Continuous Forest Sites Research Articles

Incidence of Spruce Grouse in Residual Forest Strips within Large Clear-cut Boreal Forest Landscapes

Falcipennis canadensis (Spruce Grouse) are a habitat specialist of coniferous forest in North America and typically have large home ranges (≈ 25 ha, range = 22–75 ha). Although this species is sensitive to logging, few studies report abundance and population trends in relation to habitat loss and fragmentation. In early May of 1997 to 1999, we surveyed Spruce Grouse in 21 riparian and 27 upland residual forest strips (width = 51–132 m) located within six clear-cut landscapes (23–256 km2) of southcentral Québec, Canada. We also surveyed five control sites in uncut continuous forest surrounding one landscape. Contrary to our prediction that Spruce Grouse would not be observed from such narrow strips, they were detected in 24 of the 48 strips. Both types of strips had similar occurrence rates and abundance indices (2.01 ± 0.46 [SE] males/10 stations in riparian strips, 1.29 ± 0.35 males/10 stations in upland strips). Because of the age of adjacent cuts (mean = 4.2 years, range = 2–8 years), crowding of grouse that were present at logging time into forest strips does not adequately explain the numbers of Spruce Grouse in these structures. Therefore, we conclude that Spruce Grouse observed in forest strips have either immigrated from outside or were local recruits. Despite persistence of Spruce Grouse in residual forests of logged landscapes, we suggest a cautious approach to forest management since the species' long-term persistence could be jeopardized if residual forest strips are harvested.

Composition and Abundance of Ground-Dwelling Coleoptera in a Fragmented and Continuous Forest

The structure of the ground-dwelling beetle community at two forested sites in northern Delaware was compared by examining composition and abundance at the species or genus level for five target families and at the family level for all other Coleoptera. The beetle communities at an isolated 15-ha woodlot and a nearby plot of relatively continuous forest were sampled using pitfall traps during 2000 and 2001. The results showed an absence or reduced abundance of many beetle taxa in the isolated woodlot relative to the continuous forest site and elevated abundances of a few taxa. Most notably, the species richness of poorly dispersing species of carabid beetles was lower in the isolated woodlot, with 12 of 19 brachypterous species completely absent from pitfall samples. Although this study does not constitute a test of habitat fragmentation hypotheses, the data do suggest that the beetle community in the woodlot may be suffering from isolation effects. The estimated response to fragmentation was correlated with body size for the 19 most common carabid species caught, suggesting that body size may be a useful predictor of vulnerability to forest fragmentation. At least in Carabids, larger body size seems to be associated with greater susceptibility to fragmentation. A number of staphylinid beetle taxa were also significantly less abundant or absent in the isolated woodlot, as were some beetles in other trophic groups, including carrion beetles (Silphidae) and geotrupid beetles (Geotrupidae). Although it retains some faunal similarity with nearby continuous forest, the woodlot may be too isolated to sustain populations of some beetle taxa, especially large, poorly dispersing forest species.

Can forest fragments reset physical and water quality conditions in agricultural catchments and act as refugia for forest stream invertebrates?

Forest fragments embedded within agricultural landscapes have the potential to provide a “forest reset effect” by mitigating agricultural effects on water quality, and acting as refugia and conservation reserves for aquatic species. We investigated the ability of forest fragments to reset agricultural effects using four catchments in the South Island, New Zealand. Two catchments were dominated by agricultural activities, but each had an isolated forest fragment in the lower valley, and two catchments had continuous riparian forest along the valley floor. Riffles sampled in continuous forest were generally deeper than those in agricultural and forest fragments, and not surprisingly streams in forest fragments and continuous forest received less light than those in agricultural land. All sites had circum-neutral pH, but both conductivity and temperature were significantly lower at continuous forest sites than agricultural and forest fragment sites. Taxonomic richness, Margalef’s index and numbers of Ephemeroptera, Plecoptera and Trichoptera (EPT) taxa were significantly higher in continuous forest than at forest fragment sites, but overall invertebrate densities did not differ between fragments and continuous forest. Several taxa were abundant at agricultural and forest fragment sites, but absent or at low densities in continuous forest. They included the blackfly Austrosimulium spp. and two caddisflies Pycnocentrodes sp. and Hydrobiosis parumbripennis. Conversely, the mayflies Austroclima sp. and Coloburiscus humeralis and the blepharicerid Neocurupira chiltoni were either restricted to continuous forest, or abundant in continuous forest but rare in agricultural and forest fragments. An ordination of communities separated those in agricultural and continuous forest sites, but communities at forest fragment sites were clustered among the agricultural sites. In this study forest fragments of 5–7 ha, located in the lower reaches of the catchment did not mitigate the negative upstream effects of agriculture on stream functioning. Fragment size (or riparian forest length), riparian forest width and vegetation type, and fragment location in the catchment may have critical roles in enabling forest fragments to reset the negative impacts of agriculture. Determining these characteristics of fragments has important consequences for stream remediation.

