Conservation of Heathland Ground Beetles (Coleoptera,Carabidae): The Value of Lowland Coniferous Plantations

The value of conifer-dominated plantation forestry for heathland carabids is examined, in Breckland, eastern England. Historically dominated by heathland habitats supporting scarce species of restricted distribution within the UK, approximately 47% of Breckland was afforested in the early 20th century. The carabid fauna of this forest, previously little known, was sampled by intensive pitfall trapping, and results compared to samples from heathland and arable. The results affirm that Breckland supports many range-restricted carabid beetles, including scarce species restricted to lowland heathland or sandy habitats. Approximately half of the heathland-associated species, and also half of the Nationally Scarce carabid species recorded in Breckland, were found within the forest landscape. Carabid species composition differed between closed-canopy forest, heathland and arable, while open forest habitats (clear-felled stands, young restocked stands, and track margins) contained a diverse fauna with elements from all habitats. The proportion of heathland-associated species was significantly greater in open forest habitats than in closed-canopy forest and was similar to that of heathland assemblages. Wing morphology did not differ between heathland species solely recorded from heathland and those also recorded in the forest. Nationally Scarce species from the forest had a wider UK distribution than those recorded only from heathland. Our study shows that, within conifer forest planted in lowland heathland areas, open habitats provided by trackway networks and clear-felling management can have significant value for the conservation of open-ground carabids, including species of conservation importance.     

Zygodon forsteri(Brld.) Mitt. in Buckinghamshire

Abstract

Data from hemispherical (‘fisheye’) photographs are presented from twenty bryophyte sites, including open grassland and rock outcrops of various slopes and aspects, habitats of mountain species on shaded cliffs and in corries, and wooded habitats of lowland atlantic species. These demonstrate effects of slope and aspect on both instantaneous and total energy income in direct sunshine, especially marked during the winter half of the year. Corrie habitats of northern atlantic species such as Herberta adunca and Mastigophora woodsii were exposed to direct sun only for a limited period in summer, with peak direct irradiance in June (about half that for an unobstructed level site in the examples considered). Wooded habitats of lowland atlantic species received their peak irradiance (both direct and diffuse) immediately before leaf expansion in spring; direct irradiance at this period relative to an unobstructed level site ranged from 0.8 (instantaneous)/0.35 (daily total) (Plagiochila spinulosa, Scapania gracilis: Devon) down to negligible values (Drepanolejeunea hamatifolia, Aphanolejeunea microscopica; ravine sites in N. Wales and Yorkshire).     

Evolution along the Great Rift Valley: phenotypic and genetic differentiation of East African white‐eyes (Aves,Zosteropidae)

The moist and cool cloud forests of East Africa represent a network of isolated habitats that are separated by dry and warm lowland savannah, offering an opportunity to investigate how strikingly different selective regimes affect species diversification. Here, we used the passerine genus Zosterops (white‐eyes) from this region as our model system. Species of the genus occur in contrasting distribution settings, with geographical mountain isolation driving diversification, and savannah interconnectivity preventing differentiation. We analyze (1) patterns of phenotypic and genetic differentiation in high‐ and lowland species (different distribution settings), (2) investigate the potential effects of natural selection and temporal and spatial isolation (evolutionary drivers), and (3) critically review the taxonomy of this species complex. We found strong phenotypic and genetic differentiation among and within the three focal species, both in the highland species complex and in the lowland taxa. Altitude was a stronger predictor of phenotypic patterns than the current taxonomic classification. We found longitudinal and latitudinal phenotypic gradients for all three species. Furthermore, wing length and body weight were significantly correlated with altitude and habitat type in the highland species Z. poliogaster. Genetic and phenotypic divergence showed contrasting inter‐ and intraspecific structures. We suggest that the evolution of phenotypic characters is mainly driven by natural selection due to differences in the two macro‐habitats, cloud forest and savannah. In contrast, patterns of neutral genetic variation appear to be rather driven by geographical isolation of the respective mountain massifs. Populations of the Z. poliogaster complex, as well as Z. senegalensis and Z. abyssinicus, are not monophyletic based on microsatellite data and have higher levels of intraspecific differentiation compared to the currently accepted species.     

