Conservation value of roadside verges for stenotopic heathland Carabidae: corridors or refugia?

It has been shown that, in The Netherlands, sandy roadside verges may aid the survival of stenotopic Carabidae associated with open habitats such as heathland and sandy grassland. It is suggested that roadsides serve as corridors linking heathland fragments, thus maintaining a metapopulation structure, which helps avoid regional extinction. However, a counter-argument is that verges encourage dispersal into unsuitable habitats, and function as sinks in which individuals die without reproducing. The stenotopic species recorded from roadside verges but not in adjacent open habitats all had high powers of dispersal. The least successful dispersers were found only in open habitats and in verges closely connected to them. From this, the conclusion that verges function as corridors can be drawn. We reinterpret these data to suggest that stenotopic species were breeding on the roadside verges, and that they acted as refugia rather than corridors. We also present evidence from a roadside verge in the Breckland region of East Anglia, England. We show that many stenotopic species breed on the verge, which thus functions as a refugium. For Carabidae dependent on early successional habitats, survival may be more likely on disturbed roadside verges than on nature reserves in the absence of regular management disturbance.     

Bird communities of highway verges: Influence of adjacent habitat and roadside management

We have investigated the effects of landscape traversed and roadside structure on the use of highway verges by birds. Three contrasted landscapes were chosen in terms of human land use and vegetation structure: an intensive farmland, a pine plantation, and a matoral. The roadside sections varied in vegetation structure, width and profile. We recorded birds present in roadsides and adjacent habitats by transect counts over all seasons. Roadside bird species appeared for a great part similar to those of adjacent habitats. However, diversity and abundance in verges did not depend on that of adjacent habitats. Woody roadsides were comparable to hedges, as trees (and shrubs) in verges enhanced species richness and abundance of birds in the farmland and woodland sites. Width and profile of verges had less influence. In all sites, typical species of the habitat traversed partly avoided roadsides. On the contrary, numerous species associated with ‘rare’ habitats in one site preferred roadsides, provided that verge vegetation contrasted with the dominant habitat. It is concluded that birds responses to highways can vary greatly with landscape traversed and verge vegetation. Highway verges could be favorable to birds, if they constitute a complementary habitat to the dominant habitat within a landscape.     

Oil pipeline corridor through an intact forest alters ground beetle (Coleoptera: Carabidae) assemblages in southeastern Ohio

Litter-dwelling ground beetle (Coleoptera: Carabidae) assemblages were monitored 1 yr after the construction of a corridor for installation of an oil pipeline along a xeric ridge-top forest in southeastern Ohio. After the creation of the corridor, three distinct habitats were evident in these sites: open corridor, ecotone areas around the corridor, and undisturbed forest interior. Carabidae were collected using directional pitfall traps that were placed parallel and perpendicular to the corridor in each of the three habitats. Results indicate that more carabids were present in the ecotone than in the other two habitats. Carabid diversity as estimated by rarefaction was highest in the corridor followed by ecotone and forest interior. Generalist and forest specialists such as Synuchus impunctatus (Say), Carabus goryi Dejean, and Pterostichus trinarius (Casey) were present in greater numbers in the forest interior and ecotone assemblages. In contrast, open-habitat specialists such as Harpalus pensylvanicus (DeGeer) and Selenophorus opalinus (LeConte) were present in greater numbers in the corridor assemblages. Carabid assemblages of the corridor were distinct from those of the ecotone and forest interior, whereas the latter two habitats had very similar assemblages. The successional pathway of the corridor carabid assemblage will therefore be likely different from that of the forest interior and ecotone. Overall, results indicate that construction of the oil pipeline corridor had significant short-term effects on the carabid numbers, diversity, and species composition because of ensuing habitat changes and fragmentation of the forest.     

Species decline--but why? Explanations of carabid beetle (Coleoptera,Carabidae) declines in Europe

We investigated some of the causes of ground beetle decline using atlas data from Belgium, Denmark and the Netherlands, countries in which natural environments have all but disappeared. We used ordinal regression to identify characteristics that are significantly correlated with the decline of carabid beetle species over the last 50-100 years, using a stepwise selection procedure to select the optimal model according to the Akaike Information Criterion. The results showed that large-bodied carabid populations have declined more than smaller ones, possibly because of their lower reproductive output and lower powers of dispersal. Habitat specialist populations (i.e. species with small niche breadths) have also decreased more than habitat generalist populations. Species with both long- and short-winged individuals have been less prone to decline than those that are exclusively either short-winged or long-winged. Dimorphic species may survive better in highly altered environments because long-winged individuals are good at dispersing between suitable habitats and short-winged individuals are good at surviving and reproducing in these newly colonised habitats. Finally, populations of large carabids associated with coastal, woodland or riparian habitat types were less prone to decline than populations of large carabids associated with various, open or grassland habitat types. The pattern is reversed for carabid species smaller than 8 mm in size. These results are explained in the context of habitat restoration and destruction in these highly modified western European countries.     

