This study investigates whether increases in elephant populations may influence the structure of African savannas, and consequently may affect other herbivores through changes in habitats. Two contrasting periods in terms of elephant population densities were compared in Hwange National Park, Zimbabwe: the early 1980s and the late 1990s. Elephant population density and other ungulate population densities were estimated for a c. 400-km2 area from road counts. Vegetation structure at the landscape scale was assessed using aerial photographs for the same area. All browsers and grazers declined between the early 1980s and the late 1990s, whereas elephants experienced a 16-fold increase. At the landscape scale, vegetation structure changed little with no evidence of an opening of the habitats. These results do not support any kind of medium-term facilitation between elephants and other herbivores. They rather suggest a negative effect of elephants on other herbivore species when elephants are present at high densities. This study rules out a scenario where the decrease of the different herbivore populations was caused by large changes in vegetation structure due to elephant activity.

We describe levels of parasitism of ectoparasitic bat flies (Hippoboscoidea: Streblidae) on male and female bats from an extensive Neotropical survey. The collection resulted from coordinated vertebrate-parasite surveys undertaken by the Smithsonian Venezuelan Project (SVP) from 1965–1968, which sexed 24 978 bats of 130 species. Streblid parasites were recovered from 6935 individuals of 87 bat species, but only 47 species were captured frequently enough (≥ 20 infested individuals) to permit reliable estimates of streblid parasitism on males and females. Well-sampled species included 39 phyllostomids, four mormoopids, two noctilionids, one natalid and one molossid. Prevalence of streblid parasitism (proportion of individuals infested) of male and female bats was generally not significantly different, and averaged 0.34 across infested species. In species-level analyses assessed against captures, significant sex differences in infestation levels were noted in six species; all had mean prevalence below 0.5 and females were parasitized disproportionately in each. Sex differences in total numbers of flies were noted in 21 species, and in 16 of these, females carried disproportionately heavy loads. Sex differences were also found for eight species of bat in the number of fly species infesting an individual; seven of eight showed heavier female parasitism. In analyses weighted by infestation levels, sex differences in total number of flies were found in only 12 species, with seven showing excessive parasitism of females, and no species showed sex differences in the number of fly species infesting them. These significant biases were not associated with sexual size dimorphism among the bat species. Generally higher levels of parasitism among female bats accords with theory, given their generally higher survivorship and enhanced probabilities of lateral and vertical transmission of host-specific parasites, but contrasts with patterns shown by many other parasitic arthropods. Future analyses should target social groupings of bats, not passively sampled foragers, to better address the mechanisms responsible for this pattern.

Natural forests on Java have been generally cleared and remnants are now confined to mountain areas Although legally protected, these forests are used by local people for products like firewood, timber and fodder Locally, such exploitation is carried out daily by hundreds of people, whose livelihood depends on forest products Using a landscape-guided approach, a 15,500 ha area of montane forest in East Java has been sampled and analysed at the sylvatic mosaic level Vegetation types, sociologic species groups and vegetation structure are described for landscape units The most accessible landscape units have the most degraded forest with basal area values < 15 m2 ha−1 and a canopy closure of < 20% The least accessible units have relatively undisturbed forest with basal area values of > 45 m2 ha−1 and a canopy closure of > 60% Human activities such as tree cutting have altered species composition and modified forest structure to such an extent that large stretches of forest have been replaced by woodland or shrubland Forest degradation is described as a human induced process in which basal area and canopy closure are ultimately reduced, in spite of regrowth Comparison with other forest areas on Java shows that degradation as a result of tree cutting is a widespread phenomenon

Forest edges bordering on pasture were created by cutting and burning the surrounding Amazonian lowland rain forest in the dry season (June) of 1990. Litterfall was measured for 3.5 y along transects 10, 50, 100 and 250-m into the forest from the forest edge. Litterfall along the 10-m transects increased by up to 2.5 times that on spatial controls (250-m transects) in the dry season in which the edge was created. In the second dry season after edge creation litterfall at 10-m was lower than on controls, after which it returned to control rates in the second wet season, 1.5 y after edge creation. Litterfall 50-m into the forest was less affected; there was a smaller rise in the dry season in which the edge was cut, and no significant effects after that. At 100-m there was no effect of edge creation on litterfall. Phosphorus concentrations in litterfall were elevated along 10-m transects, 10–12 wk after edge creation, possibly because of reduced retranslocation prior to abscission. The changes in litterfall described here are inextricably linked to the biomass collapse, which occurs near forest-fragment edges in the same area.