Long‐term effects of forest fragmentation on Amazonian ant communities

AbstractAim To analyse the effects of forest fragmentation on ant communities in an Amazonian landscape that has been fragmented for over a century.Location The region surrounding the village of Alter do Chão in the Brazilian Amazonian state of Pará (2°30′ S, 54°57′ W).Methods Collection of ants and measurements of tree density were performed along transects established in eight sites in continuous forest and in 24 forest fragments surrounded by savanna vegetation. Data on size, perimeter, and degree of isolation (distance to continuous forest and distance to nearest area of forest > 5 ha) of each fragment were obtained from a georeferenced Landsat image of the study area.Results There were significant differences in species richness and composition between fragments and continuous forest, and these differences were not related to intersite variation in vegetation structure (tree density). Fragments supported fewer ant species per plot, and these species tended to represent a nested subset of those found in continuous forests. Fragments had significantly fewer rare species and fewer ant genera. However, fragments and continuous forest had similar numbers of species that also occur in the savanna matrix (i.e. that are not forest specialists). Multiple linear regression analyses indicated that species richness and composition in the fragments are significantly affected by fragment area, but not by fragment shape and degree of isolation. More species were found in larger fragments.Main conclusions Forest fragmentation influences the organization of ant communities in Amazonian savanna/forest landscapes. Forest fragments harboured, on average, 85% of the species found in continuous forest. That these fragments, despite their long history of isolation, support a relatively large complement of the species found in continuous forest is surprising, especially given that in some recently fragmented landscapes the proportion of species surviving in the fragments is lower. Differences in inter‐fragment distance and type of matrix between Alter do Chão and these other landscapes may be involved. The fact that fragments at Alter do Chão are surrounded by a natural (rather than an anthropogenic) habitat, and that most of them are less than 300 m from another forest area, may have helped to ameliorate the adverse effects of forest fragmentation.

Nestedness in fragmented landscapes: a case study on birds, arboreal marsupials and lizards

AbstractAim The potential nestedness of assemblages of birds, arboreal marsupials and lizards was examined in a fragmented landscape in south‐eastern Australia. We assessed which ecological processes were related to the presence or absence of nestedness, particularly in relation to previous autoecological studies in the same study area.Location Data were collected at Buccleuch State Forest, c. 100 km to the west of the Australian Capital Territory in south‐eastern Australia.Methods Presence/absence matrices were compiled for birds (40 pine sites, 40 continuous forest sites, 43 fragments), arboreal marsupials (41 continuous forest sites, 39 fragments) and lizards (30 sites including all landscape elements) from a range of field surveys conducted since 1995. Nestedness was analysed using a standardized discrepancy measure, and statistical significance was assessed using the RANDNEST null model. For birds, species thought to be extinction‐prone were analysed separately to assess if assemblages comprising extinction‐prone species were more strongly nested than others. Also, sites with a substantial amount of Eucalyptus radiata were analysed separately to assess whether nestedness was stronger if environmental heterogeneity was minimized.Results The assemblages of lizards and arboreal marsupials were not nested, probably because of qualitative differences between species in response to environmental conditions. The assemblages of birds in fragments and pine sites were significantly nested, but nestedness was substantially stronger in fragments. For birds, nestedness appeared to be related to somewhat predictable extinction sequences, although there were many outliers in the analysis. Nestedness increased when extinction‐prone species were analysed by themselves. Nestedness decreased when environmental heterogeneity was minimized by including only sites dominated by E. radiata.Main conclusions In a given landscape, different vertebrate assemblages can respond in vastly different ways to fragmentation. Nestedness analyses can provide a useful overview of likely conservation issues in fragmented landscapes, for example by highlighting the possible roles of local extinction and immigration. However, nestedness analyses are a community‐level tool, and should be complemented by more detailed autoecological studies when applied in a conservation context.