Reduced dispersal propensity in the wingless waterstrider Aquarius najas in a highly fragmented landscape

Dispersal behaviour of animals in fragmented habitats has generated intensive theoretical attention but empirical data on the evolution of dispersal are still relatively scarce. Theory predicts reduced dispersal propensity in small and isolated habitat patches. We tested these predictions in the waterstrider Aquarius najas, a wingless species with special habitat demands. Flightlessness constrains insect dispersal and as a stream specialist A. najas cannot survive on still water. Lakes therefore represent a dispersal barrier for this species. We measured dispersal propensity of the waterstrider A. najas which originated from ten fragmented populations. In the experiment, we transplanted laboratory-grown individuals to the field. We did not find differences between sex in dispersal propensity. However, we found that waterstriders that originated from small and isolated patches moved less than individuals from large and more continuous habitats. This suggests that the cost of dispersal over hostile surrounding habitats may be high. We conclude that a low likelihood of dispersal in A. najas is an adaptation to small and isolated stream habitats.     

Road verges as habitat for small mammals in Britain

There are 370 000 kilometres of roads in Great Britain, mostly bordered by a verge that is potential habitat for small mammals. The present study assessed the importance of road verges as small mammal habitat and investigated the influence of some key features on rodent abundance. Five rodent and three shrew species were live‐trapped on 14 road verges in late summer 1994 and nine verges in autumn 1996 in north Cambridgeshire, UK. On average, between three and four species were captured per verge. Bank Voles Clethrionomys glareolus, Wood Mice Apodemus sylvaticus and Field Voles Microtus agrestis were the most abundant species, with mean densities of 45.5, 40.2 and 29.5 animals km^–1 in summer and 52.8, 181.9 and 47.2 animals km^–1 in autumn. Numbers varied between verges and this was significantly correlated with particular features on the verge. Bank Vole and Field Vole numbers showed a significant positive correlation with the dimensions of hedges and the width of the tall grass area, respectively. Wood Mice were also more numerous on verges with big hedges but the relationship between mouse abundance and verge structure was complex. The number of mice in 1994 was positively and significantly correlated with hedge features and with the width of the short grass sightline, whereas, in autumn 1996, they were only significantly correlated with total verge width (positive association) and ditch width (negative association).     

Effects of habitat transitions on rainforest bird communities across an anthropogenic landscape mosaic

We compared bird community responses to the habitat transitions of rainforest‐to‐pasture conversion, consequent habitat fragmentation, and post‐agricultural regeneration, across a landscape mosaic of about 600 km² in the eastern Australian subtropics. Birds were surveyed in seven habitats: continuous mature rainforest; two size classes of mature rainforest fragment (4–21 ha and 1–3 ha); regrowth forest patches dominated by a non‐native tree (2–20 ha, 30–50 years old); two types of isolated mature trees in pasture; and treeless pasture, with six sites per habitat. We compared the avifauna among habitats and among sites, at the levels of species, functional guilds, and community‐wide. Community‐wide species richness and abundance of birds in pasture sites were about one‐fifth and one‐third, respectively, of their values in mature rainforest (irrespective of patch size). Many measured attributes changed progressively across a gradient of increased habitat simplification. Rainforest specialists became less common and less diverse with decreased habitat patch size and vegetation maturity. However, even rainforest fragments of 1–3 ha supported about half of these species. Forest generalist species were largely insensitive to patch size and successional stage. Few species reached their greatest abundance in either small rainforest fragments or regrowth. All pastures were dominated by bird species whose typical native habitats were grassland, wetland, and open eucalypt forest, while pasture trees modestly enhanced local bird communities. Overall, even small scattered patches of mature and regrowth forest contributed substantial bird diversity to local landscapes. Therefore, maximizing the aggregate rainforest area is a useful regional conservation strategy.     

