The Brazilian Sea Turtle Conservation Programme (TAMAR) was established in 1980 to protect sea turtles in Brazil and to re-establish their natural life cycle, which had been disrupted by humans. The programme is based on grassroots involvement by the coastal communities, and provides jobs and generates alternative sources of income. Within this framework, the mini-guides programme was created. It consists of a summer training course in which local children between 8 and 13 years old learn basic information about marine ecosystems and sea turtles, strategies to promote their conservation, and skills to interact effectively with tourists. After the course, children are given onsite experience by working as interns at the TAMAR visitor centre at Praia do Forte, on schedules designed to avoid conflicts with their studies at school. The children work closely with tourists, informing them about sea turtles and TAMAR's work; they also help the biologists in field activities. In this way, they enhance their own awareness, and by extension their families' and the community's concern about environmental conservation. Their involvement provides them not only with valuable skills, but also an extra source of income, thereby providing economic benefits to their families. The success of this programme is indicated by the excellent approval ratings given by the tourists visiting the area.

A survey of diurnal primate populations was carried out in Odzala National Park, Republic of the Congo, over 3 months in 1994 and 2 months in 1995. The park contains a high diversity and species-specific abundance of primates, and has the highest number of diurnal primates (10 species) in the forest block of Central Africa. Eight species of monkey: greater whitenosed monkey Cercopithecus nictitans, moustached monkey Cercopithecus cephus, crowned guenon Cercopithecus pogonias, De Brazza's monkey Cercopithecus neglectus, talapoin Miopithecus talapoin, white-cheeked mangabey Cercocebus albigena, agile mangabey Cercocebus galeritus and guereza Colobus guereza, as well as gorilla Gorilla g. gorilla and chimpanzee Pan t. troglodytes were sighted in the survey. Monkey species richness was highest in dense inundated forest and thicket, with all eight species occurring in these habitats, whereas only four species were found in terra firma forest (consisting of the park's two main habitats, open-canopy Marantaceae forest and closed-canopy primary forest). Three of the four species (C. nictitans, C. cephus and C. albigena) present in terra firma forest were most abundant in closed-canopy primary forest (1.4,1.0 and 0.6 groups per km, respectively) while the fourth (C. pogonias) was most abundant in open-canopy Marantaceae forest. Gorilla nests were most abundant in open-canopy Marantaceae forest (12.1 nests per km), while chimpanzee nests were mostly found in closed-canopy primary forest and Marantaceae forest (14 and 12 nests per km, respectively). Odzala has the highest recorded densities of western lowland gorilla (mean = 5.4 individuals per sq km) and chimpanzee (mean = 2.2 individuals per sq km) in Central Africa. The high densities of gorillas and chimpanzees may be a result of the high productivity of the forest and low poaching pressure. Conservation measures to ensure the maintenance of conditions in the area, such as educational programmes, are suggested.

The Juan Fernandez Islands constitute two distantly separated (182 km) major islands, of which Isla Robinson Crusoe is the closest (667 km) to the Chilean mainland, and a number of smaller surrounding islets. The endemic Juan Fernandez firecrown Sephanoides fernandensis once inhabited both major islands and numbered many thousands. Today it exists only on Isla Robinson Crusoe and numbers just a few hundred. Another hummingbird, the green-backed firecrown S. sephaniodes, inhabits this island but also occurs on the mainland. The islands are in ecological meltdown as a result of the removal of much of the native (largely endemic) biota. This degradation has been exacerbated by introduced herbivorous and carnivorous mammals, which are extirpating all that remains. In a previous investigation the authors assessed the evolutionary relationships and genetic variation in the Juan Fernandez firecrown population. They showed that the species: (i) is a relative of the high Andean hummingbirds; (ii) is a close sister species of the green-backed firecrown; (iii) dispersed to the islands approximately 0.5–1 million years ago, unlike the green-backed firecrown, which is probably a very recent invader; (iv) maintains some genetic variation, although significantly lower than that found in the green-backed firecrown island population. Because of their relatively recent divergence the authors aimed to find out if the two species compete for resources. In addition, they assessed the conservation options for the Juan Fernandez firecrown and provided recommendations for its management. They concluded that the survival of the Juan Fernandez firecrown is inextricably linked to the holistic restoration of ecosystem/ community function of the island. The initial step must involve the removal of introduced mammals, followed by replanting of native flora. Although the authors do not support the use of captive propagation as a general conservation tool, they believe that, because the Juan Fernandez firecrown is composed of only one small population that is under extreme threat of extinction, a captive population should be established.

Malagasy chameleons are threatened by the relentless pace of habitat destruction. Their great popularity amongst herpetoculturalists has also resulted in collection for international animal markets. Most previous fieldwork has focused on the compilation of much needed species inventories, but little attention has been given to estimating chameleon population densities. This lack of information prevents a reliable assessment of the effects ofhabitat loss and direct exploitation on wild populations. A simple and repeatable methodology for the monitoring of chameleon populations is therefore urgently required. The present study developed survey techniques based on distance sampling within an area of relatively undisturbed rain forest at Ranomafana National Park. In total 394 individuals from six species were recorded and population densities (± SE) of 27.2 ± 5.4/ha Brookesia spp. and 21.4±4.0/ha Calumma spp. were estimated using the computer program DISTANCE. The authors propose that monitoring of chameleons in areas subject to different threats should begin immediately using the methods outlined in this paper.

In the debt-ridden, high-population-growth, resource-mining states of the Congo Basin, conservation of biodiversity is seldom the primary concern of national policy makers or of local resource users. Moreover, the recurring costs of managing protected areas and the opportunity costs of forgoing logging and farming to maintain protected areas are a substantial net drain on national and local economies. Consequently, it is becoming increasingly important that protected areas generate, from user fees or donor contributions, sufficient funds to offset the costs of maintaining them. Government and donor investment currently meet less than 30 per cent of the estimated recurring costs required to manage the protected-area network within central African countries effectively, and cover none of the growing opportunity costs. Nature tourism, the fastest growing sector of the $US3 trillion (3 million million) a year global tourism industry, may offer a source of revenue to help fill this gap in funds. Congo Basin national parks and reserves harbour many charismatic animals (okapi, lowland gorilla, mandrills, bongo, forest elephant) that are likely to attract tourists, and as a result many protected- area managers are sinking capital into the development of tourist infrastructure. This paper reviews the evidence for ecotourism's capacity to generate revenue for protected-area management and appraises the financial viability of nature tourism in the Congo Basin.

The threat that the bushmeat trade presents to primates and other taxa was assessed from the literature, including data from markets, village hunting studies and logging concessions in Central and West Africa. In many cases the numbers of both common and protected species of primate being killed throughout the region are thought to be unsustainable. This is also the case for other taxa involved in the bushmeat trade, which crosses geographic, cultural and economic boundaries. A suite of measures must be considered to mitigate the effects of this trade, and these measures will have to recognize the local, regional and national socio-economic importance of the trade if they are to result in long-term conservation success.

The African wild dog Lycaon pictus has declined dramatically over the past 30 years. Formerly distributed throughout sub-Saharan Africa, today c. 5000 wild dogs remain in total, mostly in southern and eastern Africa. Wild dogs' decline reflects the expansion of human populations and the associated fragmentation of habitat available to wildlife. Because wild dogs live at very low densities, even ‘fragments’ covering thousands of square kilometres may not support viable populations. Furthermore, packs often range beyond the borders of reserves, so even nominally protected populations are often subject to persecution, road accidents, snaring and disease contracted from domestic dogs. Such edge effects mean that reserves smaller than c. 10,000 sq km will provide only incomplete protection. The highest priority for wild dog conservation, therefore, is to maintain and promote the contiguity of areas available to wildlife. Establishing cross-border parks and buffer zones, and encouraging game ranching on reserve borders, will all be beneficial. In smaller areas, protecting wild dogs requires that edge effects be mitigated by: (i) working with local farmers to limit persecution; (ii) controlling snaring; (iii) routing roads carrying high-speed traffic away from wildlife areas; and (iv) minimizing contact between wildlife and domestic dogs. Most of these measures will also benefit other wildlife.

In 1989 an experiment was started at the Bukhara Breeding Centre, Kyzylkum Desert, Uzbekistan, to discover whether zoo-bred Przewalski's horses Equus przewalskiicould adapt to semi-wild desert conditions. One stallion and four mares of different ages were released into a 5126-ha fenced area and monitored over a period of 17 years. This paper presents the information gained from the studies concerning home ranges, interactions with other ungulate species, adaptation to new food resources and comparative changes in daily activity. The results of breeding and changes in the structure of the group are also presented. The horses appeared to adapt well to the new conditions and there appeared to be no negative effects of interspecific interaction with introduced kulans Equus hemionus kulan in the release area.

Atelognathus nitoi is a microendemic leptodactylid frog restricted to the area surrounding the pond Laguna Verde in the Nahuel Huapi National Reserve, north-west Patagonia, Argentina. Its habitat is potentially threatened by a number of anthropogenic and natural factors. The aim of this study was to obtain basic knowledge of the biology and habitat of the species, which would be useful for selecting management measures to ensure its conservation. The area was surveyed during all four seasons in 1996 and 1997 and Laguna Verde was sampled intensively in winter and in summer. Atelognathus nitoi is distributed patchily. The preferred microhabitats of adults and juveniles are the most humid areas of the forest. Reproduction and larval development occurs only in lentic environments. Laguna Verde is the only known permanent waterbody suitable for reproduction every year. During rainy years reproduction can also take place in temporary ponds. Action towards the conservation of A. nitoi should be aimed at preserving both the terrestrial and aquatic habitats. Laguna Verde appears to be a key site for the A. nitoi source population and plays an essential part in the conservation of this species. In terrestrial habitats it is vital to preserve the heterogeneity of the lower strata in the forest. The knowledge gained through this study has allowed the authors to suggest protective measures to mitigate or eliminate the impact of certain stressors on the ecology of the species.

African wild dogs Lycaon pictus have been extirpated across most of West and central Africa, and greatly depleted in eastern and southern Africa. Given an urgent need for population recovery, especially in West and central Africa, this paper discusses the possibilities for using reintroduction to re-establish wild dog populations. Reintroduction is probably now technically possible, as long as release groups include wildcaught animals; several past attempts failed because captive-reared animals lacked skills needed to survive in the wild. However, reintroduction has only a limited role to play in wild dog conservation. Ideally, it should involve animals of the appropriate local genotype. Limited genetic data indicate that wild dogs from West and central Africa may be distinct from those in eastern and southern Africa. Because there are no wild dogs with West or central African genotypes in captivity, and no wild populations in the region large enough to be harvested for translocation, future reintroductions might have to use animals with non-native genotypes. In addition, there appear to be no suitable sites for wild dog reintroduction in West or central Africa, and few in eastern and southern Africa. Releases currently planned in the Republic of South Africa will be locally valuable, but will not establish a population likely to remain viable without intensive management in perpetuity. For these reasons, protecting remaining wild dog populations currently represents a better investment than any attempt at reintroduction.

In sub-Saharan Africa conservation of biodiversity is increasingly predicated on finding ways to ensure that the economic value of maintaining a landscape in its ‘natural’ state meets or exceeds the expected returns from converting the area to an alternative land use, such as agriculture. ‘Wildlands’ in Africa must generate, directly or from donor contributions, funds sufficient to cover both the operating costs of conservation, and the opportunity costs of forgoing other forms of resource use. Government and donor investments currently meet less than 30 per cent of the estimated recurring costs required to manage the protected-area network within central African countries effectively, and cover none of the growing opportunity costs incurred to maintain protected areas. Unfortunately, few additional sources of funding are available.

Wild Bactrian camels Camelus bactrianus ferus are endangered. Surveys over the past several decades suggest a marked decline in camel numbers and reproductive success. However, most surveys were made using methods that precluded rigorous population estimation. The need for more accurate surveys resulted in an aerial survey of known and suspected camel habitat in Mongolia during March 1997. We estimated density, group density and population size of large mammals in south-western Mongolia using the interactive computer program DISTANCE. We recorded sufficient data for population modelling of wild Bactrian camels, goitred gazelles Gazella subgutturosa, Asian wild asses Equus hemionus and argali sheep Ovis ammon. We observed 277 camels in 27 groups (mean group size = 10.26 ± 2.38 SE camels/group). Modelling yielded a population estimate of 1985 ± 802 SE camels in the survey area. Population modelling for other ungulates yielded estimates of 6046 ± 1398 SE goitred gazelles, 1674 ± 506 SE Asian wild asses and 909 ± 303 SE argalis. Discrepancies between population estimates of ungulates in our survey and previous surveys are discussed with regard to methods used and robustness of results obtained. We also discuss conservation implications for wild Bactrian camels and other Mongolian ungulates.

In terms of the persistence of biodiversity, the siting of conservation areas has traditionally been ad hoc. In the Cape Floristic Region, a hot-spot of plant biodiversity and endemism, past conservation interventions have led to the mountains being over-represented in the reserve network, while the lowlands have remained very poorly conserved. Ongoing threats to the lowlands such as the rampant spread of invasive alien plants, and land transformation for agriculture and resort development, continue to undermine biodiversity in these regions. A new conservation intervention, the Agulhas National Park, is in the process of being implemented on the coastal lowlands at Africa's southernmost tip. A flexible, reserve-selection design tool is being used to guide this process. The practical challenges in implementing a new protected area in a fragmented landscape, which has a high biodiversity and vulnerability, are examined. The role of different institutions, in partic-in particular state-private partnerships, and current investigations into conservation agencies' policies, legislation and funding mechanisms are dealt with. It is imperative that future conservation planning considers the threats to biodiversity first and foremost. Institutions such as South African National Parks and the Cape Nature Conservation Board must act strategically to avoid changes in land use that will compromise the biodiversity goals of retention and persistence. Conservation efforts will only succeed if institutional and socio-economic considerations are integrated with conservation plans aimed at ensuring the long-term persistence of biodiversity.

The Chinese alligator Alligator sinensis is one of the world's most endangered reptiles. At one time widespread throughout much of the lower Yangzi River basin, the remaining wild individuals are now restricted to a small area in southern Anhui Province and perhaps in adjacent Zhejiang Province. Population estimates conducted in the 1980s suggested that only 500–735 wild individuals remained at that time. Current figures suggest that the wild population is c. 400 individuals and continues to decline. The principal factor contributing to historic population decline has been habitat loss, but deliberate killing of alligators and the heavy use of pesticides have also had significant negative effects. The current conservation programme in Anhui Province is based on captive breeding and the establishment of a reserve for small groups of wild alligators. However, the inferred recent decline in the size of the wild population suggests that the reserve design is inadequate for the long-term survival of alligators. Programmes to survey the status of the remaining wild populations and evaluate the feasibility of establishing new wild populations by reintroducing captive-bred animals are currently being developed.

The endemic bearded sakis Chiropotes satanas Satanas and Chiropotes satanas utahicki of south-eastern Amazonia are among the most threatened of this region's primates because of a combination of deforestation and hunting, and their apparent intolerance of habitat disturbance. Recent surveys at eight sites confirm that sakis are locally extinct in some areas where intense habitat fragmentation is exacerbated by hunting pressure, but also show that, in the absence of hunting, they can be relatively abundant in isolated forest fragments. Density was unexpectedly low in one protected area, however, which implies that caution is necessary for the planning of long-term conservation strategies. Well-protected forest fragments of reasonable size ( > 5000 ha) appear to have good potential for the protection of bearded saki populations. While many of the region's major landowners may thus make a significant contribution to the management of saki populations, land conflicts are a potentially serious problem for the long-term conservation of not just these primates, but the region's fauna and flora as a whole.

The conservation status of the sun-tailed guenon Cercopithecus solatus, which is endemic to central Gabon, is assessed using the most recent information available. The known range of the species covers 11,000–12,000 sq km, an area with a sparse human population but where the logging activity is widespread. There is no evidence of a recent reduction of the population size of this monkey, mainly because trees are extracted at a low rate (1–2 trees/ha), which does not cause a significant change in the forest composition and structure, and because hunting pressure is currently low. Moreover, the species is suspected to be able to thrive in degraded secondary forests. Consequently, the sun-tailed guenon does not appear to be under immediate risk of extinction. The conservation actions undertaken to date are described and the threats that may affect the monkey in the near future are discussed. Commercial bush-meat hunting is considered to be the most serious threat for the species. Recommendations for the conservation of wild populations and habitat are proposed, and emphasis is placed on co-operation with the forestry companies that are the main economic actors in the species's range.

In May 1996 there was an acute and dramatic mortality incident in the last remaining wild population of northern bald ibis Geronticus eremita. This species is Critically Endangered, comprising only about 250 wild individuals, which occur on the Atlantic coast of southern Morocco. Over a period of 10 days a total of 38 adult birds (aged 1 year or more) died or disappeared. Deaths, probably secondary to the loss of one or both parent birds, also occurred subsequently in six estlings and one recent fledgling. The incident appeared to involve no other species. This paper describes the pattern of the incident, and pathological, microbiological and toxicological investigations and findings. Several features point to a toxic aetiology but the cause of the incident has not been established.