Decreased relatedness between male prickly forest skinks (Gnypetoscincus queenslandiae) in habitat fragments

In species with low levels of dispersal the chance of closely related individuals breeding may be a potential problem; sex-biased dispersal is a mechanism that may decrease the possibility of cosanguineous mating. Fragmentation of the habitat in which a species lives may affect mechanisms such as sex-biased dispersal, which may in turn exacerbate more direct effects of fragmentation such as decreasing population size that may lead to inbreeding depression. Relatedness statistics calculated using microsatellite DNA data showed that rainforest fragmentation has had an effect on the patterns of dispersal in the prickly forest skink (Gnypetoscincus queenslandiae), a rainforest endemic of the Wet Tropics of north eastern Australia. A lower level of relatedness was found in fragments compared to continuous forest sites due to a significantly lower level of pairwise relatedness between males in rainforest fragments. The pattern of genetic relatedness between sexes indicates the presence of male-biased dispersal in this species, with a stronger pattern detected in populations in rainforest fragments. Male prickly forest skinks may have to move further in fragmented habitat in order to find mates or suitable habitat logs.

Dung beetle community (Coleoptera: Scarabaeidae: Scarabaeinae) in a tropical landscape at the Lachua Region, Guatemala

Biological diversity conservation within natural reserves has been prioritized, but conservation efforts outside protected areas (where most human activities take place) have been very little considered. In this scenario, an alternative agricultural practice that may reduce the impacts of fragmentation in outer landscapes is a perforation process, which involves conservation in agricultural fields surrounded by continuous forests. Such practices enhance the positive impact of ecological services on fields. In this study we analyzed the biological diversity state in perforation fields and their surrounding forests. The analysis was done using dung beetles as biological indicators. A nested pattern in dung beetles distribution was found, which ordered the surrounding continuous forest sites as the ones with the highest species richness, followed by the perforation fields, and placed the fragmentation practice fields (continuous agricultural fields surrounding forest patches) with the lowest one. Indicator species for perforation fields and surrounding continuous forests were chosen. In general, perforation practice fields differed in composition, based upon functional groups richness and identity; it also contained a higher species richness than the fragmentation practice. Agricultural practices that enhance biological diversity conservation such as perforation, should be recommended and considered in natural resource management by local communities in order to take advantage of ecological services that otherwise may be gradually lost.

The effect of habitat fragmentation on communities of mutualists: Amazonian ants and their host plants

The consequences of fragmentation for communities of mutualist partners are for the most part unknown; moreover, most studies addressing this issue have been conducted on plant–pollinator communities. We evaluated how the experimental fragmentation of lowland Amazonian rain forest influenced a community of ant–plant mutualists. We inventoried a total of 1057 myrmecophytes in four fragments and four continuous forest sites; the twelve plant species recorded were occupied by 33 ant morphospecies, of which 11 were obligate plant inhabitants. Neither plant species richness, ant species richness, nor total ant–plant density were significantly lower in forest fragments. However, eight of the plant species, including three of the four most common, had higher mean densities in continuous forest than fragments. Of these four species, only one ( Cordia nodosa) had significantly different colonization rates between habitat types, with higher colonization rates of plants in fragments. This may be because the Azteca species it is associated with increases in abundance in forest isolates. Although our results suggest that communities of ant–plant mutualists are likely to persist in fragmented tropical landscapes 25 years after fragment isolation, most species are rare and populations sizes in fragments are extremely low. Environmental and demographic stochasticity could therefore limit long-term population viability. We suggest future studies focus on evaluating how fragmentation has altered herbivore pressure and the dispersal of ants and plants to fragments, since the interaction of these factors is likely to have the greatest impact on long-term patterns of population persistence.

Genetic diversity of the tropical tree Terminalia amazonia (Combretaceae) in naturally fragmented populations.

The effect of long-term fragmentation on the genetic diversity of populations of the neotropical tree species, Terminalia amazonia, was studied using random amplified polymorphic DNA (RAPD) analysis. Leaf material from 104 trees was collected from three naturally fragmented gallery forest patches and three plots in nearby continuous forest in the Mountain Pine Ridge, Belize. In total, 30 RAPD bands generated by five decamer primers were used to compare the genetic diversity of the six populations in the two groups. Genetic variation within the populations (H0), as estimated by the Shannon diversity index, ranged from 0.32 to 0.38, with an overall diversity of 0.38 (Hspecies). Analysis of molecular variation revealed that most (94.4%, P<0.001) of the variation was attributable to differences among individuals within populations. Population differentiation was significantly (P=0.038) lower among the fragmented populations than among continuous forest populations. On average, the fragmented populations also had slightly, but statistically significant (P=0.046) lower levels of genetic diversity. However, one gallery forest site had a higher level of genetic diversity than two of the continuous forest sites. We suggest that the long-term effect of fragmentation on the genetic diversity of tropical trees will depend upon the amount of local forest cover in proximity to the fragmented populations.