Area mediated shifts in bird community composition: a study on a fragmented Mediterranean grassland

The effects of habitat fragmentation on birds have often been studied in forest specialist species. Here we aimed at comparing the response of open habitat birds within a range of habitat specialization. The study area was a Mediterranean pseudo-steppe, designated as important for conservation yet fragmented by tree encroachment. We defined bird species dependency on steppe-like habitat by a correspondence analysis, allowing us to distinguish between specialists, generalists and scrubland species. We studied species abundance in relation to fragment area, testing whether species representation in fragments differed from those in continuous habitat. This analysis showed a contrasted response to fragment size between “open habitat” specialist species and generalist ones. Open habitat species were under-represented in the smallest fragments, while generalist were over-represented in small fragments in comparison to their distribution in continuous habitats. We discuss how these results can be linked to species habitat requirements. We find that scrubland species seem to be favoured by encroachment of woody vegetation, as they are able to explore and use the wooded matrix; however specialist species are restricted to open patches and are sensitive to a reduction in patch size. This allows us to predict how different species can exhibit a different sensitivity to habitat fragmentation.     

The effects of forestry on carabid beetles (Coleoptera: Carabidae) in boreal forests

As compared to natural forests, managed boreal forests are younger, more homogeneous in terms of tree age and species composition, and consist of smaller fragments. Here we examine the effects of such characteristics caused by forestry on carabid beetles (Coleoptera, Carabidae) in the boreal region. The main results are the following. (1) Fragmentation of forests and the size of a fragment appear not to be crucial for the survival of the majority of forest carabids, as they tend to be distributed over various successional stages, but species requiring old-growth habitats suffer. (2) For carabids there appear to be no or very few edge specialist species, and forest-open land edges appear to be effective barriers for species associated with forest or open habitat. However, generalist species easily cross the edge, and edges of forest fragments may be invaded by species from the surrounding open habitat. (3) Habitat change following clear-cutting dramatically changes the composition of carabid assemblages: species restricted to mature forests disappear and open-habitat species invade, while habitat generalists survive at least in the short term. Carabid diversity can probably best be maintained if forest management mimics natural processes, maintains natural structures and includes the natural composition of vegetation and other structural elements (such as dead wood) within the stands, provided that these forest features can be maintained and recreated through forest management practices. At a larger scale, the whole spectrum of forest types and ages (especially old-growth forests), and different successional processes (especially fire) should be maintained. These require the development and use of innovative logging methods, and the planning, implementation, and assessment of landscape-scale ecological management strategies.     

Effects of narrow roads on the movement of carabid beetles (Coleoptera,Carabidae) in Nopporo Forest Park,Hokkaido

Forest-dwelling carabid beetles that have no flight ability were studied using mark-recapture methods in late-June to mid-October 2007. This study was done to determine the effects of narrow roads in Nopporo Forest Park, Hokkaido on carabid beetle movement and habitat use. The investigation was conducted at four sites: one site was an abandoned grassy road with a width of 3.5 m, two sites were gravel roads with widths of 3.5 and 4.5 m, and another site was an asphalt-paved road with a width of 4.5 m. A total of 3,580 individuals from six species of carabid beetles were collected using dry pit-fall traps, and recapture rates ranged from 6.1 to 36.2%. All examined roads acted as barriers against the movement of Leptocarabus arboreus ishikarinus. All roads, except the abandoned grassy road, acted as a barrier against Carabus granulatus yezoensis movement. Forest–roadside verge comparisons demonstrated that some carabid beetles avoid even narrow roadside verges. Harmful effects increase with increasing road width and both paved roads and narrow roads negatively affect the movement of carabid beetles inhabiting the bordering forest. Therefore, forest specialist beetles are influenced by a barrier effect that starts at the forest road verge, and this barrier effect may be exacerbated by vehicular traffic. Therefore, these barrier effects on carabid beetles should be considered when planning and implementing road construction and maintenance in forests.     

Comparing movement of four butterfly species in experimental grassland strips

In contrast to many studies on the habitat quality of road verges for butterflies in relation to management regimes, little is known about whether road verges also function as corridors linking fragmented grassland habitats. We experimentally compared movements of four model species, two small blues and two medium browns, with one habitat specialist and one habitat generalist in each size and phylogenetic category. A total of 425 individual butterflies were caught and translocated to an experimental arena with three 2 × 30 m grassland strips that approximated road verges; one with adult feeding resources, one sheltered from the wind, and one without food and shelter. Movements in grassland strips were compared to movements in continuous grassland habitat. Results indicated that (1) individuals did not use the low-quality strip, (2) only specialists used strips but not in the same way according to their size and phylogenic category, and (3) strip use could not be predicted from habitat selection. This finding supports the idea that corridors of intermediate quality are the most efficient to promote dispersal rates in fragmented landscapes. Road verges cover 250,000 ha in Sweden, which is nearly the total amount of seminatural grasslands. Our results suggest that, to benefit butterfly dispersal among grassland patches, road verges should be managed to create a more favourable microclimate (e.g. sheltered from wind, high temperatures).     

