Upon its adoption in 1992, the Convention on Biological Diversity (CBD) was greeted with enthusiasm by the conservation community including bird conservationists. Nine years after its entry into force in 1993, the CBD still has to live up to the original high expectations if it is to halt, or even slow, the loss of biodiversity at the ecosystem, species and genetic level. Among the reasons for this shortcoming, the rather vague nature of obligations for the 182 governments who have ratified the Convention is probably the most important. National Biodiversity Strategies and Action Plans (NBSAPs) form the most significant instrument for the implementation of the CBD. The Global Environment Facility (GEF) is supporting the NBSAPs financially, as well as local, national and regional projects, many of which are relevant for bird conservation. Being the best-researched animal taxa, birds could serve relatively easily as indicators for the success of efforts under the Convention. Some cases, where ornithological data are being used within NBSAPs and related processes, are highlighted. From the early stages of planning, bird surveyors should take the potential use of their results by decision-makers into account. For ensuring that survey data are fed into NBSAPs and other similar policy documents, a strong collaboration of ornithologists and birdwatchers with non-governmental conservation organizations is recommended.

We developed distributional models for 90 threatened bird species in China, and used heuristic complementarity algorithms to prioritize areas for conservation. The pixel-based area selection prioritized 20 areas for protection, which covered all species analysed. Area selecting for endangered species based on the existing biosphere reserve system included only 37-62 species in eleven reserves, leaving 28-53 species unprotected. We employed algorithms for area selection based on species richness and rarity and obtained two views of a most efficient reserve network. We used the distributional hypotheses to identify additions to the reserve system that would improve its effectiveness substantially. The pixel-based area selection approach includes species much more efficiently on a per-area basis, and thus offers exciting perspectives for improved protection of the country's endangered avifauna.

In view of the continuing appropriation and conversion of natural land areas in North America for human uses, there is growing concern about the impacts of changing land use on terrestrial bird species. In order to promote conservation of critical remaining habitats for birds, Partners in Flight (PIF) initiated a project in 1997 in which bird conservation plans were prepared by members in each of 60 ecologically defined physiographical areas throughout the United States. Accurate, nationwide information on the location and extent of vegetative cover types, as well as lands under state and federal management, are critically important elements in the creation of effective bird conservation plans. The National Fish and Wildlife Foundation (NFWF) awarded a challenge grant to The Nature Conservancy's (TNC) Wings of the Americas Program to assist Partners in Flight in acquiring land cover data to serve as the foundation of the planning effort. Canon U.S.A., Inc. and the American Bird Conservancy also contributed support toward this goal. The Center for Advanced Spatial Technology at the University of Arkansas was contracted to produce the needed land cover maps and associated tabular products. Digital land cover databases created by the U.S. Forest Service, the U.S. Geological Survey, the Canada Centre for Remote Sensing, the University of California-Santa Barbara Department of Geography, and the U.S. Department of Transportation, Bureau of Transportation Statistics were used in this project. The final spatial products were produced during 1998–1999 and are described in this paper. This effort represents the first nationwide habitat mapping project in the United States aimed at supporting and enhancing conservation of terrestrial bird species.

Spoon-billed Sandpiper Eurynorhynchus pygmeus is classed as globally Vulnerable, based on the only available population estimate, made in 1977, of 2,000–2,800 pairs. Surveys for breeding Spoon-billed Sandpipers were carried out in summer 2000 on the Anadyr estuary coast, the Chukotka autonomous region, Russia. Although six new breeding sites were found, only 16–17 breeding males/pairs were recorded on the northern coast of the Anadyr estuary and five males/pairs on the southern coast and more southerly lagoons. These numbers were much lower than expected, and the species was not recorded in several apparently suitable areas. Four formerly known breeding sites held only one displaying male between them. At no site was there evidence of an increasing or stable local population. Together with other indicative data these suggest a sharp decline in the population of Spoon-billed Sandpiper. The previous population size estimate obviously requires updating, both because there were some incorrect assumptions in the calculations made for the first population estimate, and also due to a recent population decline. It is likely that the current population numbers under 1,000 breeding pairs. Given that the population of this species may be substantially lower than the only previous estimate, and the evidence for a possible rapid decline in its population described in this paper, it is recommended that its IUCN status is changed from Vulnerable to Endangered. No reasons for the apparent rapid rate of decline are evident within the breeding range. “Bottle-necks” should be looked for at migration sites or wintering grounds, but currently, monitoring of the population is only possible on the breeding grounds.

Populations of Oriental White-backed Vulture Gyps bengalensis and Long-billed Vulture G. indicus declined in India between the mid 1980s and late 1990s. Regional reports from India described declines of 95–100% across a wide area. This study was conducted to investigate the breeding success and pattern of mortality in two vulture colonies (Dholewala and Changa Manga) within Punjab Province, Pakistan between December 2000 and June 2001. Breeding success was found to be 62% in Dholewala and 59% in Changa Manga. A total of 668 sick and dead vultures were collected of which 591 were less than one month post mortem. No significant variation was found in the weekly mortality rate of adult and subadult vultures during the study period spanning winter through summer. A peak in mortality rate was observed during late April and early May that corresponded to mortality of newly fledged juveniles. Minimum annual mortality rate in the adult breeding population was calculated to be 11.4% and 18.6% in Dholewala and Changa Manga respectively. In a subsample of dead vultures (n = 185) visceral gout was found in 80% of adults, 63% of subadults, 19% of juveniles and 13% of nestlings. These mortality rates were consistent with a rapid population decline. Results imply that the mortality factor responsible for the decline in Gyps vultures described in India is also present in Pakistan and will potentially lead to a population decline of a comparable magnitude.

The Moroccan population of Great Bustard Otis tarda was surveyed in the four years 1998–2001. Highest numbers were counted in December 1998 (21 males, 62–65 females, four first-year males: total c. 90) and March 2001 (15 males, 60–66 females, four first-year males: total 75–81). Combining enquiry results from local people with these data, the population size for spring is estimated at 92–111 individuals, and 117–133 for winter. Reproduction is confirmed. Threats to the population include illegal hunting, habitat loss through agricultural intensification (irrigation, use of pesticides) and infrastructural changes (construction of a motorway and a powerplant), disturbance, and accidents at powerlines and antennas. Several conservation measures are suggested: cease hunting; maximize the attention given to display sites; encourage villagers to carry out “Great Bustard-friendly” land use and adapt animal stocks to the carrying capacity of their regions; clarify the situation of genetic exchange between Spanish and Moroccan populations; study movements between the Moroccan sites; monitor the size and trends of the most important subpopulations; and conduct education campaigns in schools and increase public awareness among adults.

The Endangered Rodrigues Warbler Acrocephalus rodericanus is endemic to the Indian Ocean island of Rodrigues, one of the world's most devastated tropical islands, its native forest having been completely destroyed since human colonization three centuries ago. It is now found in thickets and woodland dominated entirely by non-native trees and shrubs. As a step to implementing conservation initiatives, a population census and habitat study was undertaken in April-June 1999. Using a combination of tape playback of song and point counts, at least 103 Rodrigues Warblers spread through nine wooded localities were observed, and a minimum population of 150 birds estimated. The majority (78%) and highest densities (2.3/ha) were found in habitat dominated by one introduced invasive tree, the rose-apple Syzygium jambos. Warblers were also found in plantations dominated by mahogany Swietenia mahagoni, tecoma Tabebuia pallida and Norfolk Island pine Araucaria cunninghamii, but at much lower densities (0.5/ha). Warbler densities were highest in habitat with a dense structure of small branches. There was a strong positive relationship between one index of human disturbance (number of cut branches) and warbler densities. This low-intensity cutting may promote the growth of new shoots thus perpetuating the dense vegetation structure that Rodrigues Warbler favours. At two localities, one supporting an existing population of warblers and the other prone to extirpation (shown from previous surveys), vegetation structure and composition were similar between sites, hinting that in some currently unoccupied areas, habitat is probably suitable and other factors are operating to preclude colonization and establishment. Further ecological studies would be desirable, especially to investigate the effect of nest-predation by introduced mammals and other factors suppressing warbler population growth.

The highland grasslands of central Kenya hold a suite of restricted-range bird species, including several of global conservation concern. These grasslands occur almost entirely on private land with no formal protection, and have received little conservation attention. This paper describes land-use change on the Kinangop Plateau, Kenya, and considers the implications for grassland bird species, especially Sharpe's Longclaw Macronyx sharpei. The Kinangop Plateau is an area of montane grassland east of the Rift Valley, used primarily for dairy farming since human settlement in 1964. However, dairy farming (which retains grassland as pasture) has given way to other forms of land-use, especially cultivation of crops. In March 1996, a questionnaire was administered to 50 landowners to investigate community attitudes towards birds and their conser vation, the extent of present land-use changes and likely future land-use patterns. The mean acreage of landholding was decreasing, and more grassland was rapidly being converted into cultivated land, with 3.2% being ploughed up during the six months from November 1995 to May 1996. Farmers expressed a preference for crop farming (66%) over livestock farming (26%) or other land uses. Thirty-two per cent expressed a positive attitude, and 60% were indifferent, towards birds. Most (76%) were not aware of biodiversity values or conservation. Though 64% were aware of declining bird populations around them, 82% of this group did not realize that their activities might be contributing to this decline. Sixty-eight per cent of landowners planned to convert all or part of their grassland holdings. Although landowners' interests seemed generally incompatible with grassland conservation, a substantial proportion (44%) were prepared to consider opportunities to enhance the area's conservation value. An informal follow-up survey in March 1999 concluded that prospects for the dairy industry and for grassland habitats on Kinangop were not as bleak as the initial survey suggested, but confirmed the delicate conservation status of Sharpe's Longclaw and its habitat. We suggest a variety of actions that might help to secure the future of Sharpe's Longclaw while promoting sustainable agricultural development on Kinangop, and propose several topics where further research is needed. Species such as Sharpe's Longclaw will only survive if we can develop ways of managing their habitats in a manner that is consistent, rather than in conflict, with the needs of the agricultural community.

Between 1995 and 2000, we made a comparative study of ranging patterns of individuals of five sympatric species of antpitta Grallaria spp. that differ in body size and extent of geographical range. Grallaria milleri is restricted to a small area in the central Andes of Colombia, G. rufocinerea is endemic to the Central Cordillera, and G. ruficapilla, G. nuchalis and G. squamigera are widely distributed in the tropical Andes. Using mist-nets, we captured and banded 76 individuals of the five species during the five year study period. We also radio-tracked six individuals for periods of 30-60 days, to obtain estimates of territory sizes. Recapture patterns revealed that antpittas exhibited high site fidelity, remaining in the same patch for documented periods of up to 34 months. However, some individuals were captured only once, and were probably transient through the area. Territory sizes estimated were 0.5-5.4 ha for G. milleri (n = 7), 1.4 and 1.5 ha for G. rufocinerea (n = 2), 1.9 ha for G. ruficapilla (n = 1), 9.3 ha for G. nuchalis (n = 1) and 2 and 4.5 ha for G. squamigera (n = 2). We compared these data with territory estimates of other Neotropical Formicariids. Our results indicated that antpittas had relatively small territories, and had population densities higher than assumed previously for these secretive birds of the forest understorey.

Little is known about Mexican Duck Anas diazi biology and populations. We analyse long-term (1960–2000) trends of Mexican Duck numbers in Mexico and employ contemporary count data (1991–2000) from the U.S. Fish and Wildlife Service midwinter surveys to identify key sites for conservation using a complementarity approach. The overall Mexican Duck population showed a significant long-term increase of 2.5% per year, with large fluctuations throughout the study period. The Northern highlands population increased at an annual rate of 7.7%, while the Central highlands population showed no significant long-term trend. During the last decade, counts in both the Northern and Central highlands exhibited no significant change. At the site level, significant long-term increases occurred in four localities in the Northern highlands (Laguna Bab�cora +13.9% annually, Laguna Bustillos +25.9%, Laguna Mexicanos +20.4% and Laguna Santiaguillo +16.9%) and in three localities in the Central highlands (Languillo +15.3% annually, Presa Solís +8.9%, Zacapu +13.4%). Two sites in the Central highlands showed significant declines, in the long term (Lago de Chapala, -5.2% per year) and during the last decade (Lerma, -11.8% per year). The Northern highlands held 16% and the Central highlands 84% of the Mexican Duck population in the period 1960–2000; during the last decade, these figures were 31% and 69%, respectively. A set of priority sites for conservation of the Mexican Duck was identified, consisting of 15 sites holding more than 70% of the midwinter Mexican Duck counts in Mexico. Ten sites from the priority set also qualify for designation as wetlands of international importance under the Ramsar Convention on Wetlands, by holding ≥ 1% of the estimated population. Four of the priority sites are in the Northern highlands and 11 in the Central highlands, of which eight are distributed along the Rio Lerma drainage. The most urgent actions that need to be undertaken are to estimate the current minimum population size in Mexico; to establish a programme for monitoring populations in the priority sites, especially those located within the highly degraded Rio Lerma drainage; and to determine the most feasible management actions for the species, concentrating efforts around the priority sites.

The island of Santa Catarina lies in the Atlantic Forest Domain of southern Brazil, one of the most threatened ecosys tems on the planet. It holds a wide variety of habitats, ranging from sand dune ‘restingas’ to mature rain forest. We report, for the first time, a survey of the bird species and their conservation status for the different habitats present on the island. We recorded 269 bird species of 54 families (25 species were seabirds). More than 50% (140) of the total number of species were residents or probable residents, 20% (58) were visitors and a relatively high number of species (61, 23%) had indeterminate residential status. Nearly 15% (35 species) of the 244 landbird species observed were endemic to the Atlantic Forest. Habitat specificity was recorded for 40% (97 species). We recorded one Endangered (Solitary Tinamou Tinamus solitarius), one Vulnerable (White-necked Hawk Leucopternis lacernulata), and three Near-threatened species (Azure Jay Cyanocorax caeruleus, Azure-shouldered Tanager Thraupis cyanoptera, Black-backed Tanager Tangara peruviana). We also recorded six locally threatened species that we predict will become extinct in the near future. The future for the birds of the island is uncertain, especially for habitat specialists, due to habitat loss for resort developments, and a lack of concern by the government and local people.

Siberian Crane Grus leucogeranus occurs only in Asia, and is Critically Endangered. The western population of the species has been almost extirpated, wintering at just two known sites, in Iran and India. To help conserve species that migrate long distances it is essential to have a comprehensive conservation plan that includes identification of migration routes and key resting areas. One Siberian Crane was satellite-tracked from the south Caspian Sea to its breeding grounds in Russia during the spring of 1996. The crane began migration on 6 March, and completed its migration on 1 May. This destination was formerly unknown as a breeding area for the species. During migration, the crane rested primarily at the eastern end of the Volga River delta. This suggests that the delta may be an important resting site for Siberian Crane.

White-bellied Shortwing is a globally threatened species found in the Western Ghats of India. There have been few records of this species over the past 119 years and its population is thought to have fallen due to habitat loss and fragmentation. The present study was conducted to assess the habitat preferences and distribution of this species in the Western Ghats of Kerala and Tamilnadu states as a prelude to a long-term study on the biology and causes of decline of this species. Several areas across varying vegetation type and altitude were surveyed after an intensive reconnaissance. There were 214 sightings from evergreen and shola forests during the study. They were found to prefer shola forests to other vegetation types. The species was also found in pine, wattle and tea plantations near shola forests. Although it was found to prefer an altitude above 1,500 m, it was also found in shola forests between 1,000 and 1,500 m elevations. White-bellied Shortwing was found to occupy an elevation range from 1,000 m to 2,200 m. Encounter rates in shola forests fragmented by human interference were four times higher than those in natural shola forests. Most encounters of shortwings were near streams that were wet (running or damp). Emphasis is laid on the need for long-term studies to unravel the basic biology of the species in order to aid active conservation efforts.

A high proportion of island birds are threatened with extinction as a result of their vulnerability to introduced predators, habitat destruction, and fragmentation/isolation effects. In order to conserve island species effectively, it is necessary to disentangle these effects on distribution and abundance. We attempt to do this for the nine native passerines in the Falkland (Malvinas) Islands, using a database of presence/absence on 59 offshore islands in the archipelago, linked to data for each island on mammal presence, habitat modification, and isolation. Falklands native passerines are of considerable conservation importance, comprising one endemic globally threatened species, several endemic subspecies, and several restricted range species. Presence of rats on islands was by far the most important predictor of passerine presence, overriding the effect of habitat modifications. The globally threatened endemic Cobb's Wren Troglodytes cobbi was absent from all islands containing rats. Some species were more likely, and others less likely to occur on islands where tussac Poa flabellata grassland had been destroyed by grazing. The former species were primarily those adapted to dwarf-heath vegetation, and/or that thrive around human settlements. Island size and isolation were important predictors of occurrence for several bird species. The analyses show that, if vegetation restoration in the Falklands is to meet conservation aims, then it should be accompanied by introduced mammal control. Secondly, they indicate that biogeographical effects on bird distribution among islands in the Falklands are important, and need to be considered when assessing the conservation status of species, and when considering conservation action.

The island of Sangihe, Indonesia, was visited in 1998–99 with theaim of producing population estimates of the island's endemic and otherrestricted-range bird species, some of which were poorly known and thoughtto be at risk of extinction due to habitat change. The study focused on thelast remaining patch of primary forest, on MountSahendaruman in southern Sangihe, home to three critically endangeredendemic species: Cerulean Paradise-flycatcher Eutrichomyiasrowleyi, Sangihe Shrike-thrush Colluricincla sanghirensis,and Sangihe White-eye Zosterops nehrkorni. Population densitieswere estimated from primary forest and secondary habitats across Sangihe toassess species dependency on primary habitat. Twenty of 24restricted-range, globally threatened or endemic taxa were recorded anddensity estimates were calculated for 15 of these. The endemic Red-and-blueLory Eos histrio is extinct on Sangihe as a result of lowlandforest loss. Six species (Nicobar Pigeon Caloenas nicobarica,Sulawesi Dwarf Kingfisher Ceyx fallax, Golden Bulbul Ixosaffinis, E. rowleyi, C. sanghirensis, and Z. nehrkorni) wereeither not recorded or have verysmall populations and are critically endangered on Sangihe; two species ofTanygnathus parrot are also facing local extinction on the island.The major threat to all these species is the destruction of primary forest;larger species are also threatened by hunting. The remaining endemic andrestricted-range species (Blue-tailed Imperial Pigeon Duculaconcinna, Sangihe Hanging-parrot Loriculus catamene, SangiheScops-owl Otus collari, Lilac-cheeked Kingfisher Citturacyanotis and Elegant Sunbird Aethopyga duyvenbodei) were morewidespread and occurred in secondary habitats. I recommend that L.catamene and A. duyvenbodei, currently treated as globallyendangered, be reclassified as near-threatened and vulnerable respectively because of their large populationsand tolerance of disturbed habitats. Species with wide global ranges thatare represented by endemic subspecies have the greatest tolerance fordisturbed habitat. The widespread deforestation of Sangihe has had seriousconsequences for many bird species and today the island supports the mostthreatened assemblage of single-island endemic species in Indonesia.Species-specific research to determine the status and ecology of E.rowleyi, C. sanghirensis and Z. nehrkorni, and monitoring ofthe Sahendaruman forest are desperately needed as a basis for futureconservation efforts.

We report new observations of the globally threatened Minas Gerais Tyrannulet Phylloscartes roquettei from two areas in the São Francisco Valley, Minas Gerais, Brazil, between July 1993 and February 2002. Four pairs, one fledged young and two lone individuals were observed in the course of our fieldwork. It was previously known only from a female taken in the mid-1920s and sight records and tape-recordings in 1977. Our records extend the species' known range 250 km south and west of the type-locality region. Details of the first-known male specimen are presented, along with novel data concerning its vocalizations and behaviour. We draw attention to the possible relationship of the species to a group of four other Phylloscartes tyrannulets with similarly patterned faces and overall plumage, which exhibit a similar circum-Amazonian distribution pattern to three Phyllomyias tyrannulets. We also take the opportunity to draw more attention to the imperilled conservation status of the dry forests upon which P. roquettei and a host of other threatened and Near-threatened avian taxa depend.

Grey-headed Quail-dove Geotrygon caniceps has traditionally been considered a polytypic species endemic to Cuba and the Dominican Republic and treated as globally threatened within the most recent Red Data Book (BirdLife International 2000). Chapman (1917) described Geotrygon leucometopius of Hispaniola as specifically distinct from G. caniceps of Cuba based on 10 specimens, taken by Rollo Beck in the Dominican Republic. Subsequently, Bond (1936, 1956) merged leucometopius within caniceps, an arrangement that has persisted, unchallenged in the technical literature, until the present. Through examination of 76 specimens, extensive field experience of Cuban birds, and less exhaustive fieldwork in the Dominican Republic, we re-evaluate the taxonomic status of the Hispaniolan population, identifying consistent differences in coloration, tail length and characteristics of the second to fifth primaries between it and the Cuban population. Based on these differences, we suggest that caniceps (endemic to Cuba) and leucometopius (restricted to the Dominican Republic) be henceforth resurrected to species status. We were unable to undertake a complete analysis of the vocalizations of the two forms, due to the lack of definite recordings from Hispaniola, but present sonograms and notes concerning Cuban birds. Further work, including molecular analyses, would be clearly desirable to test our hypothesis. Both forms are undoubtedly declining due to habitat destruction and hunting, and both certainly qualify as Vulnerable under current IUCN criteria. Indeed, leucometopius may even qualify as Endangered under the range criterion. Its status requires particularly careful monitoring, while new information, published since the BirdLife International (2000) review of globally threatened birds, suggests that the range even of nominate caniceps is considerably smaller than previously considered.

Cape Vulture Gyps coprotheres is endemic to southern Africa and is globally threatened. Colonies in Botswana comprise part of one of the two core breeding areas in the species's range, and very little has previously been published about them. Ground censusing of 11 Cape Vulture sites in Botswana was undertaken from 1992 to 1999, continuing a monitoring programme initiated by the authors in 1984. Survey methods and census procedures are documented. The potential Cape Vulture breeding population in Botswana is estimated to be about 600 pairs, comprising at least 100 pairs at Mannyelanong in the south-east and about 500 pairs in eastern Botswana. This represents an increase from previous estimates, and reflects improved census techniques rather than an increase in population size. The mean turnover rate of nest site usage from one year to the next averaged 21% at Mannyelanong, and was about 26% at Manong Yeng in eastern Botswana. Annual productivity of nest sites increased with the number of years the site had been occupied. Over the eight years of study eggs were laid in at least 436 of 477 nests (91.4%) at Mannyelanong; chicks survived to mid season (60–80 days old) in 327 nests (75% of eggs laid), and fledged (best estimate) in 248 nests (56.9% of eggs laid and 52% of pairs attempting to breed). In eastern Botswana eggs were laid in at least 1,825 of 2,101 nests (86.9%); chicks survived to mid season in 1,272 nests (69.7% of eggs laid). Two seasons have been excluded for eastern Botswana (1994 and 1995) due to incomplete data, and breeding success can be estimated only from 1997 to 1999: of 990 eggs laid out of 1,108 nests, chicks fledged in 384 nests (38.8% of eggs laid and 34.6% of pairs attempting to breed). The eastern Botswana breeding population remains in a state of flux following the collapse of a primary colony that was the country's Cape Vulture stronghold. Conservation concerns and the vulnerability of Cape Vulture sites are discussed.

Saffron-cowled Blackbird Xanthopsar flavus is a globally threatened icterid of open areas of south-eastern South America. Several sources suggest that the species is declining throughout its range, mainly due to habitat destruction. Between December 1997 and December 1999 we studied a breeding population of X. flavus at the Banhado do Vinte-e-Cinco and Banhado da Mulata area, Rio Grande municipality, Rio Grande do Sul state, Brazil. This population, previously unreported for the state's coastal plain, numbers at least 60 individuals and seems confined to an area of c. 35 km2. The area is characterized by Cyperacaea-dominated peat marshes in comparatively higher terrain, covered with sandy, rolling fields. The species was recorded in all months (not necessarily of the same year) except March, and a marked irregularity in flock size was detected. During the non-breeding period, flock size averaged 12.7 individuals (S.D. = 11.7; range = 2-37; n = 11). During the 1997-1998 breeding season (December to January), mean flock size was 25.3 individuals (S.D. = 16.2; range = 16-60; n = 7), while in the 1998-1999 and 1999-2000 breeding seasons, 33 and 4 individuals were recorded respectively. No pronounced discrepancy was found in the size of the largest flocks recorded per season (from 20 to 37 individuals), suggesting that the population is sedentary and that it undertakes only local movements. A nesting colony occupying the same part of the marsh in two successive breeding seasons was composed of six breeding pairs in 1997-1998, with an estimated nesting success of 31.8-36.3% at fledgling time. The species is highly gregarious and sociable, foraging flocks frequently following individuals of Black-and-White Monjita Heteroxolmis dominicana and, to a lesser extent, Yellow-rumped Marshbird Pseudoleistes guirahuro and Brown-and-yellow Marshbird P. virescens. Although habitat loss has been indicated as the main factor for the species' decline, we suggest that habitat perturbation and modification may affect negatively recruitment rates in the species and, theoretically, its overall population size. Furthermore, the conservation of Rio Grande's peat marshes should be regarded a regional priority, in order to preserve threatened birds and a very unusual wetland type.

The endemic Seychelles Scops-owl Otus insularis is a Critically Endangered restricted range species currently only recorded in the upland forest of Mahé, the largest (152 km2) and highest (903 m) island in the granitic Seychelles. We studied the response of colour-ringed individuals to playback of conspecific calls, to determine factors affecting playback reaction, to monitor behaviour and to assess the reliability of current monitoring techniques. Playback of the territorial waugh call was conducted at fixed points (c. 200 m apart) every month along a total transect length of 8.4 km between April 1999 and May 2001. On the basis of responses and movements of marked individuals we identified 12 territories and noted their boundaries. Mean territory length was 2.92 points (approximately 400 m), equivalent to a conservative estimated territory size of c. 12–16 ha. Males were more frequently detected than females/pairs. The probability of detection (males vs. females/pairs) was 0.87 vs. 0.50 on territory; 0.65 vs. 0.26 at fixed points within territories; and 0.58 vs. 0.22 at fixed points along the transect. There was some significant seasonal variation in detection rates at fixed points within territories for both males and pairs, and in general there was a low probability of detecting individuals around June-August and a high probability of detection around April and, to a lesser degree, October. Response times and owl-recorder distances were not influenced by moon phase or time of playback. This study provides the first baseline data on individual responses to the systematic repeated use of playback in Seychelles Scops-owl. We discuss the implications for current monitoring and its application to other similar Otus species.