Explaining bird species composition and richness in eucalypt‐dominated remnants in subhumid Tasmania

AbstractAim To determine the factors influencing the distribution of birds in remnants in a fragmented agricultural landscape.Location Forty‐seven eucalypt remnants and six sites in continuous forest in the subhumid Midlands region of Tasmania, Australia.Methods Sites were censused over a two‐year period, and environmental data were collected for remnants. The avifauna of the sites was classified and ordinated. The abundances of bird species, and bird species composition, richness, abundance and diversity were related to environmental variables, using simple correlation and modelling.Results There were two distinct groups of sample sites, which sharply differed in species composition, richness, diversity and bird abundance, separated on the presence/absence of noisy miner (Manorina melanocephala Latham) colonies, remnant size, vegetation structural attributes and variables that reflected disturbance history. The approximate remnant size threshold for the change from one group to another was 20–30 ha. Remnant species richness and diversity were most strongly explained by remnant area and noisy miner abundance, with contributions from structural and isolation attributes in the second case. Segment richness was explained by precipitation, logging history and noisy miner abundance. Bird abundance was positively related to precipitation and negatively related to tree dieback. The 28 individual bird species models were highly individualistic, with vegetation structural variables, noisy miner abundance, climatic variables, variables related to isolation, area, variables related to floristics, disturbance variables, the nature of the matrix and remnant shape all being components in declining order of incidence. Age of the remnant did not relate to any of the dependent variables.Main conclusions Degraded and small remnants may have become more distinct in their avifaunal characteristics than might otherwise be the case, as a result of the establishment of colonies of an aggressive native bird, the noisy miner. The area, isolation and shape of remnants directly relate to the abundance of relatively few species, compared to vegetation attributes, climate and the abundance of the noisy miner. The nature of the matrix is important in the response of some species to fragmentation.

Effect of forest fragmentation on dung beetle communities and functional consequences for plant regeneration

I compared dung beetle communities and assessed some of their functional effects (dung removal, seed burial, seedling establishment) in continuous forest with those in 1‐ha and 10‐ha forest fragments in Central Amazonia. I followed the fate of seeds until seedling establishment for three native tree species, using clean seeds and seeds surrounded by dung. The 1‐ha fragments had half the number of dung beetle species captured in continuous forest and in 10‐ha fragments. The continuous forest sites and the 1‐ha fragments had similar number of individuals, but in the 10‐ha fragments dung beetles were twice as abundant. Mean beetle size increased with increasing forest area. Dung removal and seed burial rates were higher in continuous forest than in forest fragments. Seed predation rates were higher in the forest fragments. In all sites, the proportion of seedlings established from seeds surrounded by dung vs clean seeds was the same, and it was the same in continuous forest vs fragments. When comparing seeds that remained on the forest floor with seeds buried by dung beetles, a higher percentage of seedlings established from the latter. Conservation programs that aim to maintain the regeneration ability of forest fragments must incorporate all the important components involved in seedling establishment; in Central Amazonia these include dung beetles as secondary dispersers. It is important that studies start measuring directly not only the first‐order effects of forest fragmentation on species, but also the higher‐order functional effects.

Experimental assessment of Heliconia acuminata growth in a fragmented Amazonian landscape

Summary Fragmentation severely alters physical conditions in forest understories, but few studies have connected these changes to demographic impacts on forest species using detailed experimental examination at the individual and population levels. Using a 32‐month, reciprocal‐transplant experiment, we show that individuals of the Amazonian understory herb Heliconia acuminata transplanted into forest fragments lost over 20% of their vegetative shoots, while those transplanted to continuous forest showed a slight gain. The leaf area of plants in fragments also increased at half the rate it did in continuous forest sites. It appears that the normal dry season stresses to which forest understorey plants are exposed are greatly exacerbated in fragments, causing plants to shed shoots and leaves. The observed shifts in size could help explain why populations in fragments are more skewed towards smaller demographic stage classes than those in continuous forest. These shifts in size structure could also result in reduced abundances of flowering plants, as reproduction in H. acuminata is positively correlated with shoot number. Fragmentation‐related changes in growth rates resulting from abiotic stress may have significant demographic consequences.