Five new species of Zingiber (Zingiberaceae) from Borneo

Recent collections from Sabah and Sarawak have revealed a wide species diversity among the Bornean members of the genus Zingiber. Four previously unknown and one incompletely known species of the genus are presented. Z. vinosum is a lowland species characterized by its burgundy colour of the lower surface of the leaves. Z. lambii was collected in primary lowland forest on ultramafic soil while Z. argenteum was collected in primary lowland forest on limestone. Both are characterized by their silvery leaves and orange inflorescences. Z. pendulum, found in primary mountain forest, has pending leafy stems and a deeply split labellum with very well developed side-lobes. Z. latifolium, found in secondary mountain forest, is vegetatively distinct by its broad leaves with a densely pubescent margin Z. vinosum, Z. Iambii, Z. pendulum and Z. latifolium are from Sabah and Z. argenteum is from Sarawak.     

Day-journey length and daily diet of solitary male gorillas in lowland and highland habitats

We compared day-journey length and daily diets of solitary male gorillas in lowland versus highland habitats. Solitary males in tropical forests of Zaire tend to travel longer distances, to visit more types of vegetation, and to consume more kinds of food than a solitary male mountain gorilla in the Virunga Volcanoes did. The number of feeding sites per day is larger and the mean distance between feeding sites is far longer for the former than the latter. These observations may reflect differences in food breadth and availability between highland and lowland habitats. The herbaceous plants that are eaten by mountain gorillas are densely and evenly distributed in the higher montane forest of the Virungas, where gorillas need not cover long distances to search for food. In contrast, herbaceous plants are scarce in primary and ancient secondary forests of lowland habitats, where gorillas travel long distances and eat various fruits and insects. The patchy and unpredictable distribution of foods may extend the distances over which gorillas search for food in the lowland habitat. However, solitary males showed a prominent reduction in day-journey length and changed their choices of food during the nonfruiting season (the long rainy and dry seasons) in the lowland habitats. This strategy may have developed during the Pleistocene and may have enabled them to enlarge their ranges to the higher montane forests, where fruits are sparse throughout the year.     

The role of landscape configuration in plant composition of floodplain forests across different physiographic areas

Questions: What is the relative importance of landscape variables compared to habitat quality variables in determining species composition in floodplain forests across different physiographic areas? How do species composition and species traits relate to effects of particular landscape variables? Do lowland and mountain areas differ in effects of landscape variables on species composition? Location: Southern Czech Republic. Methods: A total of 240 vegetation relevés of floodplain forests with measured site conditions were recorded across six physiographic areas. I tested how physiographic area, habitat quality variables and landscape variables such as current land‐cover categories, forest continuity, forest size and urbanization influenced plant species composition. I also compared how mountain and lowland areas differ in terms of the relative importance of these variables. To determine how landscape configuration affects the distribution of species traits, relationships of traits and species affinity with landscape variables were tested. Results: Among landscape variables, forest continuity, landscape forest cover and distance to nearest settlement altered the vegetation. These variables also influenced the distributions of species traits, i.e. life forms, life strategies, affinity to forest, dispersal modes, seed characteristics, flooding tolerance and Ellenberg indicator values for nitrogen, light, moisture and soil reaction. Nevertheless, physiographic area and habitat quality variables explained more variation in species composition. Landscape variables were more important in lowland areas. Forest continuity affected species composition only in lowlands. Conclusions: Although habitat quality and physiographic area explained more vegetation variability, landscape configuration was also a key factor influencing species composition and distribution of species traits. However, the results are dependent on forest geographical location, with lowland forests being more influenced by landscape variables compared to mountain forests.     

Variation in body colour in western Irish populations of Cepaea nemoralis (L.)

Samples of Cepaea nemoralis (L.) taken from sand-dune, lowland and mountain populations show differences in body colour. Overall, sand-dune populations are much paler than those in the other two areas. On sand-dunes, darker and more variable populations are found in densely vegetated habitats; in the lowland, darker forms occur further inland but there is no correlation with vegetation; mountain populations are darker and less variable on north- than south-facing slopes. Overall differences between areas can be ascribed to climatic selection and differences in variation within areas to niche width.     

