study area (e.g., Arremenops conirostris).

We report average year-round breeding density estimates for resident species. For the few wintering neotropical migrants for which we could estimate abundances, we used only data from November through March, when little or no migration is occurring. Although we censused during two years, abundances of nearly all species showed remarkably little change between years, which is generally consistent with previous results (e.g., Greenberg and Gradwohl 1986). Important exceptions were the colonial icterids and bamboo specialists. In 1994, the icterid colony hosted [greater than]85 total nests. In 1995, however, the colony declined to [less than]10 oropendola nests. Similarly, unusually large numbers of the seedeater Sporophila schistacea and the dove Claravis pretiosa were attracted to seeding bamboo in 1994, but both species were rare in 1995 when markedly less bamboo was seeding. The 1995 estimates represent more "normal" densities for those four species, so we use the 1995 estimates in our calculations of commu nity parameters. We will address the details of annual population variation elsewhere.

Throughout the paper, most of our comparisons are with the Cocha Cashu site in Peru (Terborgh et al. 1990). Cocha Cashu is located in primary, floodplain forest, whereas the other comparably censused neotropical site is in primary, terra firme forest, in Nouragues, French Guiana (Thiollay 1994). The latter study, therefore, could be considered to offer a more informative comparison because the Panama site is also located in terra firme forest. Characteristics of the bird communities at the Peruvian and French Guianan sites were so similar, however, that we are effectively comparing our results with both studies. We mention the Peruvian results most often, however, because that study made explicit reference to and comparisons with the original small-scale census from Panama, whereas the French Guianan study did not.

Territory sizes

We estimated territory (or home range) sizes by plotting sightings of color-marked birds onto a map of the study area. We also used clusters of registrations of unmarked birds and countersinging records to provide measures for some additional species. Our primary goal was to facilitate comparisons of the distribution of territory sizes among species at Limbo with those at Cocha Cashu. Because low sample sizes of observations can cause an underestimation of territory size (Stickel 1954), we made measurements on a continuous scale only for species for which we had [greater than] 15 registrations per territory, which was the average number of registrations accumulated per territory at Cocha Cashu (Terborgh et al. 1990). For the remaining species at Limbo, we used a conservative approach and estimated territory size on a categorical scale with intervals of [log.sub.2] width. Consequently, we measured territory size on a categorical scale for some species and a continuous scale for others. Measurements could not be made for all species for a variety of reasons; for example, some species were nomadic and did not regularly occupy the same area (e.g., manakins), and some species foraged in social groups that interacted with other groups and roamed over large areas (e.g., fruitcrows). We estimated territory sizes for a total of 87 species.

Data analyses

We compared territory sizes at Limbo with those at Cocha Cashu (where measures for 111 species were made; Terborgh et al. 1990). The measurements from Limbo were primarily categorical (see Methods: Territory sizes). Therefore, we used contingency analyses ([[chi].sup.2]) to make most statistical comparisons. In cases in which values in contingency tables were very sparse and too many expected values were [less than] 5, we refrained from using asymptotic P values and instead derived exact probabilities (StatXact 1995). For paired comparisons of the territory size of the same species at both sites (measured on a continuous scale), we used a paired t test.

To compare rank--abundance curves, we first converted the data to cumulative frequency distributions with abundance measured on the abscissa in intervals of 1 individual/100 ha (except for two initial intervals for densities of [leq]0.5 and 0.51-1.0 individual/100 ha and three wider intervals for common species: 36-45/100 ha, 46-76/100 ha, and [greater than] 76 individuals/100 ha). Cumulative frequency was measured on the ordinate. We then evaluated with a Kolmogorov-Smirnov test (Siegel and Castellan 1988: 144) the null hypothesis that a pair of distributions (e.g., Cocha Cashu vs. Limbo) were drawn from the same population.

When comparing the population density estimates from the original 2-ha census of Limbo with estimates from our 104-ha census, we used a Wilcoxon signed-ranks test (corrected for continuity), which, at large samples, is closely approximated by the normal (z) distribution (Siegel and Castellan 1988:87). All tests were two-tailed, with P [leq] 0.05 considered significant.

RESULTS

We generated [greater than]27 000 mapped observations of birds on the Limbo plot. Those data comprise the largest set of census results gathered for a single neotropical bird community. In addition, we accumulated 2900 mist-net captures of 2200 individual birds. Combined with the many avifaunal studies conducted by J. R. Karr, the mist-netting and census data make Limbo the most thoroughly inventoried neotropical bird community to date. In Table 5, we present the population density estimates, masses, status assignments, biomass densities, territory sizes, guild classifications, and measures of patchiness of spatial distribution for the species in the Limbo community.

Species richness

Of the 252 species we detected in the two years of the study, 165 species were present in densities of [geq]0.5 pairs/l00 ha. The remaining 87 species were either vagrants (23 species), nonwintering migrants (35 species), or were too mobile (e.g., aerial insectivores) or rare (29 species) for us to estimate densities. We classified 181 species, or 72% of the total, as residents capable of breeding on or near the Limbo forest plot (Table 1). In addition, 14 species of migrant were resident on the plot for part of the year. One species of intratropical migrant, Legatus leucophaius, bred in association with a large colony of icterines on the study area. Three species of altitudinal migrants (Phaethornis guy, Eutoxeres aquila, and Turdus albicollis), were detected, although only in very small numbers. Altogether, migrants accounted for 19% of the species total. Vagrants, most of which were species from aquatic or edge habitats, represented 9% of all species detected.

Species--abundance relationships

Distribution of species' abundances at Limbo was significantly different from distributions at Cocha Cashu and Nouragues (Fig. 5). Although rank--abundance curves for the Amazonian sites were not significantly different from one another (Kolmogorov-Smirnov test, P = 0.76), both differed dramatically from the Limbo curve (Kolmogorov-Smirnov test, P [less than] 0.001). Most of the Limbo curve lies above the Amazonian curves, indicating: (1) larger differences among species in abundances, and (2) a community composed to a greater extent of common species rather than rare ones. In the Amazonian communities, most species are relatively equally abundant, which accounts for the flat, long tails of the rank-abundance curves.

An oligarchy of eight common species, all with abundances of [greater than]95 individuals/100 ha, accounted for 36% of the individual birds at Limbo. At the opposite extreme, only 26 (17%) of the 152 resident species for which we measured abundances were present at densities of [leq]1 pair/100 ha. In contrast, 80-85 (33%) of the species in the two Amazonian communities were that rare, and the most abundant species there barely exceeded 50 individuals/100 ha. Among territorial species, 106 territories of the most abundant species at Limbo (Thamnoplillus atrinucha) occurred on the study area, whereas only 20 territories of the most abundant species at Cocha Cashu (Myrmoborus myotherinus) were present. Furthermore, the median population density was [sim]7 pairs/100 ha at Limbo, nearly three times the median of 2.5 pairs/100 ha at Cocha Cashu and Nouragues. Overall, the total of 3234 individuals/100 ha at Limbo was nearly twice as high as the 1700 individuals/100 ha at the Amazonian sites.

The Limbo oligarchy was composed of those species with abundances ranging from a conservative estimate of 96 individuals/100 ha in the case of the most common understory frugivore, Pipra mentalis, to 120-212 breeding individuals/100 ha in the cases of several understory mixed-species flock members. Two antwren species, Mymotherula fulviventris and Microrhopias quixensis, formed the core of the understory flocks and occupied territories of [sim]1 ha (see also Greenberg and Gradwohl 1985, 1986); they were joined by two other species, Thamnophilus atrinucha and Myrmotherula axillaris, that often moved between more than one flock. Three of those flocking species were present at nearly equivalent densities. Other members of the oligarchy were a 6-g hover-gleaning insectivore, Oncostoma olivaceum, which frequented the abundant midstory vine tangles; the putative core species of canopy insectivore flocks, the vireo Hylophilus decurtatus; and the omnivorous, but apparently primarily frugivorous, treetop flycatcher Z immerius vilissimus (Wetmore et al. 1984).

Territory sizes

The observation that most species at Cocha Cashu occupied large territories was one of the primary reasons that Terborgh et al. (1990) expressed concern that census efforts on small plots could produce inaccurate population density estimates of tropical forest birds. They argued that if most species occupied areas larger than the study plot, fewer than one pair of each species, on average, would be present on a given area, potentially leading to overestimation of abundances.

If, however, territory sizes were smaller in Panama than in Peru, which preliminary analyses indicated was the case (Karr et al. 1990), risk of such a sampling problem might be minimized. We therefore compared territory sizes between the two sites to determine whether territory sizes were consistently smaller at Limbo than at Cocha Cashu.

No species had a territory size [less than]3 ha at Cocha Cashu, whereas 15 species did at Limbo (Fig. 6). Furthermore, median territory size was [sim]9 ha at Cocha Cashu compared with 6 ha at Limbo. Distribution of territory sizes at Limbo was significantly different from the distribution at Cocha Cashu ([[chi].sup.2] = 50.5, df = 6, P [less than] 0.001). Because territory size is an ecological trait that could be influenced by phylogeny, a potential problem with our statistical comparison is the use of species as replicates (e.g., Harvey and Pagel 1991). To minimize these difficulties, we performed several additional analyses. First, to improve comparability of samples, we compared territory measurements only from the species in 16 families for which similar numbers of species were measured. That included 93 species from Cocha Cashu and 72 species from Limbo. The median territory sizes at the two sites did not change and we obtained a similar statistical result ([[chi].sup.2] = 34.4, df = 6, P [less than] 0 .001).

Second, we separately analyzed territory sizes for the two most speciose families. For antbirds (Formicariidae), territory sizes were measured for 28 species at Cocha Cashu and 15 species at Limbo. Autbird territory sizes were significantly larger at Cocha Cashu (median 6 ha) than at Limbo (median 2 ha; [[chi].sup.2] = 21.2, df = 5, P = 0.002). The same result was true for 18 species of flycatcher (Tyrannidae) at Cocha Cashu (median 7 ha) vs. 16 species at Limbo (median 4 ha; [[chi].sup.2] = 10.41, df = 5, P = 0.015).

Finally, if differences in species composition account for differences in the distribution of territory sizes, the best approach would be to conduct paired comparisons of territory size of individual species that are common to both sites. Thirty-eight species (or congeneric ecological equivalents with similar body mass) have wide enough geographic ranges to occur both at Limbo and Cocha Cashu. Twenty-six of these species had smaller territory sizes at Limbo, whereas five species had larger ones; the remaining seven species occupied territories of equal size at both sites. Overall, territory sizes were significantly smaller at Limbo than at Cocha Cashu (t = 3.35, df = 37, P = 0.002). We conclude that territory sizes are indeed generally smaller at Limbo than at Cocha Cashu. The importance of territory size differences is the direct relationship with higher population densities. Within a species, smaller territory sizes accommodate a greater number of birds per unit area.

Spatial patchiness

Spatial patchiness at Limbo was evident in the distribution of birds on the study area. Nearly half of the species (45%) occupied [less than]50% of the study area, and fully one-fourth of the species occupied [less than]25% of the plot. Such patchy distributions were not unique to Limbo. Species distributions at Cocha Cashu were remarkably similar, with 52% of species occupying [less than]50% of that study area (Terborgh et al. 1990). The most obvious habitat feature that influenced bird distribution at Limbo was topography, which appeared to be associated with differences in soil drainage, plant species composition, and forest age. At least four bird species were restricted in occurrence to the foothills along the northern border, and were rarely encountered outside those areas; these included Myiobius sulphureipygius, Microcerculus marginatus, Sclerurus mexicanus, and Dysithamnus puncticeps (see Fig. 4A).

In contrast, several other species avoided areas with steep topography. Cyphorhinus phaeocephalus, for example, which fully occupied flat terrain, did not occur on high, dry ridges in the southeastern portion of Limbo or in hills to the north of Limbo (T. R. Robinson, unpublished data). Similarly, Hylopezus perspicillata, Formicarius analis, and Hylophylax naevioides (see also Willis 1972) were less numerous in hilly terrain.

Species preferring treefall gaps had conspicuously patchy distributions, reflecting the distribution of gaps (Fig. 4B; see also Schemske and Brokaw 1981). At Limbo, only a few species appeared to be gap specialists, unlike some other tropical sites where gap specialists are more numerous (Robinson et al. 1990). Colonia colonus, which flycatches for stingless bees from prominent canopy perches (Sherry 1984), selected medium-to-large gaps only if a standing dead emergent tree was available in the gap, presumably because cavities in snags were required as nesting sites. Cercomacra tyrannina, an understory antbird species, also occurred primarily in treefall gaps (see Fig. 4B).