Effects of natural and seminatural elements on the composition and dispersion of carabid beetles inhabiting an agroecosystem in Northern Italy

The natural and seminatural components of agricultural landscapes play a key role in maintaining a high level of biodiversity. Being the Po Valley one of the most human‐dominated and intensively cultivated landscapes in Europe, we investigated the effect of no‐crop habitats on carabid richness and composition and evaluated the role of tree row as corridor for forest carabid dispersion. Carabids were sampled with 70 pitfall traps arranged in 35 sampling plots along three parallel transects (80, 100, and 140 m long) and encompassing five different habitats: tree row, tree row edge, grassland, forest edge, and forest. We found 5,615 individuals belonging to 55 species. Despite the similarity in species richness, all the habitats investigated showed a peculiar and distinct species assemblage. The main distinction was between the "open habitat" cluster composed of grassland and tree row edge and the “forest" cluster composed of forest, tree row, and forest edge. We found that forest species are able to penetrate the grassland matrix up to 30 m from the forest edge and that a distance of no more than 60 m between tree row and forest can allow the passage of up to 50% of the forest species. Beyond this distance, the grassland matrix becomes a barrier, preventing them from reaching other suitable habitats. Our findings confirm the importance of maintaining different types of natural habitats to significantly increase biodiversity in an intensively cultivated agroecosystem and demonstrated the role of linear elements as a corridor and “stepping stones” for many forest species.     

Large and least isolated fragments preserve habitat specialist spiders best in dry sandy grasslands in Hungary

The role of fragment size, isolation and habitat diversity in the conservation of spider assemblages living in fragmented landscape were studied in dry sandy grasslands (East Hungary, Nyírség). Spiders were collected using pitfall traps at eight dry grassland fragments from 2001 to 2009 from March to October fortnightly. We tested the rules of island biogeography, which suggest that the species richness increases with the size and decreases with the isolation of fragments. The habitat diversity is an important factor for species richness, since large areas usually have more habitats; therefore, the number of species may be higher in these areas. During the 9-year study period, altogether 10,544 individuals belonging to 106 species were collected. Contradicting the classical theory, we found a significant negative relationship between the total number of spider species and the grassland size, while the ratio of sandy grassland specialist spider species increased with fragment size. The relationship between the ratio of generalist species and the fragment size was not significant. The overall species richness and the isolation of studied grasslands did not show a significant relationship. The ratio of sandy grassland specialist species decreased, while the ratio of generalist species increased with the increasing of isolation. The habitat diversity did not show any effect on spider species richness. We concluded that to conserve the habitat specialist species it is recommended to preserve the large and least isolated grassland fragments, furthermore to increase the size of small fragments with the restoration of the adjacent croplands.     

Single large AND several small habitat patches: A community perspective on their importance for biodiversity

The debate whether single large or several small (SLOSS) patches benefit biodiversity has existed for decades, but recent literature provides increasing evidence for the importance of small habitats. Possible beneficial mechanisms include reduced presence of predators and competitors in small habitat areas or specific functions such as stepping stones for dispersal. Given the increasing amount of studies highlighting individual behavioral differences that may influence these functions, we hypothesize that the advantage of small versus large habitat patches not only depends on patch functionality but also on the presence of animal personalities (i.e., risk-tolerant vs. risk-averse). Using an individual-based, spatially-explicit community model, we analyzed the diversity of mammal communities in landscapes consisting of a few large habitat islands interspersed with different amounts and sizes of small habitat patches. Within these heterogeneous environments, individuals compete for resources and form home-ranges, with only risk-tolerant individuals using habitat edges. Results show that when risk-tolerant individuals exist, small patches increase species diversity. A strong peak occurs at approximately 20% habitat cover in small patches when those small habitats are only used for foraging but not for breeding and home-range core position. Additional usage as stepping stones for juvenile dispersal further increases species persistence. Overall, our results reveal that a combination of a few large and several small habitat patches promotes biodiversity by enhancing landscape heterogeneity. Here, heterogeneity is created by pronounced differences in habitat functionality, increasing edge density, and variability in habitat use by different behavioral types. The finding that a combination of single large AND several small (SLASS) patches is needed for effective biodiversity preservation has implications for advancing landscape conservation. Particularly in structurally poor agricultural areas, modern technology enables precise management with the opportunity to create small foraging habitats by excluding less profitable agricultural land from cultivation.     