Diversity of geometrid moths (Lepidoptera: Geometridae) along an Afrotropical elevational rainforest transect

Geometrid moths were investigated at 26 sites on 9 elevational levels along an elevational transect at Mt. Kilimanjaro (Tanzania), stretching from the fine‐grained mosaic of small agroforest plots with combined cultivation of trees, shrubs and crops at 1650 m through mountain rainforest to heathland at 3300 m. We sampled moths manually at light between 19 : 00 and 22 : 00 in the rainy seasons of March to May and October to January in the years 2000, 2001 and 2002. Along the transect, the composition of moth communities changed from a domination by Sterrhinae and Ennominae to a dominance of Larentiinae with increasing elevation. Overall, alpha diversity was very low compared to other tropical mountain regions. Fisher's alpha showed a maximum of 30 in the agroforest mosaic at 1650 m and decreased to values around 12 in the mountain rainforest. Communities of geometrid moths within the forest belt were significantly dissimilar from communities outside the forest. The diversity patterns on Mt. Kilimanjaro can be related to the young age, island‐like position and history of the mountain. These factors have led to the formation of a homogeneous upper mountain rainforest habitat which in turn houses homogeneous moth communities with a low diversity compared to habitats at lower elevations. Here, a heterogeneous habitat mosaic allowing the intrusion of savannah species into this former forest habitat may account for an increased diversity. In the heath zone above the forest, climatic conditions are very harsh, permitting only few specialists to thrive in this ericaceous woodland. Edge effects were discernible at the forest–heathland boundary where some moth species from heathland invaded the closed forest. At the boundary between agroforest and a forest mosaic of exotic Acacia and Eucalyptus forest plantations and natural mountain forest, diversity values remained low as the dominant species Chiasmia fuscataria accounted for far higher proportions than other dominant species in any of the other habitats.     

A key to the adult Costa Rican "helicopter" damselflies (Odonata: Pseudostigmatidae) with notes on their phenology and life zone preferences

We present a key to the Costa Rican species of Pseudostigmatidae, comprising three genera with the following species: Megaloprepus caerulatus, Mecistogaster linearis, M. modesta, M. ornata and Pseudostigma aberrans. Pseudostigma accedens, which may occur in the region, is also included. For each species we give a brief account of morphology, phenology and life zone preferences, including distributional maps based on more than 270 records. These are not all of the known specimens from the area, but a high enough number to give a relatively good picture of the distribution and status of the species. We found M. caerulatus to be active during the first half of the year in seasonal, tropical semi-dry lowland forest and tropical moist forest at mid-elevation, but like M. linearis, M. caerulatus was active all year round in non-seasonal, tropical wet lowland forest and tropical moist forest at mid-elevation. Mecistogaster modesta also flew year round in non-seasonal, tropical wet lowland forest and tropical moist evergreen forest at mid-elevation, and likewise in seasonal and non-seasonal, tropical premontane moist forest. Only a few findings, however, have been made of M. modesta in seasonal, tropical semi-dry deciduous forest and seasonal, tropical moist evergreen forest. Mecistogaster ornata was missing entirely from non-seasonal, tropical wet lowland forest and non-seasonal, tropical moist forest at mid-elevation, while this species was active year round in seasonal, tropical dry lowland forest and tropical semi-dry forest, as well as in seasonal, tropical moist evergreen forest and tropical premontane moist forest, both at mid-elevation. Pseudostigma aberrans has so far been found too few times in Costa Rica for any indication of flight time preference.     

Lowland panmixia versus highland disjunction: genetic and bioacoustic differentiation in two species of East African White-eye birds

East-African mountain forest species often occur in small and isolated populations, whereas species inhabiting the dry lowland savannahs exist in large and interconnected population networks. Taxa with closely related highland and lowland species, such as the East-African White-eye birds, allow testing for the potential effects of the two contrasting distribution patterns, mountain disjunction versus lowland panmixia. In this study, we compare the population genetic and bioacoustic differentiation of two representatives of the genus Zosterops: Zosterops poliogaster is exclusively found in forests at higher elevations; in comparison, Zosterops abyssinicus, only occurs in the dry and warm lowland savannahs. Both species were analysed across a similar geographical scale. Population genetic differentiation was inferred using the same set of 15 polymorphic microsatellite loci for both species. In addition, we quantitatively analyzed bioacoustic traits. Both data sets indicate a strong population differentiation among populations of the highland species, but an absence of differentiation in the lowland species. In addition, the lowland Z. abyssinicus was characterised by a twofold higher genetic diversity than detected for the highland Z. poliogaster. These two contrasting intraspecific population structures may reflect the opposite ecology and distribution of these species: the strong population isolation of Z. poliogaster resulting from long-term restriction to the cool and moist mountain forests at higher elevations has led to strong differentiation among local populations and resulted in a comparatively low level of intraspecific variability. In contrast, population panmixia in the lowland Z. abyssinicus provides a high level of gene flow allowing the maintenance of high genetic diversity and avoiding strong population structuring. These findings need to be considered when planning conservation actions.     