This study investigated the vertical stratification of Old World fruit bats (Chiroptera: Pteropodidae) in relation to habitat structure and wing morphology, in a lowland Malaysian rain forest. In total, 352 fruit bats of eight species were captured within the subcanopy of the structurally complex old-growth forest during 72 306 m2 mist net hours of sampling. Fruit bat species that were grouped in relation to capture height were also grouped in relation to wing morphology – with those species predicted to have more manoeuvrable flight (i.e. lower wing-loadings and lower aspect-ratios) captured in increasingly cluttered airspaces. Thus, small differences in wing morphology are likely to be ecologically significant to the vertical stratification of bats. Hence, habitat heterogeneity may be a key factor promoting fruit bat species diversity in old-growth palaeotropical forests.

We used tree-rings to reconstruct long-term patterns of suppression, release and growth among five sympatric canopy tree species representing the full range of shade tolerance in a seasonal tropical forest in western Thailand. We expected that the frequency and duration of suppression and release events would be positively correlated with shade tolerance. All five species showed evidence of major and moderate growth releases. As expected, Melia azederach, an extreme heliophile, had the fewest releases. However, among the other species the number of major releases was consistent across the range of shade tolerance. The most significant difference among the species was the number of moderate releases recorded. There was a general positive correlation between the number of moderate releases and shade tolerance; however, Chukrasia tabularis, a relatively shade-intolerant species, had an anomalously high number of moderate releases. The study species also showed considerable variation in canopy accession strategies. The least common canopy accession strategy was establishment in the understorey and growth into the canopy in the absence of any gaps. However, with the exception of Melia, all four study species had one or more individuals that successfully reached the canopy using each of the four canopy accession strategies. These results highlight the importance of periodic or episodic bouts of gap formation on canopy tree recruitment and the utility of tree-rings for reconstructing long-term growth patterns in tropical trees.

Seed dispersal, predation, desiccation and seedling survival are strong bottlenecks that impede tropical forest recovery on abandoned agricultural lands (reviewed in Holl 2002). The ability to resprout after disturbances overcomes these barriers, as those individuals bypass the seed stage and have more vigorous shoots than seedlings (Bond & Midgley 2001, Kammesheidt 1999, Kennard et al. 2002). Resprouting is a particularly important recovery mechanism in tropical dry forests (reviewed in Vieira & Scariot 2006). There are a number of reasons why resprouting may be more important in dry forests than in rain forests, including slower decay rates of trunk bases (Ewel 1980), or adaptation of plants to drought (Bond & Midgley 2001, Sampaio et al. 1993).

The fate of seeds of eight tree species was followed during 4 y of storage in the soil of an Afromontane forest at Gara Ades in the eastern highlands of Ethiopia. Seeds were enclosed in nylon mesh bags and buried at 5 cm soil depth. The bags were exhumed at intervals and the viability of the seeds was assessed by germination and cutting tests. Seeds of Bersama abyssinica and Ekebergia capensis germinated in the soil almost completely within a year after burial. The seeds of Juniperus procera, Olea europaea and Podocarpus falcatus also germinated to a substantial degree in the soil but with a distribution over several years, and some seeds of these species remained viable at the end of the 4-y period. Germination in the soil was very low in seeds of Acacia abyssinica and Croton macrostachyus throughout the whole burial period and the seeds kept their viability. In C. macrostachyus fresh seeds were highly dormant, but after 3 y or more in the soil they germinated readily in the laboratory suggesting an altered dormancy with time in the soil. Dormancy in seeds of A. abyssinica and Indigofera rothii was not altered throughout the study period as evidenced by marginal or no germination during incubation in the laboratory. The differential seed behaviour observed during storage in the soil can be an indicator of the regeneration strategy of the species studied. B. abyssinica, E. capensis, J. procera, O. europaea and P. falcatus form seedling banks on the forest floor and lack persistent soil seed reserves in contrast to A. abyssinica, C. macrostachyus and I. rothii which accumulate reserves of long-lived seeds in the soil. The generally high levels of dormancy and somewhat extended viability in the soil, even in several of the species producing seedlings in undisturbed forest, may have been selected for under a climate of seasonal drought and unreliable rainfall that characterizes this region.