Road effects on demographic traits of small mammal populations

Recent studies have highlighted the positive effects of road verges on the abundance of small mammals. However, most of these studies occurred in intensively grazed or cultivated areas, where verges were the last remnants of suitable habitats, which could mask the true effects of roads on population traits. We analysed the effects of roads on small mammal populations living in a well-preserved Mediterranean forest. We used the wood mouse (Apodemus sylvaticus) as a model of forest-dwelling small mammals that probably are among the species most affected by road clearings. Our study compared populations in similar habitat areas with and without road influence. We assessed abundance, survival and temporary emigration using extended Pollock’s robust design capture-recapture models. Moreover, we analysed population turnover, sex ratio, age structure and body condition. We found that wood mouse abundance and body condition were lower at the road bisected area, whereas the remaining population traits were similar. This suggests that the reduced habitat availability and quality due to the physical presence of the road and verge vegetation clearing are the main drivers of demographic differences in wood mouse populations between areas. Nevertheless, our results also suggest that in high-quality habitats surrounding national roads, wood mouse populations present similar dynamics to others living in undisturbed areas, despite the decrease in abundance and body condition. Overall, the often-reported increased small mammal abundance in road surroundings should not be generalized independently of habitat quality or to other population traits.     

Avian Use of Restoration Plantings along a Creek Linking Rainforest Patches on the Atherton Tablelands, North Queensland

Rainforest fragmentation and creek degradation are major problems on the Atherton Tablelands in North Queensland. Lake Barrine and Wooroonooran National Parks are separated by a 1.5‐km length of Toohey's Creek which has been cleared except for small remnant rainforest patches. The creek has been fenced and replanted to create a corridor between the two national parks. The aim of this study was to monitor the success of the corridor in providing habitat for rainforest birds. Avian communities were monitored in the corridor and in adjacent rainforest sites using point counts over the period of reestablishment of a continuous rainforest canopy. In terms of bird community composition, within 3 years of planting, the replanted sites were on a trajectory toward the rainforest sites. Although specialist rainforest species were not yet found in the plantings, after 3 years those species which mainly inhabit rainforest but also utilize edges and riparian strips were found in similar numbers in the plantings and in the small remnant patches along the corridor. Frugivorous species were quite abundant even in the youngest plantings. Thus, replanting of rainforest trees along Toohey's Creek has begun to provide habitat for some rainforest birds. It also shows good potential to act as a corridor to connect the previously isolated patches of rainforest for these species and to develop in habitat complexity as the frugivorous birds bring in rainforest plant species from nearby areas.     

Roadside verges and cemeteries: Comparative analysis of anthropogenic orchid habitats in the Eastern Mediterranean

Several important habitats have become threatened in the last few centuries in the Mediterranean Basin due to major changes adopted in land‐use practices. The consequent loss of natural and seminatural orchid habitats leads to the appreciation of small anthropogenic habitats, such as cemeteries and roadside verges. Colonization of cemeteries and roadside verges by orchids has long been known, but no study to date compared the suitability of these two anthropogenic habitats for orchids. Therefore, in this paper our aim was to survey cemeteries and roadside verges and to compare these two habitats regarding their role in conserving Mediterranean terrestrial orchids. We conducted field surveys in three Mediterranean islands, Cyprus, Crete, and Lesbos, where both cemeteries and roadside verges were sampled on a geographically representative scale. We found a total of almost 7,000 orchid individuals, belonging to 77 species in the two anthropogenic habitat types. Roadside verges hosted significantly more individuals than cemeteries in Crete and Lesbos, and significantly more species across all three islands. Our results suggest that although cemeteries have a great potential conservation value in other parts of the world, intensive maintenance practices that characterized cemeteries in these three islands renders them unable to sustain valuable plant communities. On the other hand, roadside verges play a prominent role in the conservation of Mediterranean orchids in Cyprus and Greece. The pioneer status of roadside verges facilitates their fast colonization, while roads serve as ecological corridors in fragmented landscapes.     