Population structure of common juniper in the Cis-Urals and Southern Urals

The population structure of common juniper (Juniperus communis L.) growing in the Cis-Urals and Southern Urals has been studied using 17 morphological traits of generative and vegetative organs. A multivariate analysis of ten coenopopulations has recognized three phenotypically different local populations: Cis-Ural, forest Cis-Ural forest-steppe, and Southern Ural mountain populations. The Cis-Ural forest population is strictly associated with lowland pine forests of the northwestern part of the Bashkir Cis-Urals. The Cis-Ural forest-steppe population is located in the northwestern part of the Bashkir Cis-Urals and the southeastern part of the Udmurt Cis-Urals. The Southern Ural mountain population is located in the central part of the Southern Urals and is associated mainly with mountain pine and dark coniferous forests. The last population is divided into forest and forest-edge subpopulations; the first one is represented by typical undergrowth locations, whereas the second is associated with open steppelike slopes and forest edges. In general, based on morphological traits of generative organs, the revealed local subpopulations hold an intermediate position between the Eastern European and Siberian populations of common juniper. Based on the morphological traits of vegetative organs, Cis-Ural populations are considered related to the populations of the European part of Russia, whereas the mountain Southern Ural population resemble Siberian populations. Concerning morphological traits of generative organs, the intrapopulation phenotypic diversity of common juniper is higher for mountain habitats; in the case of vegetative organs, the maximum diversity is observed for lowland habitats. The character of phenotypic differentiation determines the need to conserve the gene pool of common juniper of the Cis-Urals and southern Urals on a population basis.     

Invertebrate survey of coastal habitats and podocarp forest on Ulva Island,Rakiura National Park,New Zealand

Abstract

Inventory of the invertebrate fauna is important to establish taxonomic diversity, abundance and distribution, and hence the conservation of indigenous biodiversity. Invertebrate assemblages have been documented in some broadleaf‐podocarp forests and grassland habitats in New Zealand, but not in dense stands of coastal forest or in mature podocarp forest. This survey aimed to provide a taxonomic inventory of terrestrial invertebrates and their habitat associations on Ulva Island (Rakiura National Park, Stewart Island), an off‐shore sanctuary of significant conservation value in New Zealand. We systematically documented the invertebrate assemblages collected in ground litter and on tree trunks on the island. The invertebrate specimens identified represented 4 phyla, 6 classes, 25 orders and 62 species. The invertebrate fauna reported in this survey was distinct from those of lowland shrubland and broadleaved‐Nothofagus forests on the mainland, but shared species with that reported from another similar off‐shore island, Codfish Island (Whenua Hou).     

The role of tropical dry forest as a long-term barrier to dispersal: a comparative phylogeographical analysis of dry forest tolerant and intolerant frogs

We used a comparative phylogeographical approach to investigate the origins of the disjunct wet forest biota of the Golfo Dulce region along the Pacific slope of Costa Rica. This region is isolated by Pacific dry forests north and south and isolated from Caribbean wet forests by mountains. We studied three sympatric lowland frog species in the Craugastor fitzingeri species group that prefer wet forest but differ in their response to dry habitats. In dry forest, C. fitzingeri can survive along streams while C. crassidigitus and C. talamancae are entirely absent. We collected samples from across the ranges of all three species, and obtained mitochondrial DNA sequence data from the COI and cytochrome b genes. We observed significant phylogeographical structure in C. crassidigitus and C. talamancae, but much less in C. fitzingeri, demonstrating that mountain barriers and dry forest habitat have reduced mitochondrial gene flow in the strictly wet-forest species. Additionally, we discovered that the Golfo Dulce and Central Panama populations of C. crassidigitus appear to have diverged in the Pliocene or earlier, suggesting that the dry forest separating these populations is old. Our phylogenetic analysis of 12 of approximately 16 species of the C. fitzingeri species group suggests that the three lowland species are each other's closest relatives. Because of this shared phylogenetic history, we attribute the striking differences in phylogeographical structure to the different ecologies of the frogs. In summary, we find that what appear to be minor differences in the natural history of these three closely related species may profoundly impact the potential for dispersal, range size, and cladogenesis.     