Bat Species Composition in Three Localities in the Amazon Basin

We compared the bat species composition of three localities in the Brazilian Amazon: continuous forest and fragments at Alter do Chão, Santarém, an area not previously sampled, and continuous forest in two relatively well studied areas, Manaus and Belém. The number of species captured at each locality varied between 17 and 42. Most species occurred at high abundance in continuous forest in the three localities, but only a subset was common in the fragments at Santarém. The relative abundance of bat species in Santarém is different from other localities sampled in the Amazon Basin and the presence of forest fragments may be the most important factor contributing to this difference. Multidi-mensional scaling ordination of sites based on relative abundance indicated distinct assemblages of bat species in the forest fragments in Santarém, and little differentiation of continuous-forest sites from sites in other localities. Continuous forests at Santarém and near Belém were more similar to each other than to continuous forest near Manaus, indicating the possibility of an east-west gradient in bat communities in the Amazon.

Landscape and Fragment Size Effects on Reproductive Success of Forest-Breeding Birds in Ontario

To determine the minimum size requirements and influence of landscape context on reproductive success of forest-breeding songbirds, we monitored nesting success of five species of songbirds on 40 fragments (12–2350 ha in total woodlot size) and two continuous forest sites in south-central Ontario from 1994 through 1997. Woodlot size was the most important variable contributing to differences in reproductive success, with local forest cover (within a 10-km radius) having no significant additional effect on productivity for any species. For all species, except Red-eyed Vireo (Vireo olivaceous), adult female reproductive success was at or above replacement levels in large fragments (mean of 121 ha core area, 849 ha woodlot area) and continuous forest, and below replacement levels in small fragments (mean of 7.8 ha core area, 93 ha woodlot area). Red-eyed Vireo productivity was particularly low, with single-brooded females unable to maintain populations in any woodlots monitored, although populations were close to replacement levels in continuous forest. Only the very largest fragments acted as sources for Ovenbird (Seiurus aurocapillus) and Wood Thrush (Hylocichla mustelina) (i.e., >23 ha in core area, 225 ha in total woodlot area). Ovenbird and Wood Thrush were the most area-sensitive, and Rose-breasted Grosbeak (Pheucticus ludovicianus) and Veery (Catharus fuscens) the least, although all four species had significantly lower reproductive success in small as compared to large forest fragments. Reproductive success of Rose-breasted Grosbeaks and Veery were similar in large forest fragments and continuous forest and were close to maintenance levels in small fragments, despite published reports of significant declines that might be attributable to poor productivity. Nest predation was the major cause of nest failure, with brood parasitism by Brown-headed Cowbirds (Molothrus ater) further reducing the number of host young fledged from Ovenbird and Red-eyed Vireo nests. Parasitism rates on the remaining species were very low and not analyzed statistically. Local forest cover had little effect on predation rate or rate of parasitism by Brown-headed Cowbirds. Although local forest cover had little effect on reproductive success within fragmented landscapes, nest success was significantly higher in continuous forest than in all fragments combined for both Ovenbird and Red-eyed Vireo. Our data show a poor agreement with Breeding Bird Survey population trend estimates for Ontario between 1986 and 1996. As only 1% of the forest fragments locally available are large enough to function as source habitats, we strongly advocate their preservation. We recommend preservation of forest tracts at least 500 ha in size, particularly woodlots with >90 ha in core area, to function as sources that will help guard against population declines on a local scale.