Fruit-feeding butterflies in edge-dominated habitats: community structure,species persistence and cascade effect

As old-growth forests are converted into edge-affected habitats, a substantial proportion of tropical biodiversity is potentially threatened. Here, we examine a comprehensive set of community-level attributes of fruit-feeding butterfly assemblages inhabiting edge-affected habitats in a fragmented Atlantic forest landscape devoted to sugar cane production. We also explored whether the consequences of habitat loss and fragmentation can interact and cause cascading ecosystem changes, with the pervasive simplification of tree assemblages inhabiting edge-dominated habitats, altering fruit-feeding butterfly persistence. Butterflies were sampled in three forest habitats: small fragments, forest edges and patches of forest interior of a primary forest fragment. Assemblage attributes, including taxonomic composition, correlated to some patch (patch size) and landscape (such as forest cover) metrics as well as habitat structure (tree density and richness). Fruit-feeding butterfly assemblages in the forest interior differed from those in small fragments due to an increased abundance of edge-specialist species. On the other hand, several forest-dependent species were missing in both small fragments and forest edges. Our results suggest that edge-affected habitats dominated by pioneer tree species support taxonomically distinct assemblages, including the presence of disturbance-adapted species, and butterfly community structure is highly sensitive to fragmentation- and plant-related variables, such as forest cover and pioneer tree species. In this way, while the establishment of human-modified landscapes probably results in the local extirpation of forest-dependent species, it allows the persistence of disturbance-adapted species. Thus, forest-dependent species conservation and the plant–animal interaction webs they support could be improved by retaining a significant amount of core forest habitat.     

The impact of hedge-forest connectivity and microhabitat conditions on spider and carabid beetle assemblages in agricultural landscapes

Agricultural intensification in terms of decreasing landscape complexity and connectivity has negatively affected biodiversity. Linear landscape elements composed of woody vegetation like hedges may counteract this negative trend by providing habitats and enhancing habitat connectivity for different organisms. Here, we tested the impacts of habitat type (forest edges vs. hedges) and hedges’ isolation (connected vs. isolated hedges) from forests as well as microhabitat conditions (percentage of bare ground and width) on trait-specific occurrence of ground-dwelling arthropods, namely spiders and carabids. Arthropods were grouped by habitat specialisation (forest vs. open-habitat species vs. generalists), hunting strategy (web-building or hunting spiders) and dispersal ability (wing morphology of carabids). Spider and carabid assemblage composition was strongly influenced by habitat type and isolation, but not by microhabitat conditions. Activity density of forest species and brachypterous carabids was higher in forest edges compared to hedges, whereas open-habitat species and macropterous carabids showed reverse patterns, with no effects of isolation. Occurrence of generalist carabids, but not spiders, was higher in hedges compared to forest edges. Habitat type and isolation did not affect spiders with different hunting strategy. Microhabitat conditions were less important for spider and carabid occurrence. Our study concludes that on a landscape scale, type of linear woody habitat is more important for arthropod occurrence than isolation effects and microhabitat conditions, depending on traits. Hedges provide refuges for species specialised to open habitats and species with high dispersal ability, such as macropterous carabids. Forest edges enhance persistence of species specialised to forests and species with low dispersal ability, such as brachypterous carabids.     

Living on the verge: are roads a more suitable refuge for small mammals than streams in Mediterranean pastureland?

The retention of natural habitat corridors is a useful and practical conservation tool that can attenuate the effects of habitat loss and fragmentation on wildlife. Linear structures may contribute to the conservation of biodiversity by providing additional habitats for small fauna living in highly modified environments. We assessed the importance of road verges as refuge areas for small mammals, in highly intensified grazed pastures, within a Mediterranean landscape and compared the role of road verges as refuges with that of riparian galleries, which have been described as important shelter locations for small fauna. For this purpose, a small mammal trapping study was undertaken on two road verges and beside two small streams in southern Portugal. We captured 457 individuals of five different species, with Mus spretus the most common species captured, followed by Crocidura russula. Captures were 4.6-fold higher immediately beside both roads and streams than 12 m away in the surrounding matrix. Individuals captured in the matrix presented a smaller body size and lower body condition, suggesting that this suboptimal habitat is occupied mainly by subadults. M. spretus was 46% more abundant by roads than by streams, while C. russula was present in similar numbers in both habitats. M. spretus individuals were larger near streams but exhibited no difference in body condition between habitats. C. russula had a better body condition and slightly higher body lengths at roadsides. Our results show that roadside verges in intensively grazed Mediterranean landscapes act as important refuges and constitute equally vital habitats for small mammals as do riparian vegetation strips in landscapes where other suitable habitats are scarce.