Roadside edge effects

During the initial setup stage of the project, we were concerned that the presence of the road might introduce an unwanted level of heterogeneity, thereby inviting forest edge species that would otherwise not be present in a forest-interior plot. We found only four species that clearly had a higher abundance on our plot because of the road. (1) Nyctidromus albicollis frequently foraged from the road surface at dawn and dusk; one pair bred [less than]200 m from the study area. (2) Cercomacra tyrannina territories were frequently established in dense, vine-tangled vegetation along the roadside edge. However, those antbirds also occupied treefall gaps well away from the road (see Fig. 4B). (3) Amazilia amabilis maintained three leks in small clearings along the road and another lek site within the forest (Fig. 3). (4) Manacus vitellinus maintained a small lek of two males along the roadside edge (Fig. 3), although two females were captured regularly throughout the study in large gaps away from the road.

Presence of the road provided a full view of the open sky at a few locations, thus facilitating our counts of passing flocks of parrots and our inventories of aerial species such as swifts, swallows, and some raptors. Many of those species might have been overlooked otherwise. Three species were detected only along the road, including two vagrants, Melanerpes rubricapillus and Myiodynastes maculatus, and one migrant, Myiodynastes luteiventris. No resident species was encountered only along the road.

Guild structure

We classified resident birds into 20 different guilds based on primary dietary items consumed and, for most guilds, foraging substrate and method of capture (Karr et al. 1990, Terborgh et al. 1990; Table 2). The most speciose guilds were arboreal insectivores and arboreal omnivores. Of the 59 species of arboreal insectivore, which comprised nearly one-fourth of the species in the entire community, 33 captured prey by sallying and 21 by gleaning from live plant surfaces; the remaining 5 species searched dead leaves. Insectivores accounted for 47% of the species in the Limbo bird community, whereas 23% were omnivores.

Insectivores also dominated the community in terms of numbers of individuals. At least 64% of all birds counted were insectivores. In contrast, insectivores, with a small average body mass of 50 g, were markedly less important in terms of biomass, contributing [less than]25% of the total avian community biomass. Omnivores, however, were relatively equally important in terms of species richness (23% of all species in community), numbers of individuals (19%), and biomass (17%). Although frugivores made up [less than]10% of the species and individuals in the Limbo community, they represented almost one-fifth (18%) of the community biomass. Likewise, granivores were even less species rich (10 species) and accounted for [less than]3% of individuals, but represented nearly 30% of the community biomass. Combined, granivores and frugivores accounted for nearly one-half of the biomass while contributing [less than]10% of the species total.

DISCUSSION

The results include at least three noteworthy findings. (1) Spatial scale has an important effect on assessments of species richness and estimations of population densities of individual species at the Panama study site. (2) Overall measures of the total number of individual birds and of community biomass, however, are relatively unaffected by expansion of the spatial scale of measurement. (3) In spite of minimal differences among sites in community biomass, Panamanian and Amazonian bird communities are differently structured in terms of numbers of individual birds and levels of species richness. We will discuss how dissimilarities in biogeographic history may explain some observed differences in community structure among sites.

Comparison with small-scale census

Species richness.--Karr (1971) observed 161 bird species on his 2-ha plot, compared with 252 species that we found during our surveys of the 104-ha Limbo study area. We expected greater species richness on a larger area because it has been observed repeatedly that measurements of species richness increase with sampling area (e.g., Arrhenius 1921, MacArthur and Wilson 1967) as well as with sampling effort (Williams 1964). Tropical bird communities are no different. But how great is the difference in assessment of species richness? If we combine Karr's "resident," "regular," and "irregular" categories into a single resident category like ours (with three exceptions: we changed the classification of Legatus leucophaius and Butorides virescens from irregular to migrant, and of Thraupis palmarum from irregular to vagrant), we calculate that 133 resident species, 27 migrants, and one vagrant were observed by Karr on the 2-ha plot. Thus, we found 38 more resident species, 23 more migrants, and 20 more vagrants on t he 104-ha study plot.

One might wonder whether the particular 2 ha studied by Karr could be unusually depauperate relative to other areas in the 104-ha plot, thus exaggerating the appearance of a large effect of spatial scale. Two analyses are useful here. First, our preliminary comparison of rarefaction curves (see, e.g., Gotelli and Graves 1996) generated from Karr's inventory of the 2-ha plot and our inventory of the 104-ha plot suggests similar levels of species richness (W. D. Robinson, unpublished data). Second, we compared the species richness of Karr's 2-ha plot with the species richness in eight randomly chosen 2-ha subplots within our 104-ha plot (Table 3). Karr's measure of 161 total species falls within 2 standard deviations of the mean richness of the random plots, but tends to be higher than the richness of most of the random plots, primarily because of the large number of migrant species detected. Our own inventory in 1994-1995 of Karr's 2-ha plot reveals even greater species richness; the measure of 182 species fa lls more than two standard deviations above mean richness of the random plots. Consequently, the lower estimate of richness generated by Karr's 2-ha census relative to our 104-ha census cannot be explained simply by an unusually depauperate location for the small plot. Instead, we suspect that the location of the small plot at the interface of steep foothill terrain and flat basin forest, as well as the presence of a permanently flowing stream and the largest patch of old forest on the Limbo plot, actually cause the 2 ha to host relatively more species than other areas of the Limbo plot. Thus, the difference in species richness between the 2-ha and 104-ha plots is unlikely to result from poorer habitat quality.

On an even larger scale, an additional 21 forest-dwelling species that we did not detect at Limbo have been recorded in the 22000-ha Soberania National Park within the last 30 years; only 9 of these species, however, have been observed regularly within the last five years (Robinson 1998). Some species seem to have disappeared (see Extirpations). Using the latter figure as the number of species that were realistically present within the park during our surveys, our plot was large enough to detect 95% of the forest species present. Nearly all of those missing were species characteristic of steep terrain, which is virtually lacking on the study area. On the 2-ha plot, only 74% of the forest species were detected. Thus, the spatial scale of the study area had an important effect on enumeration of species richness.

Population density estimates.--Of primary concern to Terborgh et al. (1990) were the population density estimates derived from the 2-ha Limbo study area. Because densities from Panama were almost twice the densities from Peru, they suggested that the small spatial scale from which the Panama estimates were generated could have produced spuriously high estimates. Surprisingly, however, the estimate of 3600 individuals/l00 ha generated by the small-scale study was close to our estimate of 3200 individuals/100 ha. Does that close agreement justify use of a small spatial scale for population estimates in that particular forest?

To address that question, we looked at estimates for individual species. The first important result from our analysis is that density was estimated for substantially fewer species from the small-scale study than from our large-scale study. We measured densities for 152 resident species, whereas Karr presented measurements for 56 species. That difference suggests that estimates for each species from the 2-ha plot must have been disproportionately high to generate an estimate of the total number of individuals per 100 ha that was so similar to our estimate. Alternatively, most estimates might be similar, but the higher total could result from the much greater population densities estimated for a handful of species during the first census. Further, populations of those species could have declined in the intervening years. We examined those possibilities by comparing the estimates for each of the 53 species for which both studies estimated densities (Table 4).

Estimates from the small-scale study were significantly greater than those from the large-scale plot (Wilcoxon signed-ranks test, z = 3.99, P [less than] 0.0001). Estimates were higher for 33, lower for 18, and equivalent for two of the 53 species. Among species for which estimates were higher from the small-scale study, deviation from the large-scale plot estimates was 61.5 [pm] 24.2% (mean [pm] 1 SD). When estimates from the 104-ha plot were higher, they averaged only 39.2 [pm] 20.8% greater. It was more difficult to estimate densities of some species included in the analysis because their populations varied greatly among years, had patchy spatial distributions, or were very mobile (Table 4). Nevertheless, when these 19 species were removed from comparisons of the small- and large-scale estimates, the small-scale study still generated significantly larger estimates (z = 2.43, P [less than] 0.015). The number and magnitude of the differences between estimates derived from the two plots strongly imply that t he small spatial scale of the 2-ha study did indeed lead to inflated population density estimates.

Some difficulties can be attributed to problems posed by the patchy species distributions. Three species with the greatest percentage differences in population estimates, for example, are species that occur regularly, even today, within the 2-ha plot, but do not generally occur elsewhere in the 104-ha Limbo plot. Deconychura longicauda, Myiobius sulphureipygius, and Microcerculus marginatus all prefer primary forest in areas of steep terrain, a habitat generally lacking away from the 2-ha plot. Likewise, some of our estimates were higher for similar reasons. We detected many more individuals of several species (e.g., Myrmotherula fulviventris, Microrhopias quixensis, and Thamnophilus atrinucha) that are most common in secondary forest. These same species tend to be less numerous in primary forest, presumably because the dense tangles of vines and understory that they prefer are less common (Oniki 1975, Greenberg and Gradwohl 1985). Thus, the decidedly non-uniform spatial distribution of many species causes s ubstantial problems for estimating densities from a small-scale study area.

Densities of a few species, however, probably can be estimated accurately on small study areas in lowland central Panama. Fifteen species had territory sizes [leq]2 ha, whereas no species had a territory size that small at Cocha Cashu (see Fig. 6). Thus, the appropriate size of the study area will vary depending on the species to be studied, the questions to be addressed, and the geographic location of the study. In general, however, large study plots are required to estimate species richness, population densities, and other community-level parameters.

Biomass.--We cannot make many useful comparisons of measures of community biomass between the 2-ha and 100-ha Limbo plot studies because biomass calculations depend directly on population density estimates, which we have already demonstrated are not comparable. However, we should note that changes in hunting pressure appear to have had some effects. Some of the more prized species such as curassow and tinamou have apparently increased in abundance in the Limbo area since the establishment of the Soberania National Park (J. R. Karr, personal communication). Curassows are regularly present, but very wary, and tinamou populations appear to have increased. These large species contribute substantially to biomass calculations, and their presence facilitates comparisons with other tropical communities where hunting pressure is not as severe.

Comparison with Amazonian communities

Species richness.--The species richness of resident birds of the two well-studied Amazonian sites differs by only three species (245 in Peru, Terborgh et al. 1990; 248 in French Guiana, Thiollay 1994). In contrast, the Limbo plot hosts a total of 181 resident species, plus an additional 14 migrant species that spend several months each year on the area. Including migrants, the species richness of Limbo is about three-fourths as great as that of the Amazonian communities. What accounts for differences in species richness?

A primary factor is the influence of geographic location on the taxonomic affinities of the species present. Panama lies at the interface of the South American and Central American avifaunas, and the taxonomic composition of birds detected on the Limbo plot bely its geographic location. Neotropical migrant species, for example, contribute 20% of the total Limbo species count, a relatively low percentage for Central American sites (Rappole 1995:11), but much higher than most locations in South America (e.g., Pearson 1980, Bierregaard 1990, Robinson and Terborgh 1990, Thiollay 1994, Robinson et al. 1995). Furthermore, at least three species of austral migrant appear at Limbo, a high number for Central America, but much lower than many sites in South America, including Cocha Cashu (Robinson et al, 1988, Chesser 1995, Stotz et al. 1996).

Two relatively widespread, although not species rich, families are represented at the South American sites, but not in Panama: Opisthocomidae (Hoatzins, Opisthocomus hoazin) and Psophiidae (screamers). Furthermore, some families that occur at lowland sites in South America are very rare and local in lowland forest, or are restricted to middle and higher elevation locations, in Panama and southern Central America; (e.g., Threskiornithidae (ibises), Capitonidae (barbets), and Rhinocryptidae (tapaculos); Ridgely and Gwynne 1989) and, consequently, are not represented at Limbo. For example, 40 families occur at Cocha Cashu, whereas only 35 occur at Limbo. In addition, within the 35 families represented at both Limbo and Cocha Cashu, an average of one additional species per family occurs at Cocha Cashu (mean of 6.8 vs. 5.9 species per family). That pattern alone nearly accounts for the [sim]50 fewer species at Limbo.

Aside from family-level differences, probably the most conspicuous taxonomic difference between avifaunas of Panama and tropical South America is the depauperate species richness of some genera in Panama. The general pattern in lowland central Panama is for each genus to be represented by only one, or sometimes two, species. When multiple representatives of a genus are present, the various species usually have different habitat-specific centers of abundance. For example, among woodpeckers, Melanerpes pucherani is common in forest, whereas its congener, M. rubricapillus, is common in young secondary forest and edges. Likewise, of four species of Euphonia, only E. fulvicrissa is common in lowland forest, whereas E. minuta is relatively uncommon in the canopy of foothill forest, and E. luteicapilla and E. laniirostris are more numerous along edges and in shrubby second growth. As a result, the beta diversity of central Panama is extremely high, with nearly 650 species recorded (Engleman et al. 1995). The few im portant exceptions to the habitat-specific division of space in Panama include the raptors, for which there are several species in each of several genera; the pigeons, dominated by Columba and Geotrygon, each of which is represented by three species at Limbo; four species of Chaetura swifts; five species of Trogon; and three species of Myrmotherula antwrens.

