Monsoon Rainforest Research Articles

Modelling strategies for the management of the critically endangered Carpentarian rock-rat (Zyzomys palatalis) of northern Australia

The Carpentarian rock-rat (Zyzomys palatalis) is a critically endangered endemic rodent known from only four sandstone gorges in the southeast Gulf of Carpentaria, Northern Territory, Australia. These gorges harbour thickets of monsoon rainforest and broadleaf woodland, surrounded by a Eucalypt savanna matrix. The long-term persistence of Z. palatalis is threatened by altered fire regimes, grazing by feral animals and stock, weed intrusion, and the stochastic hazards associated with small, fragmented populations. To assess the relative importance of these threats and develop practical management options, a population and habitat simulation model was developed, based on the best existing data. Population viability was predicted to be highly sensitive to the frequency of hot, late dry-season fires. Progressive habitat degradation (due predominantly to intense late dry-season fires) is likely to substantially reduce population size and lead to the probable extinction of the species within the next 100 years. The most effective management strategy to counteract this threat would be regular, controlled, fuel reduction burns in the vegetation around the gorge entrances during the early dry season. Establishing a new population (through translocation of captive-bred individuals) would not appreciably reduce extinction risk, but could provide valuable additional data on the impact of threats, if conducted as an adaptive management experiment.

A model for seasonality and distribution of leaf area index of forests and its application to China

Abstract. We have constructed a phenological model of leaf area index (LAI) of forests based on biological principles of leaf growth. Field data of maximum LAI from 794 plots with mature or nearly mature stand ages over China were used to parameterize and calibrate the model. New measurements of maximum LAI from 16 natural forest sites were used to validate the simulated maximum LAI. The predictions of seasonal LAI patterns were compared with seasonal changes derived from the 1‐km satellite AVHRR‐NDVI data for nine undisturbed forest sites in eastern China. Then, we used the model to map maximum LAI values for forests in China.Model results indicated that the PhenLAI model generally predicted maximum LAI well for most forest types, even when maximum LAI is > 6. This suggests an ecological approach to the saturation problem in satellite detection of high forest LAI where the relationship between NDVI and LAI reaches an asymptote near a projected LAI value of 5 or 6. Furthermore, the predictions of seasonal LAI patterns in timing and dynamics were generally consistent with the satellite NDVI changes, except for monsoon forest and rain forest in south China where satellite detection of seasonal variation in leaf area is hardly possible. Compared with average projected LAI measurements of global forests from 809 field plots in literature data, our maximum LAI values were close to the global literature data for most of Chinese forests, but the average area‐weighted maximum LAI for all forests of China (6.68 ± 3.85) was higher than the global mean LAI of the 809 field plots (5.55 ± 4.14). We believe that forest LAI in China is commonly > 6, especially in tropical rainforest, subtropical evergreen broad‐leaved forest, temperate mixed forest, and boreal/alpine spruce‐fir forest where satellite detection of high LAI is hardly possible.

Can stable carbon isotopes (δ13C) in soil carbon be used to describe the dynamics of Eucalyptus savanna–rainforest boundaries in the Australian monsoon tropics?

AbstractThe history of isolated patches of monsoon rainforest within large tracts of Eucalyptus savanna is poorly understood because of the scarcity of reliable palaeoecological records in the Australian monsoon tropics. Elsewhere in the world, the ratio of the stable isotopes 13C to 12C (δ13C) in soil organic matter has shed light on the dynamics of rainforest–savanna boundaries because tropical grasses with the C4 photosynthetic pathway have a distinct δ13C signature (–17 to –9‰) compared with that of woody plants with the C3 photosynthetic pathway (–32 to –22‰). In order to determine the magnitude of the variation in δ13C, unreplicated soil profiles were sampled beneath different vegetation types on three boundaries between Eucalyptus savanna and rainforest that were both growing on Tertiary age laterite parent material. Replicated (n = 3) soil profiles, which were also derived from Tertiary age laterite, were sampled from beneath: (i) dense stands of African grasses within a frequently burnt Eucalyptus savanna; and within the same long unburnt Eucalyptus savanna, (ii) patches of African and natives grasses and (iii) clumps of Acacia trees. The strongly negative δ13C values of soil organic matter derived from the frequently burnt and long unburnt grassy understoreys in the Eucalyptus savannas showed that a considerable amount of the soil carbon was derived from C3 (woody) species despite the presence of a ground layer dominated by C4 grasses. However, a feature of these data was the considerable variability among the three ‘replicate’ profiles. The surface soil samples from beneath three clumps of Acacia trees in the unburnt Eucalyptus savanna had much less variable δ13C values and were similar to two of the three monsoon rainforests sampled. The pattern of δ13C values from unreplicated soil profiles from different vegetation types across three rainforest boundaries was also very variable and not always obviously related the known disturbance history of the extant vegetation. Given the considerable variability within and between vegetation types with contrasting disturbance histories, it is concluded that the use of carbon stable isotopes to advance understanding of the dynamics of rainforest and Eucalyptus savanna boundaries will require further development, such as determination of the 14C age and δ13C values of different soil carbon fractions.

