Morphological differences between two semi-arboreal, bird-eating island pitvipers, Bothrops insularis and Gloydius shedaoensis, and their mainland relatives were investigated. Mitochondrial DNA sequence data for two genes show B. insularis to be rooted within B. jararaca. Bothrops insularis has a more anterior heart, a relatively longer tail, a longer head, and shorter fangs than B. jararaca. The greater head length is paralleled in the ecologically similar G. shedaoensis. Increased head size may represent an adaptation to the abundance of larger food items (migratory passerine birds), providing a selective advantage to snakes able to switch to larger prey at an earlier age. Furthermore, B. insularis and G. shedaoensis have converged on similar body sizes from opposite ancestral states. Other characters, including fang length, tail length and size of neonates do not show parallel variation in G. shedaoensis and B. insularis, suggesting that caution is required when interpreting character state shifts coinciding with ecological shifts in a single species only.

After their oceanic migration, juvenile European eels Anguilla anguilla enter estuaries as glass eels, develop into pigmented elvers and migrate into fresh water. Fisheries often transfer such eels abruptly between salinities, principally glass eels and elvers from estuarine to fresh water. It is usually assumed that survival rates are high, but this required systematic investigation. Survival was found to be 100% over 21 days of glass eels and semi-pigmented elvers transferred abruptly from estuary conditions into fresh water, 50% sea water and full sea water. Fully pigmented elvers, however, showed significantly reduced survival when transferred into sea water. Salinity preference experiments with juvenile eels have historically been inconclusive. Here, in a choice chamber design, a clear developmental shift in salinity preference was found, with glass eels preferring 100% sea water, semi-pigmented elvers showing no clear preference and fully pigmented elvers preferring fresh water. We conclude that eel fisheries enhancement by abrupt transfer of juveniles among salinities is largely vindicated. In addition, developmental shifts in salinity preference have been clarified and this aids in the interpretation of eel migration patterns.

Feeding habits of smooth-coated otters Lutra perspicillata were studied by analysing spraints collected from around the Periyar Lake within the Periyar Tiger Reserve, Kerala, India. Remains were identified by the comparison of body parts with a reference collection of appropriate prey species. Two methods were used to estimate the proportions of different prey consumed: frequency of occurrence and score-bulk estimate. The number of prey types in a spraint varied between one and seven. More prey species were found during low water levels (March–April). Fish was the major prey, followed by frogs, crabs, birds and insects. Throughout the study, the exotic tilapia and European carp constituted the major prey in the diet. The quantity of native Deccan mahsheer and endemic Periyar barb in the diet was insignificant. A higher intake of bottom-dwelling catfish was observed during periods of low water, probably because of the increased efficiency of otters to forage at these levels. By consuming large amounts of exotic species, such as tilapia and European carp, otters might contribute to the control of the rapid expansion of these species in the Periyar Lake and adjacent water bodies in the reserve.

There are more taxa of wild pig in South-east Asia (SEA) than in any other comparable area in the world, but the number of species and subspecies is still uncertain. The taxonomy of some wild pig populations distributed in Malaysia, Indonesia and the Philippines, was investigated using molecular and morphometric techniques. Our results suggest the existence of two main evolutionary clades that are likely to have diverged during the Pliocene in SEA: one including wild pig populations distributed in the Philippines (Sus cebifrons) and Sulawesi (S. celebensis); the other including the Indonesian and Malaysian bearded pigs (S. barbatus), and the widespread Eurasian wild boar (S. scrofa). A possible scenario for pig speciation in SEA is developed, and the need for a taxonomic revision of bearded pigs suggested, particularly concerning the taxonomic status of S. b. ahoenobarbus from Palawan Island, and the existence a new species distributed in the Tawi Tawi Islands (Philippines). Unexpectedly, bearded pigs in the Malay Peninsula are closely related to the Bornean population, but distinct from Sumatran S. barbatus, and they should be considered as belonging to a different subspecies.

The behaviour and survival of captive-born least weasels Mustela nivalis nivalis released in nature were reported in this study. Altogether 27 captive-born and six wild-caught weasels were equipped with radio transmitters and their survival was studied as a function of age and season. The daily mortality rate was generally higher in captive-born than in wild-caught weasels. The survival of captive-born weasels was highest if they were released during the summer and hence when they were young. Differences in the behavioural patterns of captive-born and wild-caught weasels were apparent. Captive-born weasels were often more visible in the field and were less timid than wild-caught weasels. Results suggest that to be successful, restocking of weasels should be done with young animals, which are preferably released during the season with abundant food resources.