LANDSCAPE AND FRAGMENT SIZE EFFECTS ON REPRODUCTIVE SUCCESS OF FOREST-BREEDING BIRDS IN ONTARIO

To determine the minimum size requirements and influence of landscape context on reproductive success of forest-breeding songbirds, we monitored nesting success of five species of songbirds on 40 fragments (12–2350 ha in total woodlot size) and two continuous forest sites in south-central Ontario from 1994 through 1997. Woodlot size was the most important variable contributing to differences in reproductive success, with local forest cover (within a 10-km radius) having no significant additional effect on productivity for any species. For all species, except Red-eyed Vireo (Vireo olivaceous), adult female reproductive success was at or above replacement levels in large fragments (mean of 121 ha core area, 849 ha woodlot area) and continuous forest, and below replacement levels in small fragments (mean of 7.8 ha core area, 93 ha woodlot area). Red-eyed Vireo productivity was particularly low, with single-brooded females unable to maintain populations in any woodlots monitored, although populations were close to replacement levels in continuous forest. Only the very largest fragments acted as sources for Ovenbird (Seiurus aurocapillus) and Wood Thrush (Hylocichla mustelina) (i.e., >23 ha in core area, 225 ha in total woodlot area). Ovenbird and Wood Thrush were the most area-sensitive, and Rose-breasted Grosbeak (Pheucticus ludovicianus) and Veery (Catharus fuscens) the least, although all four species had significantly lower reproductive success in small as compared to large forest fragments. Reproductive success of Rose-breasted Grosbeaks and Veery were similar in large forest fragments and continuous forest and were close to maintenance levels in small fragments, despite published reports of significant declines that might be attributable to poor productivity. Nest predation was the major cause of nest failure, with brood parasitism by Brown-headed Cowbirds (Molothrus ater) further reducing the number of host young fledged from Ovenbird and Red-eyed Vireo nests. Parasitism rates on the remaining species were very low and not analyzed statistically. Local forest cover had little effect on predation rate or rate of parasitism by Brown-headed Cowbirds. Although local forest cover had little effect on reproductive success within fragmented landscapes, nest success was significantly higher in continuous forest than in all fragments combined for both Ovenbird and Red-eyed Vireo. Our data show a poor agreement with Breeding Bird Survey population trend estimates for Ontario between 1986 and 1996. As only 1% of the forest fragments locally available are large enough to function as source habitats, we strongly advocate their preservation. We recommend preservation of forest tracts at least 500 ha in size, particularly woodlots with >90 ha in core area, to function as sources that will help guard against population declines on a local scale.

LANDSCAPE-LEVEL CORRELATES OF SMALL-MAMMAL ASSEMBLAGES IN FOREST FRAGMENTS OF FARMLAND

We studied effects of forest fragmentation on 15 species of small mammals, including 6 species of forest-dwelling granivorous rodents, in the Indian Pine watershed of west-central Indiana. Presence‐absence and population densities of small mammals were assessed in spring in 35 woodlots of various sizes (0.1‐150 ha) and 2 continuous forest sites ( .1,000 ha) using live traps in 1992‐1996. Presence‐absence and population density were related to landscape attributes using logistic and multiple linear regression models, respectively. Species richness of forest-dwelling small mammals increased with area and was highest in continuous forest sites. Nested subsets of the full complement of species were found in smaller woodlots. White-footed mice (Peromyscus leucopus) were ubiquitous, and eastern chipmunks (Tamias striatus) were nearly ubiquitous across the landscape; densities of both species were related inversely to forested area. Fox squirrels ( Sciurus niger) were found at 84% of study sites, and they did not respond negatively to isolation of forest patches. Red squirrels (Tamiasciurus hudsonicus) were unevenly distributed across the landscape and were found most often in woodlots with large core areas and simple shapes, possibly indicating sensitivity to edge. Southern flying squirrels ( Glaucomys volans) and gray squirrels (S. carolinensis) were restricted to continuous forest sites and .4.6-ha woodlots adjacent to other wooded habitat. Species of small mammals differ appreciably in their sensitivities to agriculturally induced fragmentation of forests. Interspecific differences within this assemblage were not due solely, or even primarily, to body size. Rather, differential responses of species to fragmentation likely resulted from variation in habitat breadth and ability to move through an agricultural matrix.

Habitat use by Chiropotes satanas utahicki and syntopic platyrrhines in eastern Amazonia.