In contrast, Amazonian sites host several species-rich genera (Robinson et al. 1990). At Cocha Cashu, for example, there are no less than six species of Crypturellus tinamous, six species of Ara parrots, five species of Xiphorhynchus woodcreepers, four species of Philydor foliage gleaner, and eight species of Myrmotherula antwren (Terborgh et al. 1990). Such speciose genera are less prevalent in lowland Panama. The number of species per genus is slightly lower in Panama (mean 1.39) than at Cocha Cashu (mean 1.52).

Extirpations.--The species richness of Limbo has been reduced, to some degree, by local disappearances of several species. Although the situation in Soberania National Park is not as pronounced as on Barro Colorado Island (e.g., Willis 1974, Karr 1982, Robinson 1999), several species missing from the Limbo community most likely occurred there within recent history. For example, there are now no species of macaw, which may have depressed not only the species richness but also the biomass of the arboreal granivore guild. Because there are no census data from early in the century, we cannot say whether populations of other parrot species may have increased in the absence of macaws.

The two largest species of eagle have not been seen in central Panama for several years. The most recent report of Morphnus guianensis occurred in 1989, whereas Harpia harpyja has not been seen since at least 1984 (D. Engleman, personal communication). Both species apparently have been extirpated from central lowland Panama, although populations of both species still occur in Chagres National Park in eastern Panama province (Ridgely and Gwynne 1989; R. Ibanez, personal communication). Given the lack of estimates of historical population densities, ramifications of the absence of those members of the top trophic level are unclear. Top mammalian predators such as jaguar (Panthera onca) and puma (Felix concolor) still occur in Soberania, and both have been detected at Limbo recently (T. Robinson and C. Edwards, unpublished data).

Another important deficit, especially when considering contributions to community biomass, is the local extirpation of the arboreal cracid Penelope purpurascens, which occurs rarely in some areas of Soberania Park, but is common, even abundant, in nonhunted areas of the adjacent Barro Colorado National Monument (Willis and Eisenmann 1979; W. D. Robinson, unpublished data). Hunting pressure may have eliminated Penelope from the Limbo area, because they have not been detected on the plot for at least 30 years (Karr 1971, Karr et al. 1990).

Although no other species have been clearly extirpated by hunting, it is likely that continued hunting pressure has affected populations of several species, especially curassows, wood-quail, and tinamous. However, hunting pressure has apparently lightened somewhat in the intervening years since Karr's (1971) investigation. He detected no curassows and found wood-quail to occur only irregularly. During our censuses, at least one pair of curassow was present at Limbo (yet the pair was extremely wary, which is suggestive of continued hunting pressure), and at least two groups of wood-quail were heard regularly. Populations of tinamous, although still under some hunting pressure, have increased since the late 1960s.

Unlike some lowland tropical sites where altitudinal migrants represent an important component of the bird community (e.g., Loiselle and Blake 1991), we detected only two hummingbird species (Phaethornis guy and Eutoxeres aquila) and one thrush species (Turdus albicollis) that are altitudinal migrants at Limbo. The number of altitudinal migrants may have been higher in the past. In the early 1940s construction of a transisthmian highway broke the contiguous corridor of forest extending from the lowlands of Soberania National Park to the middle and higher elevation forests to the east (Karr 1990; G. Angehr, personal communication). Subsequent human settlement and concomitant clearing of additional land increased the gap to its current width of [greater than]15 km (G. Angehr, personal communication). Consequently, the influx of altitudinal migrants that possibly once occurred on a regular basis has been greatly reduced. Some species that once moved seasonally into the lowlands, such as the toucanet Selenidera spectabilis and the hummingbird Klais guimeti, have rarely been recorded recently and were not recorded during our censuses. The three species of altitudinal migrants that we detected were seen less than three times each during our censuses. Unfortunately, no pre-highway censuses of the Soberania forests are available to assess the magnitude of the reduction in occurrence of altitudinal migrants.

Population densities.--One of the most striking differences between the Limbo and Amazonian bird communities is the nearly twofold greater abundance of birds per 100 ha at Limbo. Three foraging guilds in particular account for much of the difference: twice as many insectivores, twice as many omnivores, and roughly three times as many nectarivores occur at Limbo than at Cocha Cashu (see Table 2). Within insectivores, of which there were 2156/100 ha at Limbo and 1063 at Cocha Cashu, only bark-gleaning woodcreepers and bark-excavating woodpeckers were present in comparable numbers; all other guilds of insectivores were distinctly more abundant at Limbo. What accounts for such a huge disparity?

Insectivores tend to have much smaller territory sizes at Limbo (typically [less than]3 ha) than at Cocha Cashu ([geq]5 ha), so investigating the determinants of territory size may be informative. If food availability is the primary determinant and is higher in secondary forest than in primary forest, differences in successional stage of the two study sites may be important. The Limbo site, at an average age of 60-120 yr, is crowded with lianas, dense tangles of vegetation caught in the understory and canopy, and frequent treefalls, all of which create an extremely heterogeneous mix of microhabitats as well as a huge volume of foliage. Mature forest [greater than]200 yr old is uncommon on the plot, being restricted to small sections in the north and northeast. Consequently, the majority of the forest at Limbo is not fully mature.

Several studies of bird populations along successional gradients have shown that total populations are highest in mid- to late-successional growth, not in primary forest (Karr 1968, 1971). Populations of 70% of the forest species at Hubbard Brook, New Hampshire, for example, declined over the course of 15 years as the forest aged from 60 to 75 years (Holmes et al. 1986). In the Manu National Park, studies of bird populations along a primary successional gradient created by meandering of the Manu River revealed dramatic changes in the communty-wide patterns of relative abundances (Robinson and Terborgh 1997). Rank--abundance curves for each seral stage changed conspicuously from the earliest successional stages dominated by a few common species, with relatively few rare species present, to "climax" communities composed of a few moderately common species and many rare species (Robinson and Terborgh 1997). Rank--abundance curves from the two late-successional plots most closely resembled the pattern of abundanc es measured at Limbo (Kolmogorov-Smirnov tests: Limbo vs. Stage 4-5. P [greater than] 0.2; Limbo vs. Stage 5-6, P [greater than] 0.4; see Robinson and Terborgh 1997: Fig. 7). In contrast, the rank--abundance curves of Limbo and the mature forest plot at Cocha Cashu differed significantly (P [less than] 0.001), which strongly suggests that forest age may account for some observed differences in the distribution of abundances at the Limbo and Cocha Cashu sites.

Furthermore, although difficult to test, one also wonders whether a paucity of insectivorous primates at the Panama site may allow insectivorous birds to increase in abundance, perhaps by releasing them from competition for food. At Cocha Cashu the density of insectivorous primates is [sim]150/100 ha (Terborgh 1983; J. Terborgh, personal communication). In contrast, the only insect-eating primate at Limbo is Cebus capucinus. Its density does not exceed 15 individuals on the study area (W. D. Robinson, unpublished data). Thus, there exists an approximately 10-fold difference in the number of insect-eating primates at the two sites. In addition, E. G. Leigh (personal communication) has calculated that birds annually consume 31 kg dry mass of insects per hectare at Limbo compared to only 20 kg/ha at Cocha Cashu. The measurements for primates account for the 11-kg difference in insect consumption by birds: primates annually eat 13.5 kg/ha at Cocha Cashu compared to [less than]2.5 kg/ha at Limbo. Exclusion experiments designed to assess the relative impacts of primate predation on insect biomass at the two sites would be informative.

Numbers of individual birds in other guilds also vary among sites. Carrion-consuming vultures, for example, are several times more common at Limbo due in part to the presence of two nests of Cathartes aura on the study area. Abundance at Limbo also may be related to the near proximity of human habitation and its large garbage dumps, to which local vultures may commute to forage. Regardless of where food of the locally breeding individuals is obtained, dumps may maintain an unusually large regional population.

At Cocha Cashu, the abundance of granivores, including parrots and tinamous, is more than twice the number found at Limbo. In part, this may be related to local extirpations of Ara macaws and potentially to reduced abundances of tinamous by hunting pressure in Panama. Further, Cocha Cashu hosts almost five times as many nocturnal raptors and slightly more diurnal raptors.

Biomass.--Despite marked differences in population densities of birds, Limbo and Cocha Cashu have nearly equivalent biomass densities (187-190 kg/100 ha). Because body sizes of most insectivorous species are small, the huge difference among sites in population densities does not translate into a large difference in biomass. Although insectivores contribute 18% of the biomass at Cocha Cashu, they contribute only 6% more biomass at Limbo. Among granivores, however, Cocha Cashu hosts more individuals as well as more biomass. Lack of Ara macaws and Psophiid screamers contributes to the observed differences in the granivore guild.

Effects of biogeographic history

Having compensated for previous methodological differences, we can now better address the potential causes of observed differences in community structure among the three sites. Here, we evaluate several differences in site histories and resulting effects on habitat availability and its role in affecting patterns of both species richness and relative abundance.

First, one important difference among the Panamanian and Amazonian sites that may influence local levels of species richness is the influence of area on the size of the regional species pool. The Amazonian sites are situated in the midst of the vast continental lowlands of tropical South America where the total species pool of breeding land birds is [sim]1836 species (Rahbek 1997). In contrast, Limbo lies on a narrow land bridge only 70 km wide. That small land mass connects tropical South America to mostly subtropical Central America, which has greatly reduced species richness. Species richness of land birds in the lowlands of Central America is [less than]800 species (Stotz et al. 1996). Consequently, the regionally available species pool is substantially lower in Panama and is probably a primary determinant of species richness measured on a local scale. Indeed, the relationship between local and regional diversity is strongly positive and has been documented across a wide variety of organisms (Ricklefs 19 87, Caley and Schluter 1997).

Second, there may be an isthmus effect analagous to the peninsula effect described by Simpson (1964), who noted a common pattern of decreasing species richness as one nears the tip of geographic peninsulas. He offered no mechanistic explanation for the pattern, but it is likely to be influenced by a reduction in the size of the regional species pool nearer the peninsular tip (Gotelli and Graves 1996). Essentially, the reduction is effected by a progressive decrease in the area of terrestrial habitats. Thus, the reduced species richness at the Limbo site on the Panamanian isthmus relative to that observed in lowland Amazonia may be analagous to Simpson's peninsula effect. To our knowledge, no detailed geographic summary of changes in avian species richness from the South American mainland through the isthmus of Panama and southern Central America has been conducted yet, although the data necessary for such an analysis have become available recently (see Stotz et al. 1996).

Third, different histories of disturbance could be at least partially responsible for different levels of species richness (Karr et al. 1990). The Panama site appears to have suffered a relatively greater amount of disturbance on several time scales. In geological time, Panama has been disturbed frequently during opening and closure of the isthmian strait (Collins et al. 1996). The connection with both the South and North American continents has been broken several times in the last 10 X [10.sup.6] years by the appearance of the isthmian strait as well as straits in the Atrato Basin of northwestern Colombia and in the San Carlos Basin of southern Nicaragua (Savin and Douglas 1985, Coates and Obando 1996). The most recent separation was [sim]3.5 X [10.sup.6] years ago (Keigwin 1982, Collins et al. 1996). Thus, Panama was effectively composed of a series of islands rather than a complete isthmian connection to the rich South American continent. In contrast, Nouragues and Cocha Cashu are imbedded in the midst o f the massive continental interior.

More recently, reductions in temperature associated with glaciation at high latitudes were responsible for downward elevational shifts in montane forest habitats in central Panama (Bush et al. 1992, Bush and Colinvaux 1994). Those shifts were apparently accompanied by drying in lowland areas and the concomitant widespread reduction of humid lowland forest (Bush and Colinvaux 1990, Bush et al. 1992). Such a reduction in the area of lowland forest may have led to local extinctions of some forest-dwelling bird species and to significant reductions in population sizes of the remaining species. This scenario assumes that lowland forest bird species did not occupy montane forest that invaded lower elevations as temperatures declined. Given present-day restrictions of many species to specific elevational ranges in Panama (Ridgely and Gwynne 1989), it seems reasonable to suspect that populations of many species did indeed decline or perhaps even disappear entirely from the Panama region.

In contrast, the situation in Amazonia appears to have been less severe. Both Cocha Cashu and Nouragues are located within proposed Pleistocene refugia, whereas Limbo is not (Haffer 1969, Prance 1987). Although the aridity of the Amazonian basin typically associated with the development of South American refugia has not been confirmed by palaeoecological data (Colinvaux 1987, Bush et al. 1990), patterns of endemism associated with the proposed refugial sites remain informative. The refugial sites often have relatively higher species richness than surrounding non-refugial areas (Haffer 1987), which, because Limbo is the only one of the three sites not in a proposed refugium, may provide a further explanation of the differences in species richness.