Recent diversification in African greenbuls (Pycnonotidae: Andropadus) supports a montane speciation model

It is generally accepted that accentuated global climatic cycles since the Plio-Pleistocene (2.8 Ma ago) have caused the intermittent fragmentation of forest regions into isolated refugia thereby providing a mechanism for speciation of tropical forest biota contained within them. However, it has been assumed that this mechanism had its greatest effect in the species rich lowland regions. Contrary evidence from molecular studies of African and South American forest birds suggests that areas of recent intensive speciation, where mostly new lineages are clustered, occur in discrete tropical montane regions, while lowland regions contain mostly old species. Two predictions arise from this finding. First, a species phylogeny of an avian group, represented in both lowland and montane habitats, should be ordered such that montane forms are represented by the most derived characters. Second, montane speciation events should predominate within the past 2.8 Ma. In order to test this model I have investigated the evolutionary history of the recently radiated African greenbuls (genus Andropadus), using a molecular approach. This analysis finds that montane species are a derived monophyletic group when compared to lowland species of the same genus and recent speciation events (within the Plio-Pleistocene) have exclusively occurred in montane regions. These data support the view that montane regions have acted as centres of speciation during recent climatic instability.     

Above ground perennial plant biomass across an altitudinal and land-use gradient in Namaqualand,South Africa

This study set out to generate estimates of the standing perennial biomass for six different vegetation types, and associated upland and lowland habitats, across the altitudinal gradient presented by the Kamiesberg mountain range in the Namaqualand region of the Northern Cape Province of South Africa. Volume-biomass regressions, established for 94 perennial species accounting for 70–80% of the plant cover, were used to generate these estimations. Comparisons to other studies give similar findings, corroborating the method adopted. Biomass was found to vary significantly in relation to the altitudinal, and associated rainfall, gradient, as well as by habitat type where the rocky uplands have considerably more biomass that the sandy lowland habitats. An examination of the impact of sustained heavy grazing associated with a communal rangeland on this standing perennial biomass, showed a significant decrease in on the lowland habitats. This loss in biomass is principally of palatable species, with no evidence of a response in terms of perennial biomass. These findings point to degradation on the lowland habitats of the communal rangeland, with negative consequences for livestock farmers in the region.     

Phylogeny and evolution of the neotropical rodent genus Calomys: inferences from mitochondrial DNA sequence data

Field mice of the genus Calomys are small, mostly granivorous rodents common to several habitats in South America. To date, phylogenies for the genus have been proposed on the basis of morphological, chromosomal, and biochemical data, often with contradictory results due to incomplete species sampling or methodological shortcomings. In this paper, we propose relationships among 10 species of Calomys based on the complete cytochrome b gene sequence. Our analyses show that Calomys is constituted by two major clades, one mostly associated with mountain habitats with subsequent invasions to lowland habitats and another with species restricted to lowland habitats both north and south of the Amazon basin. The evolution of the genus was likely accompanied by a reduction of chromosome diploid numbers that occurred independently in each of the two evolutionary lineages. A "clock" calibrated on the split between Auliscomys and Loxodontomys suggests that the almost nonexistent fossil record for the genus greatly underestimates divergence times among its species.     

The ecology of lizards on Praslin Island, Seychelles

Ecological studies of five species of lizards were carried out on the island of Praslin, central Seychelles between July and September 1976. Quadrat sampling within four vegetation types revealed distinct habitat preferences between species. Phelsuma sundbergi occurred at highest densities in coastal vegetation, particularly coconuts Cocos nucifera , whilst Phelsuma astriata occurred at highest densities in intermediate forest, particularly in endemic palms such as Lodoicea maldivica. Within each vegetation type, both species preferred particular tree species. In coastal coconuts and Lodoicea in intermediate forest, the two species tended to occur in different sized trees. Calculations of niche breadth indicated greatest values for P. sundbergi in coastal vegetation and for P. astriata in intermediate forest. Both Phelsuma species showed a diurnal pattern of activity with some indication of temporal displacement between species in lowland and intermediate forest. Ailuronyx sechellensis was nocturnal with a relict distribution on Praslin, occurring in the quadrats on only three endemic palm species in intermediate forest. Mabuya sechellensis was by far the commonest and most widespread of the two skink species, with highest density in intermediate forest where it was probably favoured by the high proportion of palm leaf litter. Scelotes gardineri was only recorded in lowland and intermediate forest, where it occurred at low densities. Both species were diurnal, although the latter was never recorded after midday. Intermediate forest had the highest lizard biomass and species diversity whilst eroded land had the lowest. The possible origins and relationships of the two closely related Phelsuma species are discussed, together with the roles that habitat modification, predation, and interspecific competition may play.     

