The Blue Swallow Hirundo atrocaerulea is a globally threatened species endemic to sub-Saharan Africa. The total population breeding south of the Limpopo River (i.e. in South Africa and Swaziland) probably numbers around 100 pairs. A significant proportion of these birds breed in Swaziland, but to date little has been published on this population. Suitable breeding habitat has been reduced significantly in the country in recent decades, due mostly to afforestation with exotic timber plantations, with urbanization playing a lesser role. In Swaziland, breeding sites are restricted to grasslands above 1,200 m, but mostly above 1,300 m. Timing of egg laying is bimodal indicating double brooding. The number of nests initiated in a season is weakly correlated with rainfall in preceding months. Nests were built predominantly in disused antbear Orycteropus afer burrows, with smaller numbers in natural sinkholes. Mean clutch size was 2.80, and 61% of eggs laid resulted in fledged offspring. Mean productivity was 1.30 fledglings per pair per nest attempt. These figures suggest that breeding success is not currently being reduced in Swaziland. The minimum total population currently thought to be breeding in Swaziland is 10 pairs, but this is based on intensive studies of only part of the suitable range. It is recommended that a complete survey be conducted covering the entire range of the species in Swaziland.

The endemic brush-turkey of Waigeo Island in eastern Indonesia, Aepypodius bruijnii, is revealed to nest only on the island's previously ornithologically unexplored highest summits, along an ecological gradient above 620 m elevation where a structurally distinctive, windsheared and possibly locally edaphically controlled, stunted cloud-forest thrives on infertile substrates. The species is confirmed to breed at two locations only, and the known population totals 47 mound-owning males or 84 mature individuals. Its global population is estimated at 349 mound-owning males or 977 mature individuals, primarily confined to 60 km2 of cloudforest habitat spread over six locations, with 98% of the population restricted to just three locations in the eastern part of the island. Details are provided on the bird's altitudinal distribution, habitat preferences, spatio-temporal use of incubation sites, population densities, locations and incidental lowland records. Threats that impinge on the species are discussed in depth, and it is concluded that its current precautionary treatment as Vulnerable warrants upgrading to Endangered in accordance with the revised IUCN Red List categories and criteria. Habitat destruction as a consequence of wild fires and a recent logging epidemic is identified as the major factor threatening this unique brush-turkey's long-term survival.

Karamoja Apalis Apalis karamojae is a globally Vulnerable warbler, restricted to north-east Uganda, north-central Tanzania and southern Kenya. Here, we describe its status and habitat in the Wembere Steppe, Sukumaland, Tanzania, an unprotected area of seasonally inundated grassland bordered partly by Acacia scrub. Although its habitat in the steppe was thought likely to span c. 175 km north–south during the 1960s to 1970s, we found the apalis to be patchily distributed over a distance of only 102 km north–south. Distance sampling yielded a density estimate of c. 7–16 birds km−2; equivalent to c. 3–7 pairs km−2, after juveniles (18% of the population) were excluded, and assuming that all adults were paired. The species occupied a very narrow foraging niche, feeding almost exclusively in Whistling Thorn Acacia drepanolobium, which accounted for 66% of woody stems in the study area, but for 95% of stems in the immediate vicinity of apalis sightings. Half of all individuals were encountered in the tallest, densest stands of A. drepanolobium, which covered less than 6% of the study area. Signs of scrub clearance, pruning and browsing were widespread in the steppe, and may have intensified during 1967–2002, when the human population in surrounding districts increased at a mean rate of c. 2–4% per year. Further research is required to determine whether our findings are representative of the species in other seasons and at other sites.

Interest in the ecology and conservation of birds in coffee plantations around the world has greatly increased since the early 1990s, especially in the Neotropical region. Much of the interest was inspired by untested hypotheses proposing that extensive conversion of traditional coffee plantations, grown under a diverse canopy of shade trees, into modern, technified plantations with severely reduced shade canopy, contributed to concurrent declines of long-distance migratory birds. This possible relationship sparked major publicity campaigns during the late 1990s and continuing today, promoting shaded coffee plantations as quality habitat for migratory birds. Based on a review of the published scientific literature concerning avian use of coffee plantations, I summarize avian ecology in coffee agroecosystems, and evaluate the hypothesis that coffee plantations are important for the conservation of migratory or resident birds. While no literature has presented strong evidence that coffee plantations in general negatively affect bird populations of conservation importance, nonetheless published studies have not tested hypotheses that birds have greater survivability, fitness or productivity in coffee plantations compared with other available habitats (natural or artificial), or that any species selects coffee plantations over other available habitats for foraging or for breeding. While coffee plantations may have higher avian richness and abundance than other highly disturbed agricultural habitats and some natural habitats, more research is needed to evaluate whether and how certain coffee agroecosystems contribute to the conservation or decline of avian diversity.

The Japanese subspecies of Hodgson's Hawk-eagle, Spizaetus nipalensis orientalis, is considered threatened and has been designated as Endangered by the Japanese government. We determined the complete mitochondrial DNA (mtDNA) sequence of this species and designed a primer set to amplify a highly variable region of mtDNA, part of the control region (CR), based on this complete sequence. Using the primers, we amplified the CR and then determined the haplotypes of 178 samples collected at different sites in Japan. A nested cladistic analysis indicated that gene flow within some clades was restricted. The inference key implied that isolation by distance had caused the restriction of gene flow. Moreover, the ranges of the clades in which restricted gene flow was detected overlapped with the ranges of other clades. These results suggest that there is no fragmental population of Hodgson's Hawk-eagle in Japan and that this species has dispersed within short distances, at least in some lineages. Genetic diversity was high in comparison with other species. Therefore, at least in terms of genetic diversity, the Japanese population of Hodgson's Hawk-eagle is probably not in a critical situation.

The nominate Lesser Black-backed Gull Larus fuscus fuscus has experienced dramatic population declines on its northern breeding grounds. The global population is assessed at 18,000–119,000 pairs (BirdLife Finland), compared with 175,000 pairs of the British form graellsii (Wetlands International 2002). Although the species as a whole is not regarded as threatened, nominate fuscus is listed in the Red Data Books of Finland, Sweden, Norway, Estonia and Russian Karelia, i.e. it is classified as Threatened over its entire current range, falling into the Endangered and Vulnerable categories in the IUCN Red List categories.

The translocation of species to a new site plays an important role in the conservation of many threatened birds; however, the problems and processes involved in planning and implementing such translocations are rarely reported. In order to establish a second secure ‘insurance’ population of the endangered Rarotonga Monarch Pomarea dimidiata, or Kakerori, 30 young birds were moved from the Takitumu Conservation Area on Rarotonga to the 2,700 ha island of Atiu between 2001 and 2003. The translocation of this single-island endemic was to a site outside the historical range of the species, because the small, but rapidly growing, population on Rarotonga was considered to be highly vulnerable to a catastrophe, such as a cyclone, or the arrival of a new bird disease or predator. The translocation followed consultation with local communities, an assessment of the suitability of islands in the southern Cook Islands, and an assessment of the disease risk posed by the translocation. The translocation appears to have been successful because Rarotonga Monarchs have bred well in a variety of forest habitats on Atiu.

Steller's Eider Polysticta stelleri has a restricted arctic breeding range. The world population declined to c. 220,000 individuals in the late 1990s from an estimated 400,000–500,000 in the 1960s. The species has a limited global wintering distribution, occurring in marine habitats in north-east Europe, islands close to Kamchatka in Russia, and the eastern Aleutian Islands and south-west Alaska. European wintering numbers were estimated at 30,000–50,000 in the early 1990s, when the population was considered of favourable conservation status. Recent census data from the most important European wintering sites show annual declines of 8% in Norway since 1984, 9% in Estonia since 1994 and 22% in Lithuania since 1995, suggesting an overall 65% reduction in Europe. Counts in 1994 suggested that 30–50% of the European population wintered in Russia at that time. Current census data from Russia show similar declines along monitored sections of the Kola Peninsula wintering grounds since 1994. Accounting for trends in Russia, the current European wintering population could possibly stand at 10,000–15,000 individuals (a more than a 50% decline in 10 years), qualifying this population as Endangered under IUCN criteria. The changes in Baltic/Norwegian wintering numbers did not correlate with changes in the extent of ice-free marine waters in the Kola Peninsula/White Sea areas, but changes in annual numbers in Norway were correlated with winter North Atlantic Oscillation indices. Variation in annual numbers in the Baltic Sea correlated with projected number of juveniles among wintering birds. However, none of the possible causes discussed in this paper could fully explain the decline in Steller's Eider, confirming the need for comprehensive monitoring of the population throughout its winter range and for cohesive demographic monitoring to target effective conservation action.

The original 24,900 km2 distribution of the Cauca Guan Penelope perspicax is currently reduced to less than 750 km2. In addition, forest within this range is fragmented and populations are isolated. We conducted surveys to determine the presence of the Cauca Guan at several locations within its present range, and mapped all known localities with guan records. Cauca Guan populations are confirmed to occur at only four sites, and the guan is absent from several localities with historic records. Some sites where the Cauca Guan is potentially present are highlighted. In particular, the possible existence of populations on the Pacific slope of the Andes requires confirmation, since a large habitat block is available. Population density estimates were obtained in two protected areas. In Yotoco Forest Reserve (559 ha) we estimated 8.6 guans km−2. At Otún-Quimbaya Flora and Fauna Sanctuary (489 ha), guans fed seasonally on leaves of Chinese ash Fraxinus chinensis, an exotic tree planted in a revegetation programme. Seasonal densities in ash plantations were over 100 guans km−2, whereas in mixed forest, mean density was 41.6 birds km−2. Ash has apparently become a key resource for this population. Although these two populations are within protected areas and have relatively high densities, they are small and isolated and require continued monitoring.

The Sierra Madre Sparrow (Xenospiza baileyi) is a highly endangered and endemic species of the highlands of south-central Mexico, where it is resident in bunchgrass (Gramineae) and adjacent marshy habitats in the southern Sierra Madre Occidental (Jalisco and Durango states) and in the mountains around the Valley of Mexico (Distrito Federal and the states of Morelos and Mexico). This species was first collected in the southern part of this range on 23 April 1945 at La Cima, D.F., where its persistence has been documented essentially continuously since 1951. The first specimens of the taxon were collected in the Sierra de Bolaños of extreme northern Jalisco on 3–10 March 1889, including the type from which the genus and species were described in 1931. Two populations have been found in southern Durango: one 30 miles (48 km) south-west of the City of Durango on 22 March 1931, and the other 5 miles (8 km) west of El Salto on 16–17 June 1951. The Sierra Madre Sparrow has not been otherwise confirmed in the northern part of its range, which in July 2004 led us to conduct an extensive search for it in these areas of Durango and Jalisco as well as south-western Zacatecas. Here we present the findings from that search, during which several sites were intensively surveyed and a single population of this sparrow was located – a new one between the city of Durango and El Salto, Durango. This rediscovery increases possibilities for understanding the biogeography, ecology and basic requirements of the Sierra Madre Sparrow, information of fundamental importance for proposing measures that promote its conservation in any of its remaining populations.

The Bawean Serpent-eagle Spilornis baweanus is endemic to the 190 km2 island of Bawean in the Java Sea (Indonesia) where it is the only resident diurnal raptor. A 15 day study in 2002 revealed that the species is present in small numbers throughout the island. The eagle's abundance was assessed by an island-wide survey and by sampling 28 1 km2 plots covering five habitat types. There was a strong positive correlation between abundance and contact time in plots. Compared with other habitat types, the number of, and contact time with, serpent-eagles was higher in tall forest. The species was not recorded in mangrove and coastal forest. There was no correlation between the eagle's abundance and the nearest distance to villages. Based on the distribution of forest and the abundance of adult pairs within these forests, the available habitat totals 92 km2, in which some 60–75 adult serpent-eagles remain. Semi-structured interviews with the islanders revealed that recreational hunting poses the largest threat to the survival of the Bawean Serpent-eagle, and that the increase in recreational hunting was a relatively recent phenomenon. The forest on Bawean, including that in two reserves, is poorly protected and illegal logging and burning are widespread. The low degree of habitat protection, the severe threat posed by hunting and the very small population size of the eagle qualify the species to be included in the IUCN Red List as Critically Endangered. In order to safeguard the Bawean Serpent-eagle hunting must be stopped immediately and the remaining habitat needs to be better protected. This is probably best achieved by a conservation body in which local and regional authorities and NGOs cooperate.

There is a clear need for protection of the migration and wintering grounds of the Globally Threatened Aquatic Warbler Acrocephalus paludicola, but little is known about them and the threats they face. To narrow this gap, a desk study was performed between May 1998 and October 2004. Information on known wintering grounds was sought by means of questionnaires, personal communications, ringing data, publication and internet searches. Results show that the Aquatic Warbler has so far been recorded in nine African countries, but with recent records since 1980 from only five countries (Egypt, Ghana, Mauritania, Morocco and Senegal). All present data suggest that the Aquatic Warbler migrates through north-west Africa in autumn and spring, with the wintering grounds limited to wetlands of western sub-Saharan Africa, with verified records only from Mauritania, Mali, Senegal and Ghana during the months of August to April. The species was almost always found in habitats similar to that of its breeding grounds, i.e. in Carex, Juncus or Phragmites (sedge, rush and reed) associations, but also in dense grasses, shrubs and other vegetation found in freshwater marshes, flooded or wet meadows, and along the edges of backwaters, flood basins, lagoons, lakes, ponds, rivers and wadis. Recent research suggests that at least some Aquatic Warblers may actually winter further south than the present data suggest, in countries such as The Gambia, Guinea-Bissau, Guinea, Sierra Leone, Liberia, Ivory Coast, Ghana, Togo and Benin, or maybe even in so far unexplored wetlands in Central or East Africa. Because wetlands throughout Africa face imminent threats from agricultural and tourist development, more fieldwork is urgently needed to further pinpoint the migration and wintering grounds of the Aquatic Warbler.

The biggest and best-known population of Brazilian Merganser Mergus octosetaceus occurs in the region of Serra da Canastra National Park, Minas Gerais, Brazil. However, prior to our study it had been recorded in only six sites in the park and its buffer zone. In 2001 and 2002 we surveyed the region to confirm the occurrence of the Brazilian Merganser in different areas within and outside the park. Samples were taken by walking along the banks of the watercourses, using tape-recording playback. We sampled 49 stretches of streams, and observed Brazilian Mergansers in nine localities, totalling 29 sightings. Taking into consideration the reports by biologists who participated in the fieldwork for the review of the park's management plan, the guaranteed reports of the occurrence of the species from local people, and the sites where its occurrence is judged to be very likely, we believe there are at least 81 individuals in the areas sampled. We point out that the absence of records in an area does not mean the absence of the species. Indeed, the presence of the Brazilian Merganser was confirmed, after the end of our field studies, in some stretches where we indicated its occurrence as very likely but had failed to detect it during the sampling. All anthropogenic activities that influence the quality and integrity of the rivers and their banks are a potential threat to Mergus octosetaceus. Efforts to preserve the species entail the conservation of its habitats, which comprise important watercourses and the surrounding natural vegetation.

To investigate whether nest predation can influence the breeding success of Ferruginous Ducks Aythya nyroca, artificial nests were used in Nagyberek, the strictly protected swamp pond of the Juniper Woodland Nature Conservation Area (Somogy county, south Hungary). Experimentation lasted for 4 weeks, a similar length of time to the incubation period of Ferruginous Ducks. After 1 week, 80% of nests were intact, after 2 weeks only 46%, and after 3 and 4 weeks only 2% remained undamaged. Nest survival rates were not affected by the width of the sedge stands, but as water levels surrounding nests decreased, nests became more accessible to Wild Boar Sus scrofa and other land mammal predators, which increased the rate of predation. Artificially maintaining water levels would not only decrease the predation rate of nests, but would also maintain feeding areas for ducks. Wild Boar were the main cause of clutch loss in this area, and therefore by management measures, such as a reduction in their abundance or attracting them away from potential nesting sites by providing food elsewhere, the breeding success of the Ferruginous Ducks may be further improved.

Censuses were conducted from 15 June to 6 July and from 15 to 24 October 2003 throughout Lake Titicaca to determine the current distribution and abundance of the Titicaca Flightless Grebe Rollandia microptera. We surveyed 22 of an estimated 25 local population distribution ranges. We defined these as shallow offshore feeding areas in combination with coastal beds of tule-rushes (Schoenoplectus tatora). At this large spatial scale we found variously sized populations of grebes in all 22 of the areas surveyed and a negative correlation between the total area of tule-bed within each area and grebe density. Absolute counts of grebes pooled across all sites produced a total of 2,582 individuals on Lake Titicaca. Given that some sites were either partially surveyed or not surveyed at all we feel that this is an underestimate of the total lake population. Limitations in methodologies used precluded the classification to development stage of approximately 43% of the individuals counted. Among the remaining 57% of the individuals counted we identified 732 mature adults within the total population of Lake Titicaca. When combined with information from other sources and current knowledge of habitat fragmentation our results support the classification of the species in the IUCN Red List as Endangered under criterion C2a. Of 20 local fisherman interviewed on the Bolivian side of Lake Titicaca, 90% confirmed that individuals of Titicaca Flightless Grebe incidentally drown in their fishing nets although responses varied from weekly to monthly estimates of by-catch. Although this study provides the first estimate of the total population of Titicaca Flightless Grebes on Lake Titicaca, a standardized monitoring programme will be required in order (1) to provide a baseline for the interpretation of future population trends in the context of impacts generated by human activities and (2) to aid in future management decisions for the long-term conservation of the species.

Alpha taxonomy involves delineation of the basic unit of biology: the species. The concepts by which we define species, however, have been controversial, with several alternatives competing at present, some creating fewer and some more species units, depending on interpretation of species limits. Although it is tempting to assume that species concepts would have little interaction with the geographic foci of species richness and endemism — and some have so argued — this assumption does not withstand careful analysis. In this paper, I develop a first-pass assessment of Philippine bird taxonomy under an alternative species concept, and compare the results with the traditional biological species concept lists. Differences between the two lists were dramatic, but not just in numbers of species; rather, new, previously unrecognized or previously underappreciated foci of endemism were noted. A thorough understanding of the taxonomic basis of species lists is therefore critical to conservation planning.

The digestive tract of many animals is morphologically flexible and can adjust over time to make the most efficient use of the foods available. Differences between captive and wild diets often cause large differences in the gut morphology of captive and wild birds. This is potentially an issue when captive-bred birds are used to establish or supplement populations in the wild, such as in the Brown Teal Anas chlorotis, an endangered duck endemic to New Zealand. We compared the size and mass of the digestive organs (proventriculus, gizzard, small intestine, caeca, rectum and liver) of 57 wild, eight captive and four captive-bred released Brown Teal. Captive Brown Teal had much shorter and lighter small intestines and caeca than wild Brown Teal. These differences could reduce the ability of captive-bred teal to efficiently digest a wild diet in the weeks following release, and are likely to contribute to the number of released teal found dead in extremely poor nutritional condition. Increased fibre and diversity in the captive diet together with supplementary feeding post-release are recommended to improve the survival of captive-bred Brown Teal released to the wild.

In this paper, we analyse the non-breeding distribution of Dupont's Lark, one of Europe's most threatened and least known passerines, using breeding and wintering censuses, published and unpublished records of individuals outside their breeding areas, and data on radio-tagged individuals and field observations. Results suggest that most breeding Dupont's Larks remain around the breeding sites during the non-breeding season, including those populations from areas with harsh winter weather conditions. The probability of winter presence of Dupont's Larks in breeding areas was significantly correlated with population size, but the altitude of the locality studied did not influence winter presence. Records outside breeding areas indicate that part of the population moves some distance after breeding, occupying lower sites and principally those with agrarian substrates. We also found several records far from well-known breeding areas during the breeding period. Our results suggest that Dupont's Lark seems to be a partial migrant, wintering in breeding areas in large numbers (probably the adult fraction of the population) but with some part of the populations moving to other areas. The movements are short-distance in most cases, although medium- and long-distance dispersive movements were also detected. We discuss the implications of these results for the design of specific conservation measures for the species.

Hornbills are useful indicators of forest condition and human disturbance because they require large tracts of unfragmented forest with large fruiting trees for feeding and nesting. They are relatively large-bodied, which makes them targets for hunting. Density estimates of such species are critical for population monitoring and serve as a baseline against which future changes can be measured. In this study we used variable-width line transect surveys to estimate the densities of nine hornbill species in the Bala portion of the Hala-Bala Wildlife Sanctuary on the Thai–Malaysia border, one of the few remaining areas of lowland forest in Thailand. The hornbill species were: Rhinoceros (Buceros rhinoceros), Great (B. bicornis), Helmeted (B. vigil), Black (Anthracoceros malayanus), Bushy-crested (Anorrhinus galeritus), White-crowned (Aceros comatus), Wrinkled (A. corrugatus), Wreathed (A. undulatus) and Plain-pouched (A. subruficollis). Between January 2001 and April 2002, 11 transects along trails, old logging roads and one paved road were surveyed once per month. A total of 1,261 observations of the nine species were made during the observation period. Estimates for Rhinoceros and Helmeted Hornbill were 2.69 and 1.21 individuals/km2, respectively, and were similar or slightly higher than densities reported elsewhere. Estimates for Great (0.12), Bushy-crested (0.64), Wrinkled (0.08), White-crowned (0.08) and Wreathed Hornbills (0.69) were generally lower than estimates from other areas in the region. It was not possible to obtain density estimates using distance sampling for Black and Plain-pouched Hornbills due to the small number of observations, but our data did suggest that their densities were also low (<0.10 individuals/km2). Potential reasons for differences between this and other areas include a scarcity of lowland habitat, habitat isolation, lower abundance of specific fruit resources and interspecific competition, but these factors require further investigation. Standardization of survey techniques among studies would significantly improve assessments of habitat requirements as well as of the effects of human disturbance on Asian hornbills, most of which are globally threatened. Our study specifically underscores the need for additional research on the Plain-pouched Hornbill, the most threatened of the species studied, as density estimates from elsewhere in its range are lacking, while more generally it highlights the need to investigate the underlying causes of the often substantial variation in hornbill densities among species and sites.