More recently, central Panama has been disturbed often by humans in the last 11,000 years (Bush et al. 1992, Bush and Colinvaux 1994). Disturbances include recent clearing of forest in central Panama on a massive scale by human populations greater in size than those present today; some estimates suggest that nearly a million humans resided in central Panama just before the arrival of Columbus (Bennett 1968, Piperno et al. 1990). In particular, flatter terrain supporting more successful agricultural practices was subjected to greater levels of disturbance than was steep topography. Consequently, the relatively flat topography in the 600-ha basin in which the Limbo study area lies, and the basin's proximity to the heavily traveled Camino de Cruces and Chagres River trading route used by the Spaniards in the 1500-1600s, have undoubtedly predisposed it to repeated disturbance by humans. Evidence of farming practices has been discovered at Limbo, including the presence of matate stones used for grinding grain and of huge, emergent Enterolobium trees that are typically able to regenerate only in pastured lands (Croat 1978). Although human presence and agricultural activities also occurred in the Amazon basin (Bush and Colinvaux 1988, Bush et al. 1989, Piperno 1990), the impacts on Panama were probably orders of magnitude greater. Thus, disturbance has been an important influence on the biogeographic history of lowland forest in central Panama.

Similarly, disturbance has also been important in Amazonian communities. Evidence indicating occurrence of repeated disturbance of various types, including fire (Sanford et al. 1985) and river dynamics (Salo et al. 1986, R[ddot{a}]s[ddot{a}]nen et al. 1987), has become increasingly available. River dynamics, in particular, are important contributors to avian species diversity in Amazonia (Remsen and Parker 1983, Robinson and Terborgh 1997). Nevertheless, the magnitude of disturbance at the Panama site appears to have been relatively greater, on several time scales, than that at the Amazonian sites.

Although historical effects on both geological and recent time scales appear to have influenced differences among the bird communities in terms of species richness, the most recent disturbance at Limbo may better explain the substantial differences in total numbers of birds present and in their patterns of abundance. The average forest age at Limbo ranges from 60 to 120 years, with only a few patches of older forest remaining. Both Amazonian study sites host forest that has been undisturbed for [geq]500 years (Terborgh et al. 1990, Thiollay 1994), although the Cocha Cashu plot is bordered by lands disturbed more recently by river dynamics. Consequently, differences in successional stage among the study areas could be important. Although the forest at Limbo is mature enough to attract nearly all local forest-dwelling species, distribution of abundances in the bird community may be different than that in primary forest. For example, diversity (as measured in terms of both richness and abundances) is often rela tively higher in secondary than in primary forest communities (Rosenzweig and Abramsky 1993). The precise mechanisms generating that pattern remain unclear, but are most likely caused by a decrease in aboveground net primary production that accompanies progression in forest age (Saldarriaga and Luxmoore 1991, Gower et al. 1996). The decrease in productivity translates into lower biomass and numbers of individuals at higher trophic levels. (We should note briefly that data from a primary successional sequence at Cocha Cashu [Robinson and Terborgh 1997] indicate a steady increase in bird species richness from youngest to oldest seral stage. However, total numbers of individual birds do indeed peak in midsuccessional forest.) Emmons (1984) demonstrated a similar relationship between soil fertility and population densities of mammals across Amazonia: population densities were positively correlated with measures of soil fertility and overall productivity. Thus, differences in productivity among Limbo and the Amazo nian bird plots may help to explain the large differences that we have documented in total numbers of birds.

Comparisons of our results with population density measurements from primary forest elsewhere in lowland Panama would help to address this possibility. The only such data are those gathered from the older sections of forest on nearby Barro Colorado Island (Willis 1974, 1980). Possible island effects notwithstanding, our initial comparisons of densities for a subset of species for which both we and Willis made measurements revealed no consistent differences (W. D. Robinson, unpublished data). More detailed inventories in lowland primary forest in Panama will be necessary to test more thoroughly the suggested effect of forest age.

Effects of spatial scale

Until very recently, the spatial scale at which both neo- and paleotropical bird communities have been investigated has often been too small. Most studies have been conducted on areas of [leq]25 ha (Davis 1955, MacArthur et al. 1966, McClure 1969, Orians 1969, Terborgh and Weske 1969, Karr 1971, Fogden 1976, Bell 1982). In most tropical localities, the extraordinarily patchy distributions of species, even within tracts of apparently "uniform" habitat, will cause an underestimation of local species richness when very small plots are used. Although the 104-ha Limbo plot hosted [greater than]95% of the locally occurring forest species, and the 97-ha Cocha Cashu plot hosted 99% of the local, mature floodplain forest species, areas even of that scale will not always be large enough (see Table 5). In the 100-ha French Guiana study, for instance, only 77% of the locally occurring forest avifauna were observed on the study area (Thiollay 1994).

A similar difficulty has beset studies of population densities. As shown in our comparison of estimates from the original 2-ha Limbo plot and the current 104-ha plot, studies of small areas can generate very different population density estimates. That appears to happen in at least three ways. First, few species present will occur within the small area frequently enough to generate sufficient observations to allow adequate estimates. In our comparison, the 2-ha allowed estimates for one-third of the number of species that the 104-ha plot allowed. Second, small plots are unlikely to adequately sample patchy distributions of birds. At both Limbo and Cocha Cashu, [sim]50% of species were present on less than half of the study areas. Plot size, then, must be large enough to sample the habitat in a manner that will encompass patchy distributions of species. Third, many species have territory sizes that are large, relative to territory sizes of temperate birds, so plots must be big enough to include the entire ter ritory before accurate estimates can be made. In central Panama, a 2-ha plot will be large enough to include an entire territory of [less than]20% of the species present (see Fig. 5).

In many cases, plots may need to be huge to encompass a sufficient number of territories to describe the variability in territory size, or to generate robust density estimates. One could argue, for instance, that in Amazonia, where the two recent studies have shown one-third of the species to occur in densities of [leq]1 pair/100 ha, a sample of 100 ha provides only a single territory from which to derive density estimates for those species, on average. Surely, with such small samples, the degree of uncertainty of measurement is great.

For conservation, as well as scientific, reasons, those methodological shortcomings require prompt resolution. Tropical forests, with the highest diversity of species on earth, are being rapidly converted to secondary habitats. Yet, we have only barely begun to quantify population densities of a handful of species at a handful of sites. Accurate estimates of tropical bird population densities require that effects of spatial scale must be considered during the design of studies (Wiens 1989, Levin 1992). Although the necessary spatial scale depends on the specific questions being asked, further advances in determining population densities of tropical birds will probably require studies on scales large enough to incorporate several pairs of each locally occurring species. Areas perhaps as large as 500-1000 ha may be required at some sites, especially where extreme rarity is a prominent feature of the community. Also, the census of multiple plots will be required before we can describe adequately the variation i n densities of tropical organisms on both spatial and temporal scales.

Last, most tropical forest species have global population sizes that are orders of magnitude smaller than those of the temperate species to which we direct so many of our monetary resources (see, e.g., Stotz et al. 1996). Thus, not only is documentation of population densities important for ecologists interested in basic issues in biology, but such measurements are critical for demonstrating the magnitude of the differences between tropical and temperate systems. Conservation efforts can only be helped when the extent to which tropical organisms are exposed to inherently higher risk of extinction, by way of their smaller population sizes, is made abundantly clear.

ACKNOWLEDGMENTS

The Smithsonian Tropical Research Institute (STRI) Environmental Sciences Program supported our investigations. W. D. Robinson was also supported by grants from the Frank M. Chapman fund; the Alexander Wetmore Memorial Award of the American Ornithologists' Union; and the Center for Latin American and Carribean Studies, the Department of Ecology, Ethology, and Evolution, the Graduate College, and the School of Life Sciences, at the University of Illinois at Urbana-Champaign. We are especially thankful to Tara R. Robinson, who contributed substantially to all aspects of the research. Karl Kaufman generously assisted with computer programming. Rick Condit shared unpublished data on vegetation composition of the study area. We also are grateful to IN.RE.NA.RE for allowing us the privelege to study birds in the Republic of Panama, and to STRI for logistical support. The manuscript has been improved by the comments of A. E. Herre, R. Holmes, E. K. V. Kalko, J. R. Karr, E. G. Leigh, T. R. Robinson, S. Russo, D. Sto tz, P. Stouffer, and S. J. Wright. Finally, we thank 3. R. Karr, who, remarkably as a graduate student, broke new ground in tropical avian ecology and inspired the current round of investigation.

(1.) Department of Ecology, Ethology, and Evolution, 515 Morrill Hall, University of Illinois, 505 5. Goodwin Avenue, Urbana, Illinois 61801 USA

(2.) Illinois Natural History Survey, 607 E. Peabody Drive, Champaign, Illinois 61820 USA

(3.) Present address: Department of Zoology and Wildlife Science, 331 Funchess Hall, Auburn University, Auburn, Alabama 36849-5414 USA.

E-mail: wdrobins@acesag.auburn.edu

LITERATURE CITED

Arrhenius, O. 1921. Species and area. Journal of Ecology 9: 95-99.

Beehler, B. M., J. B. Sengo, C. Filardi, and K. Merg. 1995. Documenting the lowland rainforest avifauna in Papua New Guinea--effects of patchy distributions, survey effort, and methodology. Emu 95:149-161.

Bell, H. L. 1982. A bird community of lowland rainforest in New Guinea. I. Composition and density of the avifauna. Emu 82:24-41.

Bennett, C. 1968. Human influences on the zoogeography of Panama. Ibero-Americana 51. University of California Press, Berkeley, California, USA.

Bierregaard, R. 0., Jr. 1990. Species composition and trophic organization of the understory bird community in a central Amazonian terra firme forest. Pages 217-236 in A. H. Gentry, editor. Four neotropical rainforests. Yale University Press, New Haven, Connecticut, USA.

Blake, J. G. 1992. Temporal variation in point counts of birds in a lowland wet forest in Costa Rica. Condor 94:265-275.

Brawn, J. D., J. R. Karr, and J. D. Nichols. 1995. Demography of birds in a neotropical forest: effects of allometry, taxonomy, and ecology. Ecology 76:41-51.

Bush, M. B., and P. A. Colinvaux. 1988. A 7000-yr pollen record from the Amazon lowlands, Ecuador. Vegetatio 76:141-154.

Bush, M. B., and P. A. Colinvaux. 1990. A long record of climatic and vegetation change in lowland Panama. Journal of Vegetation Science 1:105-119.

Bush, M. B., and P. A. Colinvaux. 1994. Tropical forest disturbance: paleocological records from Darien, Panama. Ecology 75:1761-1768.

Bush, M. B., D. R. Piperno, and P. A. Colinvaux. 1989. A 6000 year history of Amazonian maize cultivation. Nature 340:303-305.

Bush, M. B., D. R. Piperno, P. A. Colinvaux, P. E. De Oliveira, L. A. Krissek, M. C. Miller, and W. E. Rowe. 1992. A 14 300-yr paleoecological profile of a lowland tropical lake in Panama. Ecological Monographs 62:251-276.

Bush, M. B., M. Weimann, D. R. Piperno, K.-B. Liu, and P. A. Colinvaux. 1990. Pleistocene temperature change and vegetation depression in Ecuadorian Amazonia. Quaternary Research 34:330-345.

Caley, M. J., and D. Schluter. 1997. The relationship between local and regional diversity. Ecology 78:70-80.

Campbell, D. G. 1994. Scale and patterns of community structure in Amazonian forests. Pages 179-198 in P. J. Edwards, R. M. May, and N. R. Webb, editors. Large-scale ecology and conservation biology. Blackwell Science, Cambridge, UK.

Chesser, R. T. 1995. Migration in South America: an overview of the austral system. Bird Conservation International 4:91-108.

Coates, A. G., and J. A. Obando. 1996. The geologic evolution of the Central American isthmus. Pages 21-56 in J. B. C. Jackson, A. F. Budd, and A. G. Coates, editors. Evolution and environment in tropical America. University of Chicago Press, Chicago, Illinois, USA.

Colinvaux, P. A. 1987. Amazon diversity in the light of the paleoecological record. Quaternary Science Reviews 6:93-114.

Collins, L. S., A. G. Coates, W. A. Berggren, M.-P. Aubry, and J. Zhang. 1996. The late Miocene Panama isthmian strait. Geology 24:687-690.

Croat, T. B. 1978. Flora of Barro Colorado Island. Stanford University Press, Stanford, California, USA.

Davis, L. I. 1955. Bird census 27. Audubon Field Notes 9: 425-426.

Diamond, J. 1980. Patchy distributions of tropical birds. Pages 57-74 in M. E. Soul[acute{e}] and B. A. Wilcox, editors. Conservation biology: an evolutionary-ecological perspective. Sinauer Associates, Sunderland, Massachusetts, USA.

Emmons, L. H. 1984. Geographic variation in densities and diversities of non-flying mammals in Amazonia. Biotropica 16:210-222.

Engleman, D.,G. Angehr, and M. Allen. 1995. Lista de las aves de Panam[acute{a}]. Volumen 1: Ciudad de Panam[acute{a}] y alrededores. Sociedad Audubon de Panam[acute{a}], Panam[acute{a}], Rep[acute{u}]blica de Panam[acute{a}],

Fogden, M. P. L. 1976. A census of a bird community in tropical rain forest in Sarawak. Sarawak Museum Journal 45:251-267.