Local species richness and between-site similarity in species composition of parasitoid wasps (Hymenoptera: Ichneumonidae; Pimplinae and Rhyssinae) were correlated with those of four plant groups (pteridophytes, Melastomataceae, Burseraceae and Arecaceae) in a western Amazonian lowland rain forest mosaic. The mosaic structure of the forest was related to variation in soils within the non-inundated terrain. Significant matrix correlation between patterns in parasitoid wasp species composition and plant species composition was found. Most of the overall correlation was due to idiobiont parasitoids of weakly concealed hosts, which attack host larvae and pupae in exposed situations, with two of the four ecologically defined parasitoid groups showing no correlation at all. A positive correlation between the number of plant species and the number of Pimplinae and Rhyssinae species at a site was found when the latter was corrected for collecting effort. Consequently, the degree of floristic difference between sites may be indicative of the difference in species composition of ichneumonids, and the species richness of plants may serve as a predictor of the species richness of parasitoid wasps. Although these results were obtained in a mosaic including structurally and floristically clearly different types of rain forest, the correlation coefficients were relatively low, and the present results lend only weak support to the idea of using plant distributions as indicators of animal distributions.

The spatial distribution of canopy gaps was analysed on three sites (total 71 ha) in the tropical moist forest of Tai National Park, Ivory Coast. Pattern analysis revealed a clustered distribution of gaps for two of the three sites. Catena dependent gap formation processes might explain local differences in the occurrence and distribution of gaps. Gap densities, sizes and percentage forest area in gap phase are higher on the upper and middle slope than on the crest or lower slope. As a consequence, regeneration of gap dependent tree species might be directed to the catena positions with the highest disturbance regime. The spatial distribution of gap dependent species can be clumped, not only due to the regeneration within gaps, but also due to the clustered nature of gap distribution on its own.

Some of the most spectacular examples of coevolution between flowers and their pollinators are reflected in their morphologies. Tongue length of insects and bill lengths of nectar-feeding birds are some of the most significant characters in pollination studies (Kearns & Inouye 1993). As Darwin noted as early as 1862, the evolution of deep floral tubes or spurs and long tongues of flower visitors can be explained by runaway coevolution (Nilsson 1988). However, such a process has only been shown to be likely in a few cases, e.g. between Malagasy orchids and hawkmoths (Nilsson et al. 1985). The pollinator of the Malagasy orchid Angraecum sesquipedale, with a 30 cm nectar spur, is a sphingid moth with a tongue of a similar size (Darwin 1862, Nilsson 1998). However, most studies of such extreme pollination specialization also report that the interaction is asymmetrical, i.e. the pollinators interact with a guild of plants, whereas the plant often depends on only a few pollinators (Johnson & Steiner 1995).

Thirty-six of the 70 species of bats known from Belize were recorded from the area around Lamanai, Orange Walk County: two in roosts and 34 in about 680 mist net hours that produced 560 captures. Day roosts used by 35 of the species were located using radio-tracking (Sturnira lilium, Platyrrhinus helleri, Centurio senex and Bauerus dubiaquercus) or general searching for roosts (Rhynchonycteris naso, Saccopteryx bilineata, Saccopteryx leptura, Dicli durus albus, Mimon bennettii, Micronycteris schmidtorum, Carollia brevicauda, Carollia perspicillata and Eptesicus furinalis). Data on the day roosts of 23 other species were determined from the literature. Most species reported from Lamanai (19) roosted in hollows, while others used foliage (6), tents (3), sheltered sites (2), crevices (2), open sites (1), and a few species used more than one type of day roost (hollows and crevices (1); hollows and foliage (1); hollows, foliage and tents (1)). The fauna consisted of 13 aerial foragers, 9 gleaners, 11 fruit/leaf eaters, one trawler, one flower-visitor and one blood-feeder. In day roost use and foraging behaviour, the Lamanai fauna did not differ significantly from that of Paracou, French Guiana, but both these locations differed from the bat fauna of Kruger National Park, South Africa, in foraging behaviour.

Brachystegia-Julbernardia (miombo) woodlands inside and outside the Sengwa Wildlife Research Area, Zimbabwe, show major differences in structure, biomass and species composition. The woodlands inside the research area have a lower stem area, tree density and biomass than those outside it. The shrubs inside the research area occur at a higher density than they do outside it, but the biomasses of shrubs in the two woodlands are similar. The woodlands inside the research area are affected by elephants and fire, whereas those outside it are affected mainly by fire. It is postulated that the policy of maintaining the elephant population at 250 in the research area is having a beneficial effect and, provided elephants and fire are kept in check, the woodlands inside the research area will eventually recover.

Among the hypotheses formulated to explain the origin of Amazonian biodiversity, two (the riverine-barrier and the river-refuge hypotheses) focus on the role that rivers play as biotic barriers promoting speciation. However, empirical results have both supported and refuted these hypotheses. This is likely due, at least in part, to river-specific hydrologic characteristics and the biology of the focal species. The rivers of the Guiana Shield represent a model system because they have had more stable courses over time than those of the western Amazon Basin, where most tests of riverine barrier effects have taken place. We tested whether life-history traits (body size, habitat and larval development), expected to be important in determining dispersal ability, of 28 frog species are associated with genetic structure and genetic distances of individuals sampled from both banks of the Oyapock River. Thirteen of these species displayed genetic structure consistent with the river acting as a barrier to dispersal. Surprisingly, body size was not correlated with trans-riverine population structure. However, leaf-litter dwellers and species lacking free-living tadpoles were found to exhibit higher river-associated structure than open habitat/arboreal species and those with exotrophic tadpoles. These results demonstrate that rivers play an important role in structuring the genetic diversity of many frog species though the permeability of such riverine barriers is highly dependent on species-specific traits.

An understanding of the abundance of organisms is central to understanding ecology, but many population density estimates are unrepresentative because they were obtained from study areas chosen for the high abundance of the target species. For example, from a pool of 1072 lizard density estimates that we compiled from the literature, we sampled 303 estimates and scored each for its assessment of the degree to which the study site was representative. Less than half (45%) indicated that the study area was chosen to be representative of the population or habitat. An additional 15% reported that individual plots or transects were chosen randomly, but this often indicated only that the sample points were located randomly within a study area chosen for its high abundance of the target species. The remainder of the studies either gave no information or specified that the study area was chosen because the focal species was locally abundant.

A recently developed technique in soil ecology is the use of the bait lamina method, for assessing the feeding activity of soil animals in situ. Here the bait lamina method (slightly adapted to tropical conditions) was used for the comparison of four closely situated sites (a primary rain forest, a secondary rain forest and two mixed-species tree plantations) in the Brazilian Amazon (project SHIFT ENV-52). The total feeding activity (portion of empty apertures in a set of laminae) and the vertical distribution of the feeding activity were evaluated for differences between the sites. The feeding activities in the primary and secondary forest were at the same level and significantly lower than in the two plantations. Feeding activity in subplots where surface litter has been experimentally removed was significantly lower than in the controls with the litter layer intact, at all four sites. The abundance of soil macrofauna or Enchytraeidae was not correlated with the results of the bait lamina tests, while the abundance of mesofauna (probably mainly Oribatida) was consistent with the pattern of feeding activity. The feeding activities as determined with bait laminae did not correlate with the decomposition activities determined with litterbags. Due to the easy applicability in combination with its feasibility for statistical evaluation and its consistent responses to site differences in experimental treatments we consider the bait lamina method to be a promising approach for the biological assessment of tropical soils.