Populations of Gouldian Finch Erythrura gouldiae, an estrildid endemic to northern Australia, declined markedly during the last two or three decades. A survey revealed the species in two areas of Northern Territory, with an estimated 2,000 individuals, including juveniles. Recaptures were low either because of mortality or dispersal. Gouldian Finches eat predominantly Sorghum spp. seeds, but consume other seeds before Sorghum ripens. Other finches declined as cattle increased. Gouldian Finches breed in tree hollows of predominantly two species of eucalypts that grow on rocky slopes. They require water within about 4 km of the nest site. There was no shortage of nesting hollows at the known breeding sites. Fires in the early to mid-dry season allow birds access to seed without damaging trees, but later fires can destroy trees with nest hollows and remove shade. Average clutch-size is 5.2. About 72% of eggs laid, and 63% of nests, fledged young. Pairs lay up to three clutches in a season (February-August), the length of which may depend on rainfall during the preceding wet season. Air-sac mite (Sternostoma tracheacolum), found in 62% of Gouldian Finches sampled, may be preventing the species recovering to former numbers. New colonies of Gouldian Finches should be identified, populations monitored, and habitat managed by effecting patchy burns by low-intensity fires early in the dry season.

Juniper forest once covered much of the upper slopes (2,000-3,000 m) of the Rift Valley escarpment in the Asir mountains of south-west Saudi Arabia. One of the best preserved tracts is at Raydah near Abha, where a complete altitudinal floristic zonation persists. Although only about 125 bird species have been recorded in the area, the community is of considerable conservation interest due to high densities of endemic species and resident or breeding Afrotropical forest species, together with a wide range of diurnal raptors, owls and nightjars. The forest does not appear to support many Palearctic— African migrants during periods of passage, although several Palearctic species overwinter in considerable numbers. Scrub and mixed deciduous and riparian forest scattered throughout the junipers hold highest bird diversities but the juniper stands are vitally important to the well-being of Yemen Thrush Turdus menachensis, Yemen Warbler Parisoma buryi and Yemen Linnet Carduelis yemenensis populations. The disappearance of juniper forest from neighbouring Yemen necessitates that highest priority conservation action should be given to the remaining intact forests such as Raydah, in an attempt to preserve their unique biodiversity.

Seabird populations suffer from a variety of natural and human-induced sources of mortality and loss of lifetime reproductive output. On the outer coast of Washington State, Common Murre Uria aalge populations have been in decline for approximately the last decade and are currently reproductively active only at Tatoosh Island. These murres nest in two basic habitat types: crevices (25% of the population) and larger cliff-top subcolonies (75%). Murres in cliff-top subcolonies have suffered dramatic reductions in reproductive success in recent years relative to conspecifics nesting in the crevices, primarily due to egg predation by Glaucous-winged Gulls Larus glaucescens and Northwestern Crows Corvus caurinus, facilitated by the presence of Bald Eagles Haliaeetus leucocephalus. Because predator removal is not feasible and creation of additional crevice habitat is difficult, expensive and potentially ineffective, we have designed a temporary habitat modification (the “silk forest”) which replaces the natural vegetation cover and modifies the interaction between murres and eagles. Within the test subcolony, murres nesting under and immediately adjacent to the silk forest produced nearly twice as many eggs per square metre as their conspecifics nesting in adjacent exposed-ground areas.

When formulating conservation priorities, conservation biologists often rely on published lists of species-level taxa. This paper discusses the nature and taxonomic status of “species” and “subspecies” and different ways of defining “species”. Species are here taken to be terminal and evolutionarily independent units which are qualitatively diagnosable and reproductively cohesive; genealogical biodiversity is thus taken for what it is in the first place, namely the observable result of evolutionary history, an approach which has become known as the phylogenetic species concept (PSC). In contrast to the widely applied Mayrian or “biological” species concept (BSC), no inferences are made about how the degree of morphological differentiation of allopatric but seemingly closely related taxa translates in the absence or presence of reproductive isolation. Many diagnosably distinct endemic island taxa have traditionally been treated as “subspecies” of widespread “polytypic biological” species. At the same time, the “subspecies” category is also used to name arbitrarily delimited sections of intraspecific clinal variation. Thus, the “subspecies” category subjects entirely different evolutionary phenomena to the same hierarchical level through the use of trinomials. Nevertheless, and despite the discrepancy in ontological status among its contents, “subspecies” are usually considered to be of lower evolutionary and/or conservation status than “species” and this has resulted in low conservation priorities allotted to diagnosably distinct island endemics, many of which have traditionally been considered to be “merely Mayrian subspecies”. This has been recognized by some authorities who, because of the threatened status of certain island taxa, advocated treating them binomially in order to generate appropriate conservation measures to save them from extinction, without however justifying their action by any sort of phylogenetic reasoning. Although well intended and sometimes quite successful as regards the follow-up by conservationists, this demonstrates the arbitrary manner in which “species” can be defined under the BSC. Some examples of endemic taxa from eastern Atlantic islands are discussed, demonstrating the way “list thinking” and the lack of phylogenetic reasoning among conservationists translates into the presence or absence of conservation actions. Some of the criticisms of the PSC by adherents of the BSC are discussed. It is advocated that conservationists replace “list thinking” with “lineage thinking”.

The breeding activity of birds in the Oku region of Cameroon appears to be highly seasonal, with the majority of species breeding exclusively during the months of the dry season, November to April. Forest fires, which in the dry season often destroy large areas of suitable habitat, are likely to strongly influence nest failures in many species, including some endemic and endangered ones.

Whether they are “mere subspecies” or “phylogenetic species”, the omission of such forms from the ICBP (now BirdLife International) Red Data Book (RDB) was my personal decision, and I have long been sensitive to the type of charge contained in Hazevoet's paper (preceding pages). This decision, taken back in 1982, was based on the time, space and cost that would have been involved in the systematic review and inclusion of subspecies (RDBs being much fuller treatments than those in Birds to watch 2, the work criticized for the omission by Hazevoet), and on the fact that the future of the many “full” species needing urgent documentation and action consequently stood to be compromised.

The Socorro Mockingbird Mimodes graysoni is endemic to Socorro Island, Mexico, and has declined dramatically in this century. Study of its population size, distribution, and structure is currently under way to help assess its conservation status. In 1993 and 1994 215 Socorro Mockingbirds were ringed; a modified Lincoln index yielded a population estimate of 353 ± 66 individuals in 1994. Most of the birds observed occupied a small region at high elevation that covered less than 10% of the island's area. Average territory size was 1.48 ± 0.71 ha, n = 11. Of the 86 birds banded in 1993 67% were adults and 33% subadults. Of the 170 birds detected in 1994, including new captures and ringed individuals from the previous year, 44% were adults and 56% subadults. In 1993 62% were male and 38% female; in 1994 56% were male and 44% were female. The sex ratio of the population was similar to that of other undisturbed Mimid populations, which argues against the possibility that cats might prey disproportionately on incubating females. Moreover, the presence of young birds in both years indicates that successful reproduction was occurring and that the population was not suffering catastrophic nest failure, as might be caused by cat predation on nestlings or fledglings. Northern Mockingbirds Mimus polyglottos, which first arrived on Socorro in the 1970s, occupied mainly open areas produced by sheep grazing, which suggests that the Socorro Mockingbird is not being competitively displaced. Our observations indicate that habitat degradation by sheep is the most probable cause of the Socorro Mockingbird's decline.

The Norfolk Island Boobook Ninox novaeseelandiae undulata is confined to the small, isolated Norfolk Island group, an Australian territory. On morphological and biogeographical grounds, it is here classified as a large, distinctive subspecies of the New Zealand Morepork N. novaeseelandiae. In 1986 only one specimen, a female, survived. A shortage of large trees with suitable nesting holes appeared to be the immediate problem. The Australian Nature Conservation Agency, islanders and New Zealand wildlife authorities have cooperated in an attempt to re-establish an owl population in situ. Nest-boxes were erected in trees in the area frequented by the female and were used readily as roosts. In September 1987, two male New Zealand Moreporks were introduced. The female paired with one male and produced four hybrid F offspring (in 1989 and 1990). Two of these paired in mid-1991 and have since produced five F offspring (two in 1993 and three in 1994). The original female remains paired but now appears to be reproductively senile. At present there seems to be a shortage of mature males, since two female offspring are paired and both lay eggs and attempt to incubate them in the same nest; and a lone female has established a territory. In early 1995 all eleven owls appeared to be alive in the wild. The effort is low-cost, requires relatively little manpower, is carried out with minimal disturbance to the owls, and goes hand in hand with other conservation programmes.

New Zealand's endangered Black Stilt Himantopus novaezelandiae is confined largely to South Island's upper Waitaki River Basin, where it breeds on braided riverbeds and associated wetlands. It is under pressure from nest predators (particularly introduced carnivorous mammals), habitat loss and hybridization with the Pied Stilt H. himantopus leucocephalus. Management focuses on localized predator control, habitat enhancement, egg manipulation and cross-fostering, and more recently captive breeding and release. Future management may be extended to establish an island population.

The Vanuatu Megapode Megapodius layardi is endemic to Vanuatu (formerly New Hebrides). On the volcanic island of Ambrym it exhibited three different incubation strategies. It incubated its eggs (1) in burrows between decaying roots of trees; (2) in burrows at large communal nesting grounds in volcanically heated soils; and (3) in burrows on sun-exposed beaches. No mounds were found. In a study by the Vanuatu Protected Areas Initiative (VPAI) nesting grounds were mapped to monitor the population. Three communal nesting grounds were surveyed. Two were on promontories and one on a beach. The burrows were classified into three categories: active, probably active and old. The density of active burrows per hectare of the three sites was 82.4 for Promontory A, 43.5 for Promontory B and 10.5 for Buwoma Beach. Additional information of this poorly known species was collected from villagers in the north-west and west of the island. They have noticed a decrease in the numbers of birds and have expressed concern about the future of the species.

Le Mègapode de Vanuatu, Mègapodius layardi, est uniquement originaire de Vanuatu, (dit 'Nouvelles Hebrides' autrefois). Sur l'île volcanique d'Ambrym, il fait preuve de trois stratégies d'incubation différentes. II couve ses oeufs premièrement dans des terriers situés entre les racines pourries des arbres, duexièmement dans des terriers situés dans des grands terrains communaux de nids, et troisièmement dans des terriers situés sur des plages exposées au soleil. Aucan monticule n'est évident. Lors d'une étude par l'lnitiative des Terrains Protégés de Vanuatu (Vanuatu Protected Areas Initiative, VPAI), les terrains de nids furent plannifiés afin d'étudier la population. Trois terrains communaux de nids furent enquêtés. Les terriers furent classifiés en trois catégories; actifs, probablement actifs, et anciens. La densité des terriers actifs par hectare des trois terrains étaient de 82.4 pour le Cap A, 43.5 pour le Cap B, et de 10.5 pour la plage de Buwoma. Des données additionelles pur cet espèce peu connu furent rassemblées des villageois dand le nord ouest et l'ouest de Iîle. Ils ont constaté une diminuation des nombres d'oiseaux et ils ont exprimé leurs soucis vis-à-vis le future de l'espèce.

Wan pijin, nem blong hem Namalau, Megapodius layardi, i stap long Vanuatu nomo (bifo, New Hebrides i nem blong Vanuatu). Long Ambrym, wan aelan wetem volkeno, Namalau i yusum trifala defren fasin long makem nes blong hem so heg blong hem i stap gud nomo; 1) ol i putum heg blong hem nekis rus bio tri wea ol rus i rotin 2) ol i putum heg blong hem wetem plenti narafala Namalau heg long graon wea volkeno i mekem groan i hot, 3) ol i putum heg blong hem insead sanbij wea san i makem i hot. Long Ambrym ol man i no fanem ples wea Namalau putum heg blong hem insaed wanples wea i gat plenti lif mo tri antap graon. Vanuatu Protected Areas Initiative (VPAI) i mekem wan stadi