Foster, R. B., and N. V. L. Brokaw. 1982. Structure and history of the vegetation of Barro Colorado Island. Pages 67-81 in E. G. Leigh, Jr., A. S. Rand, and D. M. Windsor, editors. Ecology of a tropical forest. Smithsonian Institution Press, Washington, D.C., USA.

Gentry, A. H. 1990. Four neotropical rainforests. Yale University Press, New Haven, Connecticut, USA.

Gotelli, N. J., and G. R. Graves. 1996. Null models in ecology. Smithsonian Institution Press, Washington, D.C., USA.

Gower, S. T., R. E. McMurtrie, and D. Murty. 1996. Aboveground net primary production decline with stand age: potential causes. Trends in Ecology and Evolution 11:378-382.

Greenberg, R., and J. Gradwohl. 1985. A comparative study of the social organization of antwrens on Barro Colorado Island, Panama. Pages 845-855 in P. A. Buckley, M. S. Foster, E. S. Morton, R. S. Ridgely, and F. G. Buckley, editors. Neotropical Ornithology. Ornithological Monographs Number 36.

Greenberg, R., and J. Gradwohl. 1986. Constant density and stable territoriality in some tropical insectivorous birds. Oecologia 69:618-625.

Haffer, J. 3. 1969. Speciation in Amazonian forest birds. Science 165:131-137.

Haffer, 3. 1987. Biogeography of neotropical birds. Pages 105-150 in T. C. Whitmore and G. T. Prance, editors. Biogeography and quaternary history in tropical America. Oxford Monographs on Biogeography Number 3. Clarendon Press, Oxford, UK.

Harvey, P. H., and M.D. Pagel. 1991. The comparative method in evolutionary biology. Oxford University Press, Oxford, UK.

Holdridge, L. R. 1967. Life zone ecology. Occasional Papers of the Tropical Science Center. San Jose Costa Rica.

Holmes, R. T., T. W. Sherry, and F. W. Sturges. 1986. Bird community dynamics in a temperate deciduous forest: longterm trends at Hubbard Brook. Ecological Monographs 56:201-220.

Hubbell, S. P. 1979. Tree dispersion, abundance, and diversity in a tropical dry forest. Science 203:1299-1309.

Hubbell, S. P., and R. B. Foster. 1986. Canopy gaps and the dynamics of a neotropical forest. Pages 77-95 in M. J. Crawley, editor. Plant ecology. Blackwell Scientific, Oxford, UK.

Karr, J. R. 1968. Habitat and avian diversity on strip-mined land in east-central Illinois. Condor 70:348-357.

Karr, J. R. 1971. Structure of avian communities in selected Panama and Illinois habitats. Ecological Monographs 41:207-233.

Karr, J. R. 1980. Geographical variation in the avifaunas of tropical forest undergrowth. Auk 97:283-298.

Karr, J. R. 1981. Surveying birds in the tropics. Studies in Avian Biology 6:548-553.

Karr, J. R. 1982. Avian extinction on Barro Colorado Island, Panama: a reassessment. American Naturalist 119:220-239.

Karr, J. R. 1990. The avifauna of Barro Colorado Island and the Pipeline Road, Panama. Pages 183-198 in A. H. Gentry, editor. Four neotropical rainforests. Yale University Press, New Haven, Connecticut, USA.

Karr, J. R., and K. E. Freemark. 1983. Habitat selection and environmental gradients: dynamics in the "stable" tropics. Ecological Monographs 64:1481-1494.

Karr, J. R., S. K. Robinson, R. O. Bierregaard, Jr., and J. G. Blake. 1990. Birds of four neotropical forests. Pages 237-269 in A. Gentry, editor. Four neotropical rainforests. Yale University Press, New Haven, Connecticut, USA.

Karr, J. R., M. F. Willson, and D. J. Moriarty. 1978. Weights of some Central American birds. Brenesia 14-15:249-257.

Keigwin, L. D., Jr. 1982. Isotopic paleoceanography of the Caribbean and East Pacific: Role of Panama uplift in late Neogene time. Science 217:350-353.

Kendeigh, S. C. 1944. Measurement of bird populations. Ecological Monographs 14:67-106.

Levin, S. 1992. The problem of pattern and scale in ecology. Ecology 73:1943-1983.

Loiselle, B. A., and J. G. Blake. 1991. Temporal variation in birds and fruits along an elevational gradient in Costa Rica. Ecology 72:180-193.

Lynch, J. F. 1995. Effects of point count duration, time-of-day, and aural stimuli on detectability of migratory and resident bird species in Quintana Roo, Mexico. Pages 1-6 in C. J. Ralph, J. R. Sauer, and S. Droege, editors. Monitoring bird populations by point counts. USDA Forest Service General Technical Report PSW-GTR-149.

MacArthur, R. H., H. Recher, and M. L. Cody. 1966. On the relation between habitat selection and species diversity. American Naturalist 100:319-332.

MacArthur, R. H., and E. O. Wilson. 1967. The theory of island biogeography. Princeton University Press, Princeton, New Jersey, USA.

May, R. M. 1994. The effects of spatial scale on ecological questions and answers. Pages 1-18 in P. J. Edwards, R. M. May, and N. R. Webb, editors. Large-scale ecology and conservation biology. Blackwell Scientific, Cambridge, UK.

McClure, H. E. 1969. Estimations of bird population density in primary forest of Malaya. Malay Nature Journal 22:179-183.

Munn, C. A. 1985. Permanent canopy and understory flocks in Amazonia: species composition and population density. Pages 638-712 in P. A. Buckley, M. S. Foster, F. S. Morton, R. S. Ridgely, and F. G. Buckley, editors. Neotropical ornithology. Ornithological Monographs Number 36.

Oniki, Y. 1975. The behavior and ecology of the Slaty Antshrike on Barro Colorado Island, Panama Canal Zone. Annals de Academia Brasil Ciencia 47:477-514.

Orians, G. H. 1969. The number of bird species in some tropical forests. Ecology 50:783-801.

Pearson, D. L. 1980. Bird migration in Amazonian Ecuador, Peru, and Bolivia. Pages 273-283 in A. Keast and E. S. Morton, editors. Migrant birds in the Neotropics: ecology, behavior, distribution, and conservation. Smithsonian Institution Press, Washington, D.C., USA.

Pierpont, N. 1986. Interspecific aggression and the ecology of woodcreepers (Ayes: Dendrocolaptidae). Dissertation. Princeton University, Princeton, New Jersey, USA.

Piperno, D. R. 1990. Aboriginal agriculture and land usage in the Amazon Basin, Ecuador. Journal of Archaeological Science 17:665-677.

Piperno, D. R., M. B. Bush, and P. A. Colinvaux. 1990. Paleoenvironments and human settlement in late-glacial Panama. Quaternary Research 33:108-116.

Prance, G. T. 1987. Biogeography of neotropical plants. Pages 46-65 in T. C. Whitmore and G. T. Prance, editors. Biogeography and quaternary history in tropical America. Oxford Monographs on Biogeography Number 3. Clarendon Press, Oxford, UK.

Rahbek, C. 1997. The relationship among area, elevation, and regional species richness among Neotropical birds. American Naturalist 149:875-902.

Rappole, J. H. 1995. The ecology of migrant birds: a neotropical perspective. Smithsonian Institution Press, Washington, D.C., USA.

R[ddot{a}]s[ddot{a}]nen, M. E., J. S. Salo, and R. Kalliola. 1987. Fluvial perturbance in the Western Amazon Basin: regulation by long-term sub-Andean tectonics. Science 238:1398-1401.

Remsen, J. V., Jr., and D. A. Good. 1996. Misuse of data from mist-net captures to assess relative abundance in bird populations. Auk 113:381-398.

Remsen, J. V., Jr., and T. A. Parker III. 1983. Contribution of river-created habitats to bird species richness in Amazonia. Biotropica 15:223-231.

Ricklefs, R. E. 1987. Community diversity: relative roles of local and regional processes. Science 235:167-171.

Ricklefs, R. E. 1990. Ecology. Third edition. W. H. Freeman and Company, New York, New York, USA.

Ricklefs, R. F., and D. Schluter. 1993. Species diversity in ecological communities: historical and geographical perspectives. University of Chicago Press, Chicago, Illinois, USA.

Ridgely, R. S., and J. A. Gwynne, Jr. 1989. A guide to the birds of Panama with Costa Rica, Nicaragua, and Honduras. Second edition. Princeton University Press, Princeton, New Jersey, USA.

Robinson, S. K., J. W. Fitzpatrick, and J. Terborgh. 1995. Distribution and habitat use of Neotropical migrant landbirds in the Amazon basin and Andes. Bird Conservation International 5:305-323.

Robinson, S. K., and J. Terborgh. 1990. Bird communities of the Cocha Cashu Biological Station in Amazonian Peru. Pages 199-216 in A. H. Gentry, editor. Four neotropical rainforests. Yale University Press, New Haven, Connecticut, USA.

Robinson, S. K., and J. Terborgh. 1995. Interspecific aggression and habitat selection by Amazonian birds. Journal of Animal Ecology 64:1-11.

Robinson, S. K., and J. Terborgh. 1997. Bird community dynamics along primary successional gradients of an Amazonian whitewater river. Pages 641-672 in J. V. Remsen, Jr., editor. Studies in neotropical ornithology honoring Ted Parker. Ornithological Monographs Number 48, American Ornithologists' Union, Washington, D.C., USA.

Robinson, S. K., J. Terborgh, and J. W. Fitzpatrick. 1988. Habitat selection and relative abundance of migrants in southeastern Peru. International Ornithological Congress XIX:2298-2307.

Robinson, S. K., J. Terborgh, and C. A. Munn. 1990. Lowland tropical forest bird communities of a site in Western Amazonia. Pages 229-258 in A. Keast, editor. Biogeography and ecology of forest bird communities. Academic Press, The Hague, The Netherlands.

Robinson, W. D. 1998. The ecology of a tropical bird community, with emphases on community structure, nesting success, and isolation effects. Dissertation. University of Illinois, Urbana-Champaign, Illinois, USA.

Robinson, W. D. 1999. Long-term changes in the avifauna of Barro Colorado Island, Panama, a tropical forest isolate. Conservation Biology 13:85-97.

Rosenzweig, M. L., and Z. Abramsky. 1993. How are diversity and productivity related? Pages 52-65 in R.E. Ricklefs and D. Schluter, editors. Species diversity in ecological communities: historical and geographic perspectives. University of Chicago Press, Chicago, Illinois, USA.

Saldarriaga, J., and R. Luxmoore. 1991. Solar energy conversion efficiencies during succession of a tropical rain forest in Amazonia. Journal of Tropical Ecology 7:233-242.

Salo, J., R. Kalliola, I. Hakkinen, Y. Makinen, P. Niemela, M. Puhakka, and P. D. Coley. 1986. River dynamics and the diversity of Amazon lowland forest. Nature 322:254-258.

Sanford, R. L., Jr., J. Saldarriaga, K. E. Clark, C. Uhl, and R. Herrera. 1985. Amazon rain-forest fires. Science 227:53-55.

Savin, S. M., and R. G. Douglas, 1985. Sea level, climate, and the Central American land bridge. Pages 303-324 in F. D. Stehli and S. D. Webb, editors. The great American biotic interchange. Plenum Press, New York, New York, USA.

Schemske, D. W., and N. Brokaw. 1981. Treefalls and the distribution of understory birds in a tropical forest. Ecology 62:938-945.

Schluter, D., and R. E. Ricklefs. 1993. Species diversity: an introduction to the problem. Pages 1-10 in R. E. Ricklefs and D. Schluter, editors. Species diversity in ecological communities: historical and geographical perspectives. University of Chicago Press, Chicago, Illinois, USA.

Sherry, T. W. 1984. Comparative dietary ecology of sympatric, insectivorous neotropical flycatchers (Tyrannidae). Ecological Monographs 54:313-338.

Siegel, S., and N. J. Castellan, Jr. 1988. Nonparametric statistics for the behavioral sciences. Second edition. McGraw-Hill, New York, New York, USA.

Simpson, G. G. 1964. Species density of North American recent mammals. Systematic Zoology 13:57-73.

Skutch, A. F. 1996. Antbirds and ovenbirds, their lives and homes. University of Texas Press, Austin, Texas, USA.

StatXact. 1995. Statistical software for exact nonparametric inference. Cytel Software, Cambridge, Massachusetts, USA.

Stickel, L. F. 1954. A comparison of certain methods of measuring ranges of small mammals. Journal of Mammalogy 35:1-15.

Stiles, F. G., and A. F. Skutch. 1989. A guide to the birds of Costa Rica. Cornell University Press, Ithaca, New York, USA.

Stotz, D. F., J. W. Fitzpatrick, T. A. Parker III, and D. K. Moskovits. 1996. Neotropical birds: ecology and conservation. University of Chicago Press, Chicago, Illinois, USA.