Primates were surveyed at two sites in the Xingu-Tocantins interfluvium, in eastern Amazonia, where at least eight platyrrhines are known to occur, including the endemic Chiropotes satanas utahicki, vulnerable to extinction. Only three other forms; Alouatta belzebul belzebul, Cebus apella apella, and Saguinus midas niger; were recorded at both sites. Data on habitat use (forest type and strata) were collected in standard line transect surveys and analyzed with relation to the availability of forest types, as well as between sites and species. The smallest- (S. midas) and largest-bodied (A. belzebul) species were relatively common at the continuous forest site, where they exhibited a significant preference for primary terra firme forest. At this site, Cebus demonstrated a significant preference for liana and flooded forest in contrast with primary or secondary terra firme forests. The medium-sized Cebus and Chiropotes were more common in the isolated forest fragment (where they were also observed together frequently), but no clear habitat preferences were found at this site for any species. A. belzebul occupied significantly higher forest strata than other species, which all used relatively similar levels. C.s. utahicki was active in much lower forest strata than other bearded sakis, whereas S. midas was observed at much higher levels than at other sites in eastern Amazonia. It remains unclear whether and to what extent observed patterns are determined by differences between taxa, populations, or ecosystems, but the data indicate that C.s. utahicki is relatively tolerant of habitat disturbance.

Forest fragmentation in central Amazonia and its effects on litter-dwelling ants

We assessed responses of ants nesting in twigs in the litter layer to habitat changes associated with forest fragmentation in central Amazonia. Ants were collected along transects located at nine distances (5, 20, 40, 60, 100, 200, 300, 400, 500 m) from the edges of two isolated 100-ha fragments and two continuous-forest sites. In total, 2880 m 2 of litter were examined for the presence of ant colonies. We detected a significant decrease in litter depth with increasing distance to forest edge, and an increase and then decrease in the average diameter-at-breast-height (DBH) of large trees (DBH⩾10 cm), and in tree basal area. Distance to forest edge significantly affected ant species composition and this effect was partly attributable to variation in litter depth. With the exception of one fragment, no significant changes in ant nest densities or species richness were found with increasing distance to forest edge. However, species richness of ants was greater in continuous forest than in both fragments. Furthermore, most species (65.8%) had greater nest densities in continuous forest. These results suggest that edge and isolation effects both play a role in structuring litter-dwelling ant communities in Amazonian forest remnants.

Shrinking forest shrinks skink: morphological change in response to rainforest fragmentation in the prickly forest skink ( Gnypetoscincus queenslandiae)

Large-scale fragmentation of rainforest occurred on the Atherton Tableland in the Australian Wet Tropics from 50 to 100 years ago, leaving numerous fragments of varying sizes. Eleven fragments (from <1 to 75 ha in area) and eight continuous-forest sites were studied to assess the effects of fragmentation on the morphology and demography of the rainforest-endemic prickly forest skink ( Gnypetoscincus queenslandiae). Skink abundance (number of individuals captured per hour of search effort) was significantly greater in continuous forest than in forest fragments. Moreover, both skink abundance and the availability of decaying logs, which provide key habitat for this species, increased with fragment area. Fragments contained a smaller proportion of adults, and individuals in fragments were smaller on average for all measured morphological features, than those in continuous forest. Thus, although prickly forest skinks appear to be maintaining populations in rainforest remnants, they are nonetheless being affected by fragmentation. These demographic and morphological changes may be caused by alterations in habitat and prey availability and/or by microclimatic changes associated with edge effects.

Effects of Plot Size and Habitat Characteristics on Breeding Success of Scarlet Tanagers

We studied the effects of forest patch size and habitat characteristics on breed- ing success of Scarlet Tanagers (Piranga olivacea) in western New York in 1995 and 1996. Twenty forest stands were grouped into four size classes: Group I ( 50 to 150 ha, n = 5), and Group IV (>1,000 ha, n = 2). Group I, II, and III sites were habitat patches located in fragmented landscapes, whereas Group IV sites were located in continuous forests. Although densities of male tanagers were similar in Group II, III, and IV sites, tanagers were absent from all forest patches smaller than 10 ha. Territory size did not differ among males in Group II, III, and IV sites. Pairing success exceeded 75% in all forest size classes in 1995 and 1996, and 100% of the observed males were paired in continuous forest sites. Pairing success differed significantly among forest size classes in 1995 and approached significance in 1996. Fledging success increased significantly with area and was highest (64%) in continuous forest sites. Stepwise multiple regression and principal components analysis indicated that male tanagers breeding in for- est patches with higher canopy cover and lower density of oaks had higher pairing success than males in patches with lower canopy cover and higher density of oaks, and that males breeding in larger forest patches with more surrounding forest cover had higher fledging success than males in small patches with less surrounding forest cover. Our results indicate that: (1) breeding density is not a good indicator of habitat quality for forest-interior Neo- tropical migrants, and (2) large tracts of continuous forest are important for maintaining populations of these species. Received 19 March 1997, accepted 5 May 1998.