The leopard cat Prionailurus bengalensis is a relatively common, small felid ranging throughout Asia. During the past 15 years the natural history of leopard cats has been the topic of five studies; however, the mean sample size of study animals has been low ($ \bar{x}$ = 6.8, range 4–10). We report on the most comprehensive study of leopard cats to date. Between June 1999 and February 2003, 20 leopard cats (14 males and six females) were radio-collared and tracked from 3 to 20 months in Phu Khieo Wildlife Sanctuary, Thailand. Spatial organization of the leopard cats was characterized as weakly territorial with similar intrasexual range sizes and minimal seasonal variation. Mean (± SD) annual home-range (95% minimum convex polygon) size for males was 12.4 km2 (n = 1211 locations, ± 7.1, range 2.2–28.9), whereas females exhibited a mean home-range size of 14 km2 (n = 470, ± 12.2, range 4.4–37.1). Core area (50% minimum convex polygon) averaged 2.0 km2, and the mean 1-day movement was 1298 m (± 981, range 35–8653). Habitat use was generally in proportion to occurrence, and the mean activity (52%) was arrhythmic with crepuscular and nocturnal peaks. Analysis of scats indicated that murids dominated leopard cat diet.

The purpose of this study was to investigate if there was any species-specific type and structure of sleeping sites of common marmosets Callithrix jacchus and their key determinant. The study was conducted in an urban fragment of the Atlantic forest of north-eastern Brazil, where a group of common marmosets was followed for 488 h. For this purpose we used ad lib observations, performed twice a week, when the animals were entering and leaving the sleeping sites, identified potential predators of the common marmosets through interviews with local people, and identified the trees that provided them with fruit and/or exudate during the study. Five sleeping sites were identified, all invariably in the border of the native forest and the orchard. No predators of common marmosets were seen by local people or recorded during this study, and the sleeping trees were located invariably in the place where there was the highest concentration of feeding trees, regardless of its structure. There was no species-specific type and structure of sleeping places of common marmosets. The key variable defining the choice of a sleeping place was the availability and location of immediate sources of food.

Although many mouflon Ovis gmelini musimon populations have been introduced to continental southern and central Europe, little is known about their reproductive output. Based on post-mortem analysis of 344 harvested females, the variation in reproductive performance was investigated in three wild populations: two located in the French Alps and one in the south of France. Examination of tracts indicated a high pregnancy rate (> 80%) for females ${\ge}1.5$ years old. We found that a significant proportion of female lambs were pregnant in the Alps, c. 1 year earlier than generally reported for first reproduction in the mouflon, whereas female lambs did not conceive in the southern France population. Pregnant female lambs and yearling females always bore a single foetus. Among pregnant females ${\ge}1.5$ years old, twinning occurred less often in southern France (2.5%) than in the Alps (10.8% and 20.7%). The data required to relate body mass and reproductive output of ewes at individual level were lacking, but female lamb–ewe body mass ratio was used to test that the population with high reproductive output should have the highest ratio. As expected, the index was higher in the Alps than in southern France, suggesting between-population differences in the quality of the local environment. Differences in growth pattern and birth timing of lambs, or different tactics of maternal care could also have influenced the female lamb–ewe body mass ratio. Crossings between wild and domestic sheep during the recent history of these populations and differences in environmental conditions might have interplayed to shape age at first reproduction and twinning rates.

Carpal glands (CG) of 105 feral pigs Sus domesticus, caught in the tropical lowland rainforest in north-east Queensland, Australia, between 1999 and 2004, were investigated to examine their function in chemical communication between animals, and their histology. Female feral pigs show significantly larger CG on the right leg than on the left leg while there were no side-specific differences in males. CG on both legs were significantly larger in reproductive than in non-reproductive females, but they did not differ between pregnant and lactating females. The results suggest that CG are involved in the defensive behaviour of reproductive females but not in the identification of the mother by piglets. The area of the left CG was significantly bigger in males compared to females, but no significant difference could be shown for the CG on the right legs. CG of same-aged boars did not change significantly in size throughout the year while females showed smaller CG on the left leg in January and February suggesting that CG may be involved in intra-matriarchal group communication. Same sized and aged boars did not show any correlations between the size of the CG and the weight of their testes and the serum levels of testosterone. These results suggest that CG are not involved in advertising dominance in boars. The histological investigation of CG showed that they are active in feral pigs in the lowland rainforest, consist mainly of apocrine tissue and that their hairs may play a role in distributing secretion.

Distance sampling is a common and increasingly used method of estimating animal abundance in conservation and management programmes. The precision of a distance sampling survey can be estimated from the field data, but accuracy can only be evaluated by comparison with the true population size. For wild mammal populations, such opportunities are rare. The high-density badger Meles meles population at Woodchester Park in Gloucestershire, U.K. has been routinely and intensively live-trapped for the past 30 years. The estimates of abundance based on mark–recapture analyses of the live-trapping data provided a reliable baseline against which to evaluate distance sampling for a badger population. A distance sampling survey was carried out in autumn 2002 by using spotlighting along line transects. Owing to the impractical nature of spotlighting in closed habitats, such as woodland, surveys were conducted in open habitat only. A radio-tracking study provided behavioural data that were used to estimate the proportion of the population available for sampling in open habitats. This was incorporated as a multiplier into the DISTANCE analysis. The estimate of abundance from distance sampling was 76 badgers (% CV=42.2), using a multiplier based on radio-tracking data from autumn only. When the multiplier was derived from radio-tracking data from across the whole year, based on a larger sample of tracked badgers, the distance sampling abundance estimate was 77 badgers (% CV=25.9). These compared favourably to the mark–recapture estimate of 68 badgers (% C.V=7.4). The lack of precision in the distance sampling density estimate was largely owing to the inclusion of the multiplier. Distance sampling combined with data on habitat use can produce reliable estimates of badger abundance in areas of medium to high density, particularly in landscapes of primarily open habitat.

The availability of a population of mostly known-age African elephants Loxodonta africana from Amboseli National Park, Kenya, provided a unique opportunity to assess the use of dung bolus diameter for estimating age. A predictive equation for estimating dung bolus diameters from elephants of known age was derived and was found to follow the typical growth pattern exhibited by changes in shoulder height and foot length. The relationship between measurements of dung bolus and age was particularly strong when growth rates were high (age 0–25 years). The dung bolus growth curve from Amboseli elephants was similar to that derived from another wild population of African elephants, suggesting that dung bolus diameter can be used to assess age structure in areas where it is impossible to construct independent prediction curves of age and dung bolus.

Most mammals exhibit seasonal variation in the reproductive capacity of one or both sexes. While the female southern hairy-nosed wombat Lasiorhinus latifrons is a known seasonal breeder, the extent of seasonality in the male has not been documented. To examine this, gross body measurements including scrotal diameter and the dimensions of the accessory gland bulge were recorded and male reproductive tracts were examined between 1993 and 2000. Testes, epididymides and accessory glands from all males were dissected free of connective tissue and weighed. In addition, matched semen samples were collected over four time points in 2000 corresponding to the breeding season (September), immediately post-breeding season (November), during the non-breeding season (January), and immediately before the onset of the next breeding season (June) as determined from female reproductive status. Semen was collected by electro-ejaculation and analysed for volume, sperm number and motility characteristics. Ejaculate volume, total ejaculate sperm number, percentage motile sperm, and the sperm motility rating and index were all significantly elevated in September and significantly lower in November and January. This correlated with a significant increase in body weight, peri-cloacal gland width, and the weights of the prostate, Cowper's glands, urethral bulb and crus penis. The data confirm that male reproduction in the southern hairy-nosed wombat, like that of the female, is highly seasonal with a peak in reproductive capacity occurring in August–September and a reduction by November.

To monitor the biodiversity quality of rainforest fragments in the eastern region of the Ivory Coast, the density of at least one important invertebrate group needs to be determined. This has been possible for nocturnal hunting spiders from the family Ctenidae using a novel combination of sampling and census techniques. The reflective characteristics of the eyes of these spiders meant that individuals were readily located using headlamps. Collections were made along transects (total length 5.5 km), adopting a distance sampling protocol. Density was estimated using distance sampling models. More than 2600 ctenids (15 species) were captured and identified. Seven of these species were present in sufficient numbers for density assessment. The combined density of these seven most common species in degraded forest fragments (0.124 m−2) was only half that in less disturbed forest (0.240 m−2). Ctenidae are a promising candidate for future monitoring of invertebrate groups in the rainforest.