Stouffer, P. C., and R. O. Bierregaard, Jr. 1993. Spatial and temporal abundance patterns of Ruddy Quail-Doves (Geotrygon montana) near Manaus, Brazil. Condor 95:896-903.

Terborgh, J. 1971. Distribution on environmental gradients: theory and a preliminary interpretation of distributional patterns in the avifauna of the Cordillera Vilcabamba, Peru. Ecology 52:23-40.

Terborgh, J. 1983. Five new world primates: a study in comparative ecology. Princeton University Press, Princeton, New Jersey, USA.

Terborgh, J., J. W. Fitzpatrick, and L. H. Emmons. 1984. Annotated checklist of bird and mammal species of Cocha Cashu Biological Station, Manu National Park, Peru. Fieldiana, Zoology 21:1-29.

Terborgh, J., R. B. Foster, and V. Percy Nunez. 1996. Tropical tree communities: a test of the nonequilibrium hypothesis. Ecology 77:561-567.

Terborgh, J., S. K. Robinson, T. A. Parker III, C. A. Munn, and N. Pierpont. 1990. Structure and organization of an Amazonian forest bird community. Ecological Monographs 60:213-238.

Terborgh, J., and J. S. Weske. 1969. Colonization of secondary habitats by Peruvian birds. Ecology 50:765-782.

Thiollay, J.-M. 1994. Structure, density, and rarity in an Amazonian rainforest bird community. Journal of Tropical Ecology 10:449-481.

Tuomisto, H., K. Ruokolainen, R. Kalliola, A. Linna, W. Danjoy, and Z. Rodriguez. 1995. Dissecting Amazonian biodiversity. Science 269:63-66.

Verner, J. 1985. Assessment of counting techniques. Current Ornithology 2:247-302.

Wallace, A. R. 1878. Tropical nature and other essays. Macmillan, New York, New York, USA.

Wetmore, A., R. F Pasquier, and S. L. Olson. 1984. The birds of the Republic of Panama, part 4. Smithsonian Institution Press, Washington, D.C., USA.

Wiens, J. A. 1989. Spatial scaling in ecology. Functional Ecology 3:385-397.

Williams, C. B. 1964. Patterns in the balance of nature. Academic Press, New York, New York, USA.

Willis, E. O. 1972. The behavior of spotted antbirds. Ornithological Monographs Number 10.

Willis, E. O. 1974. Populations and local extinctions of birds on Barro Colorado Island, Panama. Ecological Monographs 44:153-169.

Willis, E. O. 1980. Ecological roles of migratory and resident birds on Barro Colorado Island, Panama. Pages 205-225 in A. Keast and E. S. Morton, editors. Migrant birds in the Neotropics: ecology, behavior, distribution and conservation. Smithsonian, Institution Press, Washington, D.C., USA.

Willis, E. O., and E. Eisenmann. 1979. A revised list of birds of Barro Colorado Island, Panama. Smithsonian Contributions to Zoology Number 291.

Windsor, D. M. 1990. Climate and moisture variability in a tropical forest: long-term records from Barro Colorado Island, Panama. Smithsonian Contributions to Earth Sciences, Number 29.
                 Distribution of bird species at Limbo by
                  residency status (see Methods: Species
                   status and presentation of population
                            density estimates).
Residency status                 No. species
Residents (n = 181 species)
 Present in measurable densities     152
 Too rare to measure                  19
 Too mobile to measure (swifts,       10
swallows)
Migrants (n = 48 species) [+]
 Latitudinal (n = 45 species)
 Transient only                       22
 Also wintering                       22
 Also breeding                         1
 Altitudinal                           3
Vagrants (n = 23 species)
 From aquatic habitats                 6
 From open/edge habitats              17
Total no. species                    252
(+.)Thirteen wintering species were
present in measurable densities. Thus, a
total of 165 species had measureable
population densities.
                     Guild structure of the Limbo bird
                                community.
                            Number of individuals
                                                  Difference Mean mass
Guild          No. spp. [+]      Limbo plot        (%) [++]     (g)
Aquatic           5 (7)              13               +37       325
Carrion           2 (3)               3              +500      1550
Frugivore
  Arboreal       12 (15)            241               +26       142
  Terrestrial     4 (4)              61              +259       135
Granivore
  Arboreal        6 (7)              41               -49       269
  Terrestrial     3 (3)              36               -64      1736
Insectivore
  Aerial          1 (17)              2                 0        75
  Ant follower    8 (8)             109              +230       120
  Arboreal
   Dead leaf      5 (5)             215              +412        21
   Gleaner       12 (21)            916              +123        23
   Sallier       26 (33)            540               +76        52
  Bark
   Interior       4 (6)              37                -4       133
   Surface        5 (5)             116               +37        30
  Terrestrial
   Gleaner       15 (16)            219               +71        59
   Sallier        1 (2)               2               -89        53
Nectarivore       8 (12)            131              +195         5
Omnivore
  Arboreal       35 (52)            645             + 114        48
  Terrestrial     1 (4)               8            [ldots]      300
Raptor
  Diurnal         9 (17)             17               +26       489
  Nocturnal       3 (5)              12               -65       487
Total           165 (242)          3364 [ss]          +69
               Biomass density (g/100 ha)
                                          Difference
Guild                  Limbo plot          (%) [++]
Aquatic                    3202                +4
Carrion                    4900              +717
Frugivore
  Arboreal               27 127                +6
  Terrestrial              8221               -16
Granivore
  Arboreal               15 572               -14
  Terrestrial            43 400               -29
Insectivore
  Aerial                    150               -14
  Ant follower             4041              +158
  Arboreal
   Dead leaf               2791              +188
   Gleaner               15 456               +89
   Sallier               10 407               -10
  Bark
   Interior                3340                +3
   Surface                 3297                +8
  Terrestrial
   Gleaner                 8509               +73
   Sallier                  106               -77
Nectarivore                 644              +184
Omnivore
  Arboreal               30 407               +43
  Terrestrial              2400            [ldots]
Raptor
  Diurnal                  5659               +23
  Nocturnal                3520               -54
Total                   193 199                -3


Notes: Resident species only are included. For assignment of species to guilds, see Table 5.

(+.)In parentheses, the total number of species present; outside parentheses, the total number of species for which densities were measured. The latter is used in calculations of the remaining three parameters.

(++.)Percentage difference of the Limbo data relative to those from Cocha Cashu (Terborgh et al. 1990).

(ss.)The total number of individuals, adjusted to 100-ha plot size, is 3234. The adjusted biomass density is 187 kg/100 ha.
                    Comparison of the number of species
                  detected on eight randomly located 2-ha
                  subplots within the 104-ha plot vs. the
                    numbers detected by Karr (1971) in
                    1968-1969 and by us in 1994-1995 on
                             Karr's 2-ha plot.
                            No. spp. on
          No. spp. on       Karr's plot
          random plots
Residency                      1968-    1994-
status        Mean     1 SD    1969     1995
Residents    111.0     18.5     133      156
Migrants       9.6      6.6      27       24
Vagrants       1.1      1.6       1        2
All          121.8     26.6     161      182
                     Comparison of population density
                   estimates of the 53 bird species for
                    which both Karr (1971) and we made
                                 esimates.
                          No. individuals/100 ha
Species                      Karr's estimate     This study
Tinamus major                       20               30
Geotrygon montana                    5               30
Phaethornis superciliosus           70               40
P. longuemareus                     50               16
Trogon rufus                        20               42
Trogon massena                      40               40
Baryphthengus martii                40               48
Electron platyrhynchum              30               30
Notharchus pectoralis               20               28
Malacoptila panamensis              40               28
Ramphastos sulfuratus               30               16
R. swainsonii                       20               12
Melanerpes pucherani                30               16
Celeus loricatus                    10               14
Automolus ochrolaemus               60               10
Xenops minutus                      40               54
Sclerurus guatemalensis             70               20
Dendrocincla fuliginosa             50               24
Deconychura longicauda              10                1
Glyphorhynchus spirurus             10                8
Xiphorhynchus lachrymosus           10               17
Thamnophilus atrinucha             100              212
Myrmotherula fulviventris          150              166
M. axillaris                       150              120
Microrhopias quixensis             100              170
Hylophylax naevioides               50               40
Myrmornis torquata                  60                4
Gymnopithys leucaspis               20               24
Formicarius analis                  80               30
Hylopezus perspicillata             40               42
Mionectes oleaginea                100               57
Oncostoma olivaceum                 40              160
Cnipodectes subbruneus              70               32
Rhynchocyclus olivaceus             70               24
Tolmomyias marginatus               90               31
Platyrinchus coronatus             100               20
Terenotriccus erythrurus           120               50
Myiobius sulphureipygius            60                3
Attila spadiceus                    10               14
Myiarchus tuberculifer              80                6
Querula purpurata                   5                14
Schiffornis turdinus                30                8
Pipra coronata                     200               36
P. mentalis                        400               96
Henicorhina leucosticta             10               18
Microcerculus marginatus            80                1
Cyphorhinus phaeocephalus          100               52
Microbates cinereiventris           10               17
Polioptila plumbea                  90               28
Hylophilus decurtatus              200              180
Vireolanius pulchellus             170               50
Pitylus grossus                     70               19
Cyanocompsa cyanoides               50                6
Species                   Difference (%) [+] Notes [++]
Tinamus major                    -33
Geotrygon montana                -83             A
Phaethornis superciliosus        +43             C
P. longuemareus                  +68             C
Trogon rufus                     -52
Trogon massena                     0
Baryphthengus martii             -20
Electron platyrhynchum             0
Notharchus pectoralis            -40
Malacoptila panamensis           +30
Ramphastos sulfuratus            +47             C
R. swainsonii                    +40             C
Melanerpes pucherani             +47
Celeus loricatus                 -29
Automolus ochrolaemus            +83
Xenops minutus                   -35
Sclerurus guatemalensis          +71
Dendrocincla fuliginosa          +52             C
Deconychura longicauda           +90             B
Glyphorhynchus spirurus          +20
Xiphorhynchus lachrymosus        -41
Thamnophilus atrinucha           -53
Myrmotherula fulviventris        -10
M. axillaris                     +20
Microrhopias quixensis           -35
Hylophylax naevioides            +20
Myrmornis torquata               +93             B
Gymnopithys leucaspis            -17             C
Formicarius analis               +62
Hylopezus perspicillata           -5
Mionectes oleaginea              +43             C
Oncostoma olivaceum              -75
Cnipodectes subbruneus           +54
Rhynchocyclus olivaceus          +66
Tolmomyias marginatus            +66
Platyrinchus coronatus           +80
Terenotriccus erythrurus         +58
Myiobius sulphureipygius         +95             B
Attila spadiceus                 -29
Myiarchus tuberculifer           +92             B
Querula purpurata                -64             C
Schiffornis turdinus             +73             C
Pipra coronata                   +82             C
P. mentalis                      +76             C
Henicorhina leucosticta          -44
Microcerculus marginatus         +99             B
Cyphorhinus phaeocephalus        +48
Microbates cinereiventris        -41
Polioptila plumbea               +69
Hylophilus decurtatus            +10
Vireolanius pulchellus           +71
Pitylus grossus                  +73
Cyanocompsa cyanoides            +88


(+.)Percentage difference was calculated by taking the absolute value of the difference between each pair of estimates and dividing by the higher estimate. Positive values indicate that the estimate derived from the 2-ha study was higher, whereas negative values indicate that the 2-ha estimates were lower than estimates from the 104-ha study.