A model for seasonality and distribution of leaf area index of forests and its application to China

We have constructed a phenological model of leaf area index (LAI) of forests based on biological principles of leaf growth. Field data of maximum LAI from 794 plots with mature or nearly mature stand ages over China were used to parameterize and calibrate the model. New measurements of maximum LAI from 16 natural forest sites were used to vali- date the simulated maximum LAI. The predictions of seasonal LAI patterns were compared with seasonal changes derived from the 1-km satellite AVHRR-NDVI data for nine undis- turbed forest sites in eastern China. Then, we used the model to map maximum LAI values for forests in China. Model results indicated that the PhenLAI model generally predicted maximum LAI well for most forest types, even when maximum LAI is > 6. This suggests an ecological approach to the saturation problem in satellite detection of high forest LAI where the relationship between NDVI and LAI reaches an asymptote near a projected LAI value of 5 or 6. Furthermore, the predictions of seasonal LAI patterns in timing and dynam- ics were generally consistent with the satellite NDVI changes, except for monsoon forest and rain forest in south China where satellite detection of seasonal variation in leaf area is hardly possible. Compared with average projected LAI measure- ments of global forests from 809 field plots in literature data, our maximum LAI values were close to the global literature data for most of Chinese forests, but the average area-weighted maximum LAI for all forests of China (6.68 ± 3.85) was higher than the global mean LAI of the 809 field plots (5.55 ± 4.14). We believe that forest LAI in China is commonly > 6, espe- cially in tropical rainforest, subtropical evergreen broad-leaved forest, temperate mixed forest, and boreal/alpine spruce-fir forest where satellite detection of high LAI is hardly possible.

Spatial pattern of dominant tree species of the secondary monsoon rain forest in Lianjiang, Guangdong Province

Based on the summed dominance ratios of species in sample plots, the first three dominant species (Litchi chinensis, Elaeocarpus sylvestris andCanarium album) of the secondary monsoon rain forest of Mt. Royal Shoe in Lianjiang City, western Guangdong, were chosen for analyzing their spatial distribution pattern with the analysis methods such as frequency models of Poisson Distribution, Two Negative Items Distribution, Neyman Distribution, aggregate indexes, Taylor exponential equation and Iwao's equation modeling. The results showed that these three species distributed in the congregate spatial pattern.Litchi chinensis andElaeocarpus sylvestris had the characteristic of basic congregate population and attractive characteristic between their plants. The patterns forCanarium album may change and become more evenly distributed with the increase of density. The overall species spatial pattern also depended on the conservation of the secondary monsoon rain forest besides it was affected by the species reproduction characteristics and its growing environment. The congregate spatial patterns of three dominant species showed that it is important to conserve forest urgent conservation of the forest.

Seasonal habitat use by flying‐foxes, Pteropus alecto and P. scapulatus (Megachiroptera), in monsoonal Australia

AbstractDistributions of flying‐fox (Pteropus alecto and P. scapulatus) were examined in relation to use of habitat in the essentially natural landscape of northern Australia. There were differences between the species in terms of the vegetation used for roosting and foraging, which were related to the reproductive cycle and seasonal variation in temperature, rainfall and the availability of preferred foods. Important habitats of P. alecto varied seasonally and included floodplain, mangrove, monsoon rainforest, Melaleuca open‐forest, and Eucalyptus miniata/E. tetrodonta open‐forest and woodland. The minimum scale at which conservation of P. alecto should be attempted is in the order of 5000 km2, based on seasonal patterns of habitat use. The size of this area will make conservation via traditional reserves difficult and conservation of important habitats outside reserves will be needed. Habitats protected for the benefit of P. alecto will also benefit P. scapulatus but because P. scapulatus is more mobile, displays greater yearly variation in distribution and is less well understood than P. alecto, appropriate conservation actions are less certain.