The contribution of different habitat types to species diversity of Baltic uplift islands

Islands are vulnerable ecosystems worldwide, increasingly exposed to human pressure, global climate change and invasive species. Thus, understanding island species diversity is key for nature conservation. Recent studies on insular plant communities indicated that habitat-specific species composition and richness might largely affect diversity patterns observed at the island scale. In consequence, habitat-based approaches are needed to (i) estimate how environmental changes at the habitat scale may affect island diversity, and to (ii) estimate the contribution of different patches of the same habitat to island diversity with respect to habitat-specific environmental constraints.In the present study, we tested these habitat-to-island diversity relationships for shoreline habitats (brackish reeds, salt marsh, rocky shore, tall herbs) and island interior habitats (rocks, semi-natural grassland, pioneer forest, coniferous forest, mixed forest) using 108 islands of three Baltic archipelagos in Sweden. These islands differed in terms of island-scale variables describing effects of island configuration and distance, and habitat-scale variables representing the effects of habitat area, abiotic environment and land-use.The studied habitats differed in their contribution to island species diversity, called habitat specificity. Shoreline habitats shared many common specialist species adapted to extreme conditions like sea salt or bird grazing, while habitats of the island interior harbored mainly species adapted to the specific conditions of a single habitat. We found high variability in habitat specificity as a consequence of habitat-specific environmental factors. Variability was highest for grasslands, where it was related to abandonment and soil fertility, stressing the importance of grassland management for maintaining island biodiversity. Habitats with high habitat specificity through either high species richness or many habitat-specific specialists should be the primary targets for biodiversity management.     

Pervasive threats within a protected area: conserving the endemic birds of São Tomé, West Africa

The importance of the rainforests on the island of São Tomé for biodiversity is well known. However, the area only recently received full legal protection as a National Park and currently few resources are available to enforce that legislation. With rapid economic development forecast for the island, active conservation efforts are essential. Here we study the distribution and density of the island's endemic bird species, including nine that are Globally Threatened, within the National Park. Sites, covering the full range of primary forest types, were surveyed using distance sampling methods. No introduced species were observed. The highest number of species, including eight Globally Threatened species, were found in lowland rainforest, although many were infrequently encountered. Higher altitude sites were less diverse, but supported some of the common endemic species at extremely high densities. The least diverse assemblage, with generally lower species population densities, occurred at the most accessible mid-altitude forest site. Distance from settlements was a key explanatory variable for the presence of all Globally Threatened species, indicating that human habitation has negative effects on the suitability of nearby forest habitats. This suggests that, as infrastructure improvements proceed, populations of endangered species will come under growing pressure. Integrating the needs of biodiversity conservation and development represents a major challenge for many biodiverse countries and on São Tomé, as elsewhere, may best be achieved by preserving the still intact functioning forest ecosystem.     

The maintenance of extensively exploited pastures within the Alpine mountain belt: implications for dung beetle conservation (Coleoptera: Scarabaeoidea)

The abandonment of marginal areas and the decline of traditional small-scale cattle-breeding is bringing about substantial changes in many areas of the Alpine mountain belt. Extensively exploited pastures, which were colonized by typical ecological communities, are now being replaced by shrublands and forests. In this paper, we evaluate the effectiveness of the maintenance of small pastures for dung beetle (Scarabaeoidea: Geotrupidae, Aphodiidae, Scarabaeidae) conservation in a protected area of the north-western Italian Alps. From July until September 2007, we studied the dung beetle coenosis in the main habitats of the mountain belt: beech forest, scots pine forest, mountain pine forest and pasture. Twenty-three species were sampled. We compared abundance, species richness, α-diversity, evenness and species turnover between the four habitat assemblages and performed a correspondence analysis based on dominance patterns of the three families. Two main ensembles were identified: a pasture ensemble, highly diverse and dominated by Scarabaeidae, and a closed habitats ensemble, less structured and dominated by Aphodiidae and Geotrupidae. Combining habitat specificity and fidelity, the IndVal method showed that half of the species collected made a strong habitat selection toward pasture, while only C. granarius (Aphodiidae) was a reliable indicator of woody areas. We therefore conclude that the loss of extensively exploited pastures may bring about fundamental changes in dung insect communities. In protected areas, a management policy intended to preserve the traditional, sustainable human activities within the mountain belt is highly recommended for the conservation of these insects.