(++.)A, highly temporally variable populations (Stouffer and Bierregaard 1993); B, very spatially patchy distribution; C, mobile species difficult to census.
                    Population densities and ecological
                     information for 242 bird species
                  recorded on the 104-ha Limbo study area
                  in Panama, January 1994-December 1995.
                                                           Density/100 ha
                                      Mass     Census           No.
Species [+]               Status [++] (g) [ss] method [II]   pairs [n]
Tinamidae
 Tinamus major                 R      1160     SM               15
 Crypturellus soui             R       250     SM               2
Ardeidae
 Tigrisoma lineatum            R       840     VO               1
 Butorides virescens           M       175     VO
 Agamia agami                  R       500     VO               0.5
 Cochlearius cochlearius       V       550     VO
Cathartidae
 Coragyps atratus              R      1800     VO
 Cathartes aura                R      1300     VO
 Sarcoramphus papa             R      3200     VO
Accipitridae
 Leptodon cayenensis           R       500     SM               0.5
 Elanoides forficatus          M       450     VO
 Harpagus bidentatus           R       185     SM               1.5
 Ictinia mississippiensis      M       275     VO
 Ictinia plumbea               M       247     VO
 Accipiter superciliosus       R       100     SM               1
 Accipiter bicolor             R       350     VO
 Geranospiza caerulescens      R       377     SM               0.5
 Leucopternis plunibea         R       482     VO
 Leucopternis semiplumbea      R       278     SM               1
 Leucopternis albicollis       R       736     VO
 Buteo platypterus             M       386     VO
 Buteo brachyurus              R       495     VO
 Buteo swainsonii              M       988     VO
 Spizaetus tyrannus            R      1005     VO               0.5
 Spizaetus ornatus             R      1305     VO
Falconidae
 Micrastur ruficollis          R       179     SM               2
 Micrastur mirandollei         R       450     SM               0.5
 Micrastur semitorquatus       R       650     SM               1
 Falco rufigularis             R       150     VO
Cracidae
 Ortalis cinereiceps           V       536     VO
 Crax rubra                    R      3800     SM               1
Phasianidae
 Odontophorus gujanensis       R       300     SM
Rallidac
 Aramides cajanea              R       405     SM               1.5
Eurypygidae
 Eurypyga helias               R       210     SM               2
Columbidae
 Columba speciosa              R       259     SM               12
 Columba nigrirostris          R       160     SM               10
 Claravis pretiosa [IIII]      R        69     SM               4, 1
 Leptotila cassinnii           R       155     SM              11
 Geotrygon veraguensis         R       155     SM               0.5
 Geotrygon violacea            R       102     SM               4
 Geotrygon montana             R       128     SM, MN          15
Psittacidae
 Brotogeris jugularis          R        63     SM X GC
 Pionopsitta haematosis        R       145     SM X GC
 Pionus menstruus              R       235     SM X GC
 Amazona autumnalis            R       416     SM X GC
 Amazona farinosa              R       687     SM X GC
Cuculidae
 Coccyzus americanus           M        60     VO
 Piaya cayana                  R       105     SM               12
 Dromococcyx phasianellus      R        86     SM               1
 Neomorphus geoffroyi          R       340     SM               1
                                       Biomass
                             No.       density               Terr.
Species [+]               inds. [#] (g/100 ha) Guild [++] size [++++]
Tinamidae
 Tinamus major                          34 800 F/G, T
 Crypturellus soui                        1000 F/G, T          B
Ardeidae
 Tigrisoma lineatum                       1680 Aq.
 Butorides virescens          +                Aq.
 Agamia agami                              500 Aq.
 Cochlearius cochlearius      +                Aq.
Cathartidae
 Coragyps atratus             2           3600 Carr.
 Cathartes aura               1           1300 Carr.
 Sarcoramphus papa            +                Carr.
Accipitridae
 Leptodon cayenensis                       500 R, D            G
 Elanoides forficatus         +                I, Aer.
 Harpagus bidentatus                       555 R, D            G
 Ictinia mississippiensis     +                I, Aer.
 Ictinia plumbea              +                I, Aer.
 Accipiter superciliosus                   200 R, D            G
 Accipiter bicolor            +                R, D
 Geranospiza caerulescens                  377 R, D            G
 Leucopternis plunibea        +                R, D            G
 Leucopternis semiplumbea                  556 R, D            G
 Leucopternis albicollis      +                R, D            G
 Buteo platypterus            +                R, D
 Buteo brachyurus             +                R, D
 Buteo swainsonii             +                R, D
 Spizaetus tyrannus                       1005 R, D            G
 Spizaetus ornatus            +                R, D
Falconidae
 Micrastur ruficollis                      716 R, D            F
 Micrastur mirandollei                     450 R, D            G
 Micrastur semitorquatus                  1300 R, D            F
 Falco rufigularis            +                R, D
Cracidae
 Ortalis cinereiceps          +                O, A, G
 Crax rubra                               7600 F/G, T
Phasianidae
 Odontophorus gujanensis      8           2400 O, T
Rallidac
 Aramides cajanea                         1215 I, T
Eurypygidae
 Eurypyga helias                           840 Aq.
Columbidae
 Columba speciosa                         6216 F, A
 Columba nigrirostris                     3200 F, A
 Claravis pretiosa [IIII]             552, 138 F/G, A
 Leptotila cassinnii                      3410 F, T            B
 Geotrygon veraguensis                     155 F, T
 Geotrygon violacea                        816 F, T            D
 Geotrygon montana                        3840 F, T
Psittacidae
 Brotogeris jugularis         3            189 F/G, A
 Pionopsitta haematosis       5            725 G, A
 Pionus menstruus             6           1410 G, A
 Amazona autumnalis          15           6240 G, A
 Amazona farinosa            10           6870 G, A
Cuculidae
 Coccyzus americanus          +                I, A, G
 Piaya cayana                             2520 I, A, G         B
 Dromococcyx phasianellus                  172 I, T, G
 Neomorphus geoffroyi                      680 I, AF/T, G      G
                             Plot
                            occup.
Species [+]               (%) [ssss]
Tinamidae
 Tinamus major                4
 Crypturellus soui            1
Ardeidae
 Tigrisoma lineatum           1
 Butorides virescens
 Agamia agami                 1
 Cochlearius cochlearius
Cathartidae
 Coragyps atratus             4
 Cathartes aura               4
 Sarcoramphus papa
Accipitridae
 Leptodon cayenensis          3
 Elanoides forficatus
 Harpagus bidentatus          4
 Ictinia mississippiensis
 Ictinia plumbea
 Accipiter superciliosus      4
 Accipiter bicolor
 Geranospiza caerulescens     3
 Leucopternis plunibea        1
 Leucopternis semiplumbea     2
 Leucopternis albicollis      2
 Buteo platypterus
 Buteo brachyurus
 Buteo swainsonii
 Spizaetus tyrannus           4
 Spizaetus ornatus
Falconidae
 Micrastur ruficollis         4
 Micrastur mirandollei        4
 Micrastur semitorquatus      4
 Falco rufigularis
Cracidae
 Ortalis cinereiceps
 Crax rubra                   4
Phasianidae
 Odontophorus gujanensis      1
Rallidac
 Aramides cajanea             2
Eurypygidae
 Eurypyga helias              1
Columbidae
 Columba speciosa             4
 Columba nigrirostris         4
 Claravis pretiosa [IIII]     1
 Leptotila cassinnii          3
 Geotrygon veraguensis        1
 Geotrygon violacea           2
 Geotrygon montana            4
Psittacidae
 Brotogeris jugularis         4
 Pionopsitta haematosis       4
 Pionus menstruus             4
 Amazona autumnalis           4
 Amazona farinosa             4
Cuculidae
 Coccyzus americanus
 Piaya cayana                 4
 Dromococcyx phasianellus     2
 Neomorphus geoffroyi         4
Strigidae
  Otus guatemalae           R 100   SM       4       800 R, N    15 3
  Lophostrix cristata       R 510   SM       1      1020 R, N       4
  Pulsatrix perspicillata   R 850   SM       1      1700 R, N     F 4
  Glaucidium minutissimum   R  61   SM            +      R, N
  Ciccaba nigrolineata      R 458   SM            +      R, N
Caprimulgidae
  Lurocalis semitorquatus   R  75   VO       1       150 I, Aer.    1
  Chordeiles minor          M  62   VO            +      I, Aer.
  Caprimulgus vociferans    M  50   VO            +      I, T, S
  Nyctidromus albicollis    R  53   SM       1       106 I, T, S    1
Nyctibiidae
  Nyctibius grandis         R 585   SM            2 1170 I, A, S    2
  Nyctibius griseus         R 185   SM            1  185 I, A, S    1
Apodidae
  Streptoprocne zonaris     R  80   VO            +      I, Aer.
  Chaetura pelagica         M  21   VO            +      I, Aer.
  Chaetura vauxi            M  18   VO            +      I, Aer.
  Chaetura brachyura        R  19   VO            +      I, Aer.
  Chaetura spinicauda       R  15   VO            +      I, Aer.
  Panyptila cayennensis     R  20   VO            +      I, Aer.
Trochilidae
  Threnetes ruckeri         R   6.1 VO            +      N
  Phaethornis guy           M   6   VO            +      N
  Phaethornis superciliosus R   6   MN, T        40  240 N          4
  Phaethornis longuemareus  R   3   T            16   48 N          4
  Eutoxeres aquila          M  11   VO            +      N
  Florisuga mellivora       R   6.3 VO           10   63 N          4
  Lophornis delattrei       R   3   VO            +      N
  Thalurania colombica      R   5   MN, T        30  150 N          4
  Damophila julie           R   4   MN, T        12   48 N          4
  Amazilia amabilis         R   3.3 MN, T        16   53 N          2
  Chalybura buffoni         R   6.0 VO            2   12 N          1
  Heliothryx barroti        R   6   VO            5   30 N          4
Trogonidae
  Trogon viridis            R  80   SM       6.5    1040 O, A     D 4
  Trogon violaceus          R  57   SM       8       912 O, A    10 4
  Trogon rufus              R  52   SM      21      2184 O, A     4 4
  Trogon melanurus          R 115   SM       3.5     805 O, A    12 3
  Trogon massena            R 140   SM      20      5600 O, A     C 4
Momotidae
  Momotus momota            V 105   VO            +      O, A
  Baryphthengus martii      R 162   SM      24      7776 O, A     C 3
  Electron platyrhynchum    R  62   SM      15      1860 O, A     5 3
Alcedinidae
  Chloroceryle inda         R  59   SM       1       118 Aq.        1
  Chloroceryle aenea        R  16   SM       2        64 Aq.        2
Buconidae
  Notharchus macrorhynchus  R  96   SM       1       192 I, A, S    1
  Notharchus pectoralis     R  68   SM      14      1904 I, A, S    4
  Notharchus tectus         R  33   SM       0.5      33 I, A, S    1
  Malacoptila panamensis    R  44   SM, MN  14      1232 I, A, S    2
Galbulidae
  Jacamerops aurea          R  63   SM       0.5      63 I, A, S    2
Ramphastidae
  Pteroglossus torquatus    R  65   SM X GC       8  520 O, A       4
  Ramphastos sulfuratus     R 375   SM X GC      16 6000 O, A       4
  Ramphastos swainsoni      R 750   SM X GC      12 9000 O, A       4
Picidae
  Melanerpes pucherani      R  54   SM           16  864 I, B, I  8 3
  Melanerpes rubricapillus  V  53   VO            +      I, B, I
  Celeus loricatus          R  74   SM           14 1036 I, B, I 25 4
  Dryocopus lineatus        R 180   SM       1.5     540 I, B, I  G 3
  Campephilus haematogaster R 224   VO            +      I, B, I
  Campephilus melanoleucos  R 225   SM       2       900 I, B, I 40 4
Furnariidae
  Automolus ochrolaemus      R 40 SM       5       400 I, A, DL   11 3
  Xenops minutus             R 11 SM      27       594 I, B, S     5 4
  Sclerurus mexicanus        R 26 SM            1   26 I, T, G       1
  Sclerurus guatemalensis    R 34 SM      10       680 I, T, G     7 3
Dendrocolaptidae
  Dendrocincla fuliginosa    R 41 SM      12       984 I, AF/B, S  6 4
  Dendrocincla homochroa     R 41 SM            1   41 I, AF/B, S    2
  Deconychura longicauda     R 24 SM       0.5      24 I, B, S       1
  Glyphorynchus spirurus     R 15 SM, MN        8  120 I, B, S     D 2
  Dendrocolaptes certhia     R 68 SM       3       408 I, AF/B, S 10 2
  Xiphorhynchus guttatus     R 47 SM      18      1692 I, B, S     C 4
  Xiphorhynchus lachrymosus  R 51 SM       8.5     867 I, B, S     6 3
Formicariidae
  Cymbilaimus lineatus       R 37 SM      25      1850 I, A, G     2 3
  Thamnophilus atrinucha     R 22 SM     106      4664 I, A, G     1 4
  Thamnistes anabatinus      R 20 SM            +      I, A, G
  Dysithamnus puncticeps     R 15 SM       3.5     105 I, A, G     A 1
  Myrmotherula brachyura     R  7 SM      21       294 I, A, G     1 2
  Myrmotherula fulviventris  R 10 SM      83      1660 I, A, DL    1 4
  Myrmotherula axillaris     R  8 SM      60       960 I, A, G     2 4
  Microrhopias quixensis     R  8 SM      85      1360 I, A, G     1 4
  Cercomacra tyrannina       R 17 SM      13       442 I, A, DL    1 1
  Myrmeciza exsul            R 27 SM      13       702 I, A/T, G   7 3
  Hylophylax naevioides      R 17 SM      20       680 I, AF/T, G  4 4
  Myrmornis torquata         R  4 SM       2       192 I, T, G     E 2
  Gymnopithys leucaspis      R 30 SM      12       720 I, AF      25 4
  Phaenostictus mcleannani   R 51 SM      48       408 I, AF       F 4
  Formicarius analis         R 57 SM      15      1710 I, T, G     5 4
  Hylopezus perspicillata    R 42 SM      21      1764 I, T, G     B 4
Tyrannidae
  Zimmerius vilissimus       R  9 SM, T   55       990 0, A, S     A 3
  Ornithion bruneicapillum   R  7 SM      19       266 I, A, S     B 3
  Tyrannulus elatus          R  8 SM      13       208 0, A, S     4 2
  Myiopagis gaimardii        R 14 SM      11.5     322 I, A, S     3 2
  Myiopagis caniceps         R 11 SM       3        66 I, A, S     C 2
  Mionectes olivaceus        R 16 MN            +      0, A, S
  Mionectes oleaginea        R 13 SM, MN       57  741 0, A, S       4
  Myiornis atricapillus      R  5 SM       3.5      35 I, A, S     3 1
  Oncostoma olivaceum        R  7 SM, T   80      1120 I, A, S     A 4
  Todirostrum nigriceps      R  6 SM       0.5       6 I, A, S       1
  Cnipodectes subbrunneus    R 23 SM      16       736 I, A, S     C 3
  Rhynchocyclus olivaceus    R 22 SM      12       528 I, A, S     C 3
  Tolmomyias assimilis       R 14 SM      15.5     434 I, A, S     4 3
  Platyrinchus coronatus     R  9 SM, T   10       180 I, A, S     6 2
  Onychorhynchus coronatus   R 15 VO       1.5      45 I, A, S       1
  Terenotriccus erythrurus   R  7 SM      25       350 I, A, S     B 3
  Myiobius sulphureipygius   R 12 SM, T         3   36 I, A, S     B 1
  Contopus virens            M 14 SM            +      I, A, S
  Empidonax flaviventris     M 11 SM            +      I, A, S
  Empidonax virescens        M 12 SM           24  288 I, A, S     B 3
  Empidonax alnorum          M 12 SM            +      I, A, S
  Colonia colonus            R 16 SM       3        96 I, A, S     B 1
  Attila spadiceus           R 38 SM       7       532 I, A, S     8 2
  Laniocera rufescens        R 49 SM           12  588 0, A, S       1
  Rhytipterna holerythra     R 38 SM       9       684 0, A, S       2
  Myiarchus tuberculifer     R 20 SM       3       120 I, A, S       1
  Myiarchus crinitus         M 34 SM            8  272 I, A, S       2
  Conopias parva             R 22 SM            +      I, A, S       1
  Myiodynastes maculatus     V 45 VO            +      I, A, S
  Myiodynastes luteiventris  M 47 VO            +      I, A, S
  Legatus leucophaius        M 26 SM            1   26 0, A, S       1
  Tyrannus tyrannus          M 38 VO            +      0, A, S
  Pachyramphus polychopterus V 18 SM            +      I, A, S
  Tityra semifasciata        R 80 VO       1       160 0, A, S       1
  Tityra inquisitor          R 41 VO            +      0, A, S
Cotingidae
  Lipaugus unirufus           R  86 SM      0.5        86 O, A, S      1
  Cotinga nattererii          R  55 VO           +        F, A
  Querula purpurata           R 104 SM          14   1456 O, A, S   16 4
Pipridae
  Schiffornis turdinus        R  33 SM      4         264 O, A, S      2
  Manacus vitellinus          R  17 SM, MN       6    102 F, A, S      1
  Pipra coronata              R  10 SM, MN      36    360 F, A, S      3
  Pipra mentalis              R  15 SM, MN      96   1440 F, A, S      4
Hirundinidae
  Progne chalybea             V  39 VO           +        I, Aer.
  Neochelidon tibialis        R   9 VO           +        I, Aer.
  Stelgidopteryx serripennis  M  16 VO           +        I, Aer.
  Hirundo pyrrhonota          M  20 VO           +        I, Aer.
  Hirundo rustica             M  18 VO           +        I, Aer.
  Tachycineta bicolor         M  21 VO           +        I, Aer.
Corvidae
  Cyanocorax affinis          R 205 SM           +        O, A, G
Troglodytidae
  Thryothorus fasciatoventris R  23 SM      6         276 I, A, DL   B 2
  Thryothorus nigricapillus   R  22 VO           +        I, A, G
  Henicorhina leucosticta     R  17 SM      9         306 I, T, G    A 2
  Microcerculus marginatus    R  19 SM      0.5        19 I, T, G      1
  Cyphorhinus phaeocephalus   R  25 SM     26        1300 I, T, G    1 3
Sylviinae
  Microbates cinereiventris   R  12 SM      8.5       204 I, A, G    B 2
  Ramphocaenus rufiventris    R  10 SM     22         440 I, A, G    A 2
  Polioptila plumbea          R   7 SM     14         196 I, A, S    B 3
Turdinae
  Catharus minimus            M  27 MN           +        O, A, G/S
  Catharus ustulatus          M  30 MN           +        O, A, G/S
  Hylocichla mustelina        M  46 MN           4    184 I, T, G      1
  Turdus assimilis            M  68 VO           +        O, A, G/S
Vireonidae
  Vireo flavifrons            M  18 SM           0.5    9 I, A, G      1
  Vireo philadelphicus        M  11 VO           +        I, A, G
  Vireo olivaceus             M  17 VO, T        +        O, A, G
  Hylophilus ochraceiceps     R  10 VO           +        I, A, G
  Hylophilus decutatus        R  10 SM     90        1800 I, A, G    A 4
  Vireolanius pulchellus      R  25 SM     25        1250 I, A, G    5 4
Parulinae
  Vermivora peregrina         M   9 VO           3     27 O, A, G/S    4
  Dendroica Pensylvanica      M  10 SM, T        7     70 O, A, G/S    4
  Dendroica fusca             M  10 VO           +        O, A, G/S
  Dendroica castanea          M  10 SM, T       10    100 O, A, G/S    4
  Dendroica cerulea           M   9 VO           +        I, A, G
  Mniotilta varia             M  11 VO           +        I, A, G
  Helmitheros vermivorus      M  13 VO           1     13 I, A, DL     1
  Seiurus aurocapillus        M  19 VO           1     19 I, T, G      1
  Seiurus noveboracensis      M  17 VO           +        I, T, G
  Seiurus motacilla           M  19 SM           2     38 I, T, G      1
  Oporornis formosus          M  14 SM          13    182 I, T, G    B 1
  Wilsonia citrina            M   9 VO           +        I, A, G
  Wilsonia canadensis         M  10 VO           +        I, A, G
Thraupinae
  Tangara inornata            R  18 VO           4     72 O, A, G      4
  Tangara gyrola              R  22 VO           1     22 O, A, G      2
  Tangara larvata             R  19 VO, T   1.5        57 O, A, G      4
  Dacnis venusta              R  16 VO           +        O, A, G
  Dacnis cayana               R  13 VO, T       20    360 O, A, G      4
  Chlorophanes spiza          R  18 VO, T        9    162 O, A, G      4
  Cyanerpes lucidus           R  12 VO, T       12    144 F, A, G      4
  Cyanerpes cyaneus           R  13 VO, T        8    104 F, A, G      4
  Euphonia luteicapilla       V  12 VO           +        F, A, G
  Euphonia laniirostris       R  15 SM      1          30 F, A, G      2
  Euphonia fulvicrissa        R  11 SM     10         220 O, A, G    1 2
 Euphonia minuta              R   11    VO, T 0.5               11 F, A, G
 Eucometis penicillata        R   30    SM    2                120 0, AF/A, G
 Tachyphonus luctuosus        R   15    SM          42         630 0, A, G/S
 Tachyphonus delatrii         R   18    VO          +              F, A, G
 Habia fuscicauda             R   39    SM          3          117 0, A, G
 Piranga rubra                M   30    VO          1           30 0, A, S
 Piranga olivacea             M   30    VO          +              0, A, G
 Ramphocelus dimidiatus       V   28    VO          +              0, A, G
 Ramphocelus flammigerus      V   34    VO          +              0, A, G
Cardinalinae
 Pitylus grossus              R   43    SM    9.5              817 0, A, G
 Pheucticus ludovicianus      M   45    VO          +              O, A, G
 Cyanocompsa cyanoides        R   32    SM    3                192 O, A, G
Emberizinae
 Arremon aurantiirostris      R   31    VO          +              O, T, G
 Arremonops conirostris       V   41    VO          +              O, T, G
 Sporophila schistacea [IIII] R   11    SM        20, 12  220, 132 O, A, G
 Oryzoborus funereus          V   12    VO          +              G, A, G
Icterinae
 Scaphidura oryzivora         R   187   VO          +              O, T, G
 Icterus chrysater            V   43    VO          +              O, A, G
 Cacicus uropygialis          R   57    SM, T       20        1140 O, A, G
 Cacicus cela                 R 68-113  VO, T       1          110 O, A, G
 Psarocolius wagleri [IIII]   R 113-214 VO, T     21, 6  2500, 779 O, A, G
 Euphonia minuta                 1
 Eucometis penicillata        G  3
 Tachyphonus luctuosus        B  2
 Tachyphonus delatrii
 Habia fuscicauda             25 1
 Piranga rubra                   1
 Piranga olivacea
 Ramphocelus dimidiatus
 Ramphocelus flammigerus
Cardinalinae
 Pitylus grossus              C  3
 Pheucticus ludovicianus
 Cyanocompsa cyanoides        E  2
Emberizinae
 Arremon aurantiirostris
 Arremonops conirostris
 Sporophila schistacea [IIII]    1
 Oryzoborus funereus
Icterinae
 Scaphidura oryzivora
 Icterus chrysater
 Cacicus uropygialis          D  4
 Cacicus cela                 G  4
 Psarocolius wagleri [IIII]   G  4