Conservation and genetics in the fragmented monsoon rainforest in the Northern Territory, Australia: a case study of three frugivore-dispersed species

Monsoon rainforest in the Northern Territory exists as an archipelago of tiny patches thought to be previously more continuous but having had retracted to a fragmented state by the Pleistocene. However, there is evidence of some more recent rainforest expansion, and many species have seeds that are dipsersed by small frugivores (birds and bats). Carpentaria acuminata and Ptychosperma bleeseri are endemic to the Northern Territory and Syzygium nervosum is endemic to the Northern Territory within its Australian distribution. Carpentaria acuminata and S. nervosum are common through moist rainforest, but P. bleeseri is rare, found in eight locations near Darwin. All have frugivore-dispersed seeds. The three species were surveyed across their geographic range to compare their population genetics. The results from C. acuminata and S. nervosum were analysed to investigate the distribution of allelic diversity, the effect of patch size and isolation on genetic diversity, and the effectiveness of seed dispersal by vagile frugivores. Genetic diversity (expected heterozygosity) was correlated with mean annual rainfall in both S. nervosum and C. acuminata. Genetic diversity (alleles per locus, percentage of polymorphic loci, expected heterozygosity) was also negatively correlated with population isolation but not patch size in C. acuminata. There was significant heterogeneity and low gene flow among C. acuminata populations (FST 0.379, Nm = 0.39) but not among S. nervosum populations, which had substantial evidence of gene flow (FST 0.118, Nm = 1.67). The distribution of rare alleles in C. acuminata was consistent with the theory of Pleistocene retraction to refugia. Ptychosperma bleeseri was genetically uniform across all sites. Only four individuals differed in their genotypes. The results suggest that the present populations originated from a single founder population rather than having independently contracted from a more extensive distribution. All three species had evidence of some more recent population expansion.

Very large area requirements for frugivorous birds in monsoon rainforests of the Northern Territory, Australia

Frugivorous birds in the Northern Territory, Australia move among patches of monsoonal rainforest and require many patches to maintain their populations. Using bird survey data from 149 rainforest patches and GIS coverage of all of the ca. 15,000 patches in the Northern Territory, we analyzed the relationship between the occurrence of six frugivorous bird species, patch area, and various measures of patch isolation. Rose-crowned fruit-doves ( Ptilinopus regina) and pied imperial-pigeons ( Ducula bicolor) exhibited strong relationships with a particular isolation measure: the percentage of rainforest within 50 km of the study patch. This result was not a statistical artifact of correlations between the measure and floristic or geographical variation between patches. Yellow orioles ( Oriolus flavicintus) and common koels ( Eudynamys scolopacea) were also influenced by this isolation measure, but various other factors (e.g. patch size and proximity to coastal areas) were more significant. Figbirds ( Sphecotheres viridis) and great bowerbirds ( Chlamydera nuchalis) were not particularly affected by patch isolation, and are probably not strongly dependent on rainforest. These relationships can be used to predict the consequence of destroying patches: for example, if half of the rainforest patches in the Darwin region were destroyed, our models predict that 80% of the pigeons and 90% of the doves would disappear. A threshold value was apparent for the pigeon, dove, and koel, suggesting that a critical degree of connectedness between patches over a very large area (32 km 2 of rainforest in a 7900 km 2 circle) is necessary to maintain these species.

Abandoned Orange-footed Scrubfowl (Megapodius reinwardt) nests and coastal rainforest boundary dynamics during the late Holocene in monsoonal Australia

Abstract The late Holocene history of monsoon rainforest retreat in northern Australia was explored by radiocarbon dating abandoned Orange-footed Scrubfowl ( Megapodius reinwardt ) nests. Scrubfowl can only build ‘nests’ (large heaps of soil and leaf litter) in coastal rainforests in northern Australia. A detailed study from a site with a known history of rainforest contraction demonstrated that it is impossible to accurately determine the date of rainforest boundary retreat by radiocarbon dating abandoned Scrubfowl nests. Nonetheless radiocarbon dates from abandoned nests at three sites on the coast of the Northern Territory suggested that the rainforests contracted sometime within the late Holocene. It is argued that the cause of such localised rainforest contraction is the combined effect of tropical cyclone damage and subsequent severe fires occurring in the storm debris, and not regional climatic change, a cause shown to be inconsistent with existing ecological and geomorphological data. The observed maintenance of rainforest boundaries under a regime of Aboriginal burning and the geographically patchy occurrence of abandoned Scrubfowl nests does not support the idea that Aborigines have been the primary cause of widespread rainforest retreat in the late Holocene.

The patterns of genetic diversity in Carpentaria acuminata (Arecaceae), and rainforest history in northern Australia

Carpentaria acuminata occurs in monsoon rainforest and is endemic to the Northern Territory, Australia. The genetic diversity of C. acuminata populations was surveyed across the geographical range of the species using isozyme analysis. Genetic diversity within C. acuminata populations (HE = 0.143) was typical of rainforest species and woody angiosperms generally. Genetic diversity was not correlated with rainforest patch size. However, there was significant heterogeneity among populations (FST = 0.379), with infrequent effective gene flow among populations (Nm = 0.39). Genetic diversity was negatively correlated with increasing distance between neighbouring C. acuminata populations, but geographical distance was not a good predictor of genetic similarity. C. acuminata is a favoured food of mobile frugivores such as Torres Strait pigeons and flying foxes. The decreased diversity with decreasing density of populations indicated that seed dispersal by frugivores has been important for the maintenance of diversity in this species. Populations known to have originated on relatively young, Holocene landforms were not necessarily genetically depauperate. Gene flow by pollen is apparently limited because C. acuminata populations are significantly inbred regardless of genetic diversity (F = 0.641). The distribution and diversity of rare alleles, i.e. those occurring in few populations, is consistent with the theory of rainforest contraction during the Pleistocene.

The genetics of Ptychosperma bleeseri, a rare palm from the Northern Territory, Australia

Ptychosperma bleeseri is known from only eight small monsoon rainforest patches located in the vicinity of Darwin, Australia. One population consists of only one adult plant and only one patch contains a population of more than 50 adult plants. A survey of the genetic diversity within the species was undertaken using isozyme analysis. Almost no genetic variation was recorded. Of the 223 wild collected samples tested, only four individuals varied at a single locus out of the 10 loci analyzed. The lack of diversity among populations suggests that the existing populations are derived from a common, genetically depauperate, source which has acted as a founder population. P. bleeseri was compared with the other Ptychosperma species which occur naturally in Australia ( P. macarthurii and P. elegans). Each of these species was found to differ consistently at several enzyme loci thus indicating that P. bleeseri does not simply represent an outlying population of either of these two species. Cultivated specimens of P. bleeseri were shown to have hybridized with other cultivated Ptychosperma species when grown together. This presents a problem for ex-situ conservation. Since palms are particularly favoured by Darwin residents in their gardens and public places, the location of urban development in close proximity to wild populations of P. bleeseri has the potential to introduce other Ptychosperma species into the wild populations. This could have serious effects on the genetic integrity of the species.

Alien plant invasions on the Tiwi Islands. Extent, implications and priorities for control

The data from five vegetation surveys on the Tiwi Islands in north Australia were compiled to examine the habitat preference of naturalised alien plant species. Eucalyptus forest, Callitris forest, Melaleuca woodland, Acacia scrubland and monsoon rainforest had very few alien plant species. Of the ten alien plant species that have become naturalised in native habitat, most were restricted to semideciduous Eucalyptus woodland and especially grassland and open scrubland on clay-rich substrate. All of these habitats have relatively fertile soils and it is argued that this greatly facilitates the spread of exotics, which may also be exacerbated by the activities, of Asian water buffalo, Bubalus bubalis. The herb Hyptis suaveolens is the most abundant alien plant species. Despite few exotics in native vegetation, 95 alien species have colonised disturbed habitat around settlements, with species richness related to the age of the settlement. The relatively low numbers of naturalised exotics enhances the natural value of the islands and the sea barrier provides valuable opportunity for quarantine. However, some alien species such as Pennisetum polystachion with small populations on the islands have elsewhere demonstrated ability to degrade large areas of their habitat and should be urgently eradicated.

Breeding biology, demography and success of the rufous-banded honeyeater, Conopophila albogularis, in Darwin, a monsoonal tropical city

The rufous-banded honeyeater, Conopophila albogularis, is probably the commonest small bird species in the suburbs of Darwin, Northern Territory. Nearly twenty pairs of this species were colour-banded on the Casuarina campus of the Northern Territory University, where they occupied territories of 0.15–0.47 ha all year-round, indicating a density of up to 12 birds ha-1. Six out of 48 birds survived 5 or more years, one individual being 9 years old at the time of writing. Males were larger in the four morphological dimensions measured. Breeding behaviour was recorded in every month of the year, but was concentrated in the late dry and wet seasons (September–March), commencing about two months before the rains. Over half of 274 nests were built in black wattles, Acacia auriculiformis, a common pioneer species both within urban Darwin and monsoon rainforest ecotones of the Northern Territory. Contrary to the literature, both sexes participated in building the nest, and the incubation and nestling periods each lasted 14 days. The clutch size was usually two (78%), and mean clutch size for 85 nests was 2.1 eggs. Nest success was about 70%, hatching success of eggs was 74% and fledging rate of nestlings 87%. Four broods per season were common, and two pairs successfully raised five broods in one season. Average annual pair productivity was 5.8 fledglings (possibly the highest yet recorded for an Australian passerine species), one pair raising a remarkable 32 young over five seasons (6.4 fledglings per season). The exceptionally high nest success and productivity were probably mainly due to the scarcity of predators, and the long breeding season, respectively, the latter being facilitated by artificial watering of gardens and lawns during the dry season. The colonising success of this species in Darwin is attributed to Darwin’s coastal location and the close proximity of favoured natural habitats, as well as the generalised diet of the species and its predilection for the abundant black wattle.

The role of seedling age and size in the recovery ofAllosyncarpia ternatafollowing fire

AbstractThis paper examines the effects of seedling size and age on fire tolerance ofAllosyncarpia ternata(Myrtaceae), a dominant tree in patches of monsoon rainforest of the wet‐dry tropics in the Northern Territory, Australia. We address the following questions: how large does a seedling have to be to tolerate fire; how old does it have to be to reach this fire‐tolerant size; and how can land‐management authorities best manage fire regimes to maintainAllosyncarpiaforest? In a field experiment, shadehouse‐grown seedlings aged from 8 months to 5 years were subjected to low‐ and high‐intensity fires in September 1994. Among 5‐year‐old seedlings, mortality was independent of fire intensity. However, mortality of young (8‐month‐old) seedlings was significantly higher in the high‐intensity fire. Three‐year‐old seedlings behaved in an intermediate manner; their survivorship and growth were marginally favoured by low‐intensity fire, rather than high‐intensity fire or no fire at all, and were dependent on pre‐treatment seedling height. Thus, the critical age that distinguishes fire‐tolerant from fire‐sensitive seedlings is somewhat more than 3 years for relatively short seedlings and somewhat less than 3 years for taller seedlings. In August 1993, a wildfire penetrated several hundred metres intoAllosyncarpiaforest growing on a steep, rocky escarpment, where it caused severe damage toA. ternataseedlings. More than three‐quarters of the ≥ 3.5‐year‐old seedlings (including some that had suffered the total loss of above‐ground parts) recovered during the following wet season and showed higher growth rates than their unburned neighbours. New growth was also promoted in those tall seedlings and saplings that had sustained only partial leaf scorch. In contrast, all 18‐month‐old seedlings were killed by the fire. Measurements of leaf‐scorch height in burnedAllosyncarpiaforest on the escarpment indicated a general uphill decrease in fire intensity, matching trends in increasing site rockiness and decreasing fuel density. An important implication for land management is that a fire‐free interval of at least 3 years following a seed‐fall event is required for a new generation ofA. ternatagerminants to progress into the cohort of established seedlings.

Regional Biodiversity Planning and Lemur Conservation with GIS in Western Madagascar

Primary forest cover in western Madagascar declined from 12.5% in 1950 to 2.8% in 1990. Approximately 23% of remaining forest is located within reserves, but this is no guarantee of protection. Forest cover in the Andranomena Reserve in western Madagascar has been reduced by 44% since 1950. The identification of priority areas for establishment of development projects has been constrained by lack of information on the distribution, abundance, and habitat requirements of threatened species and the size, condition, and threats to survival of forest remnants. We evaluate a rapid fauna survey and habitat modeling procedure specifically designed to generate information for reserve selection and design using Geographical Information Systems (GIS). The study was conducted in the largest remnant (94,000 ha) of primary monsoon rainforest in western Madagascar, using lemurs as an indicator group. Lemur abundance and microhabitat variables were measured at 64 stratified survey sites dispersed throughout the region. These records were incorporated into a GIS with other environmental and landuse data derived from maps and satellite imagery. Statistical procedures were applied to analyze species habitats, predict and map species distributions, and estimate species population sizes throughout the study area. Patterns of slash and burn agriculture and anthropogenic disturbance were also analyzed to identify areas at low risk of cultural disturbance. A map of disturbance risk was overlaid with a map of relative lemur biodiversity to identify conservation priority areas with high biodiversity and low risk of short term disturbance. Lemur diversity was most strongly influenced by habitat clearing and human disturbance (harvesting, stock grazing, and hunting) within 8 km of villages. Lemur diversity was highest in a region of higher elevation and rainfall, distant from villages and roads. The only existing reserve in the study area was located in a region of high disturbance risk. Almost half the reserve has been degraded during the past 45 years. There is considerable scope for re‐allocation of land use within the study area to include a core protected area, a buffer zone with controlled hunting and timber harvesting, and an adjacent ecotourism facility. We found that information generated by a stratified biodiversity and landuse survey at a sampling intensity of less than 0.1% was sufficient to provide an objective foundation for regional biodiversity planning.

Contribution of China to the global carbon cycle since the last glacial maximum

A better understanding of the long-term global carbon cycle requires improved estimates of the changes in terrestrial carbon storage (vegetation and soil) during the last glacial-interglacial transition. A set of reconstructions of palaeovegetation and palaeoclimate in China for the last glacial maximum (LGM) and the mid-Holocene (MH) allows us to use the Osnabrück biosphere model (OBM), which needs as input only 3 climatic parameters that are easily derivable from palaeodata, to reconstruct the past terrestrial carbon storage since the LGM. The change from the conditions of the LGM (colder and drier than present) to the MH (warmer and wetter than present) resulted in a gain of 116 Pg of terrestrial carbon in China mainly due to the build-up of temperate forest and tropical monsoon rain forest, and to the effects of changes in climate and CO2 levels. However, a loss of 26 Pg of terrestrial carbon (which does not include anthropogenic disturbances) occurred in China between the MH and the present due to shifts in the area covered by the main vegetation types. Results also show that glacial-interglacial changes in climate and vegetation distribution, both associated with variations in the Asian monsoon system, significantly affected terrestrial carbon storage in China which strongly contributed to the global carbon cycle.

Regeneration of monsoon rain forest in northern Australia: the sapling bank.

. As part of a wider study examining regeneration pathways in monsoon rain forest vegetation in northern Australia, a one‐off, dry season census of saplings was undertaken along transects sampled at each of 33 relatively undisturbed sites broadly representative of the range of regional monsoon rain forest vegetation. Four floristic quadrat groups were derived through TWINSPAN classification. Subsequent analyses involved: (1) comparison of mean dry season stockings of juveniles occurring in different rain forest types, and their structural and environmental correlates; (2) comparison of the contributions of different life forms, and the influence of clonal reproduction in the sapling regeneration banks of different forest types; and (3) exploration of relationships between the distributions of saplings of common tree species with respect to seed bank, floristic, structural, and environmental variables. While data presented here require cautious interpretation given that processes of seedling/sapling recruitment and mortality are highly dynamic, sapling banks were found to be most dense on coarse‐textured, moist soils, and least dense on coarse‐textured, seasonally dry soils. Canopy cover and fire impact were shown to be highly influential on sapling distribution, especially for saplings of tree species and those growing on seasonally dry sites. Sapling densities were little influenced by proximity to rain forest margins, except for shrubs. The potential for clonal reproduction was significantly greater on dry sites, especially for trees. The majority of saplings sampled were derived from relatively few common, non‐clonal, canopy tree species. Sapling distributions of 20 out of 23 common tree species were clumped in the vicinity of conspecific adults; for most species the strength of this relationship was greater than that for any other variable. These data support observations in the literature concerning the distribution of sapling banks in moist and dry tropical forests.

Convergence in community structure and dietary adaptation in Australian possums and gliders and Malagasy lemurs

Abstract The possums and gliders of Australia and the lemurs of Madagascar are ideally suited for investigation of radiation and convergence in arboreal mammal communities because of their long history of independent radiation in similar biophysical environments. Possum and glider communities at 22 sites in Australia and lemur communities at 18 sites in Madagascar were compared to evaluate patterns of community and species convergence in two shared habitats (evergreen and deciduous monsoon rainforests) and three derived habitats (Australian eucalypt forests and heaths, and Malagasy thorn scrubs). Arboreal mammals were classified into five dietary guilds for comparison of community structure and species adaptation. There is little evidence of convergence in trophic structure at the community level, but strong evidence of convergence in dietary adaptation at the species and niche level. Malagasy rainforest communities were characterized by a higher proportion of frugivore‐folivores, and Australian rainforest and moist eucalypt communities by a higher proportion of folivores and folivore‐frugivores. Possum and glider communities in temperate eucalypt forests and heaths supported a higher proportion of nectarivores and exudivores. Overall species richness was significantly higher in Madagascar. Four hypotheses are erected to account for these differences. The relative scarcity of frugivores in Australian evergreen rainforests is accounted for by the predominance of bird‐dispersed fruits, and fruiting phenologies unsuitable for mammal exploitation and dispersal. Higher folivore diversity in Australian evergreen rainforests is associated with a more pronounced seasonal shortage of edible fruits and new leaf, which has increased dependence on mature leaf. Folivore species richness is greatest in Australian moist eucalypt forests where structural complexity, sustained by serai responses to wildfire, permits a high level of vertical segregation. Increased nectarivory and exudivory in Australian temperate eucalypt forests and heaths is associated with Mediterranean winter rainfall regimes, which permit year round exudate production, and not with the absence of nectar feeding bats as previously supposed. A lower overall species richness in Australian rainforests is attributed to a longer history of contraction and fragmentation by anthropogenic fires (monsoon rainforests) and Pleistocene climatic change (tropical evergreen rainforests). A high degree of convergence is apparent between genera occupying folivore and wood gouging niches, in terms of gastrointestinal morphology (e.g. Pseudocheirus‐Lepilemur) and dental morphology (e.g. Dactylopsila‐Daubentonia). Divergences are most apparent in adaptations associated with frugivory, including larger body size, diurnality, and bipedal suspension and leaping modes of locomotion (in Eulemur spp.). This study highlights the importance of founding effects, competition from other vertebrate taxa, coevolution between animals and their food plants, and differences in biophysical environments, as determinants of mammalian radiation and convergence.

A comparison of ground-dwelling small mammal communities in primary and secondary tropical rainforests in China

ABSTRACTTrapping was conducted in primary and secondary forests to investigate the effects of past logging on small mammal communities. Thirteen I-ha sites were snap-trapped for three consecutive days in tropical monsoon rainforest in Xishuangbanna Nature Reserve, southern China. A total of 3900 trap-days yielded 198 captures, of which 118 were in the primary forest and 80 in the secondary forest. A total of 12 ground-dwelling species (eight Rodentia, Muridae; three Insectivora, Soricidae; one Insectivora, Erinaceidae) were caught, five (all Rodentia, Muridae) occurred in the primary forest and 11 in the secondary forest. The most abundant species was Niviventer confucianus, accounting for 62% and 39% of the total captures in the primary and the secondary forests, respectively. In the secondary forest the mean abundances of the two dominant species, N. confucianus and Maxomys surifer, were reduced, and a non-commensal form of Rattus rattus disappeared, but N. fulvescens showed no change while Mus pahari increased. The seven species which occurred only in secondary forest had very low abundance. Overall, the multivariate Mantel test showed the two communities to be significantly different. After logging, species diversity increased, and the two dominant species had increased body mass and their mean abundance decreased.

Distribution, habitat and satus of the Carpentarian rock-rat, Zyzomys palatalis

A survey was conducted in the Northern Temtory and Queensland to determine the distribution, status and habitat use of the recently described Carpentarian rock-rat, Zyzomys palatalis. The species was previously known from two localities in the Gulf of Carpentaria, and this survey added only one new population to its known range. The species is restricted to monsoon rainforest on scree slopes. In the Gulf region this habitat is very limited and highly fragmented. The only areas where suitable habitat exists are in relatively inaccessible gorges where permanent springs provide enough moisture to maintain the monsoon rainforests. These patches of relict vegetation may be threatened by feral animals and changing fire regimes. Because of the low population size of Z. palatalis and the small size of suitable habitat available, the species is considered to be critically endangered.