Note: Additional vagrant and migrant species detected but not observed using the study area or the space above it for foraging: Phalacrocorax olivaceus, Anhinga anhinga, Fregata magnificens, Pandion haliaetus, Chondrohierax uncinatus, Buteogallus urubitingo, Buteo magnirostris, Falco peregrinus, Columba cayennensis, and Ceryle torquata.

(+.)Species names and sequence follow Ridgely and Gwynne (1989).

(++.)M, migrant; R, resident; V, vagrant.

(ss.)Data from mist-net captures at Limbo, and from Stiles and Skutch (1989) and Karr et al. (1978, 1990).

(II.)MN, mist-netting; SM, spot mapping; SM X GC, spot mapping supplemented by counts of average number of birds in groups; T, transect counts used to supplement spot mapping; VO, visual observation (e.g., counts at flowering or fruiting trees, or nesting colonies).

(n.)Number of breeding pairs estimated to occur on the plot.

(#.)No. individuals/100 ha, estimated for species for which identification of classical, pair-occupied territories was problematic, (e.g., parrots and hummingbirds). Plusses (+) designate species for which no estimate was made.

(++.)Assignments are based on personal observations and data in Karr et al. (1990); categories are those of Terborgh et al. (1990) to facilitate comparisons among the two studies. Abbreviations are: Aq., aquatic species found primarily along forest streams or, for some vagrants, near larger bodies of water; Carr., carrion consumers; F, A, arboreal frugivores; F, T, terrestrial frugivores; G, A, arboreal granivores; G, T, terrestrial granivores; I, Aer., aerial insectivores that capture and consume insects while in flight); I, AF, army ant followers; I, A, DL, arboreal insectivores that search primarily dead leaf clusters; I, A, G, gleaning arboreal insectivores; I, A, S, sallying arboreal insectivores; I, B, I, insectivores that extract food from the interior of bark substrates (e.g., woodpeckers); I, B, S, insectivores that glean food from bark surface (e.g., woodcreepers); I, T, G, gleaning terrestrial insectivores (e.g., leaftossers); I, T, S, sallying terrestrial insectivores (e.g., common pauraque); N, nectarivores; most species also consume some small arthropods; O, A, arboreal omnivores; O, T, terrestrial omnivores; R, D, raptors, diurnal; R, N, raptor, nocturnal.

(++++.)Average territory size (ha) is presented for some species, but for most species, measurements were on a categorical scale: A, 0-2 ha; B, 3-4; C, 5-8; D, 9-16; E, 17-32; F, 33-64; G, [greater than]64 ha.

(ssss.)Percentage of plot occupied: 1, 0-25%; 2, 26-50%; 3, 5 1-75%; 4, 76-100%.

(IIII.)For species with large differences in abundance among years, the 1994 estimate is listed first, then the 1995 estimate. These include two species responding to the availability of seeding bamboo and one colonial icterid.
COPYRIGHT 2000 Ecological Society of America
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2000 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:ROBINSON, W. DOUGLAS; BRAWN, JEFFREY D.; ROBINSON, SCOTT K.
Publication:Ecological Monographs
Geographic Code:2PANA
Date:May 1, 2000
Words:20540
Previous Article:IMPACTS OF ROOT COMPETITION IN FORESTS AND WOODLANDS: A THEORETICAL FRAMEWORK AND REVIEW OF EXPERIMENTS.
Next Article:STRONG IMPACTS OF GRAZING AMPHIPODS ON THE ORGANIZATION OF A BENTHIC COMMUNITY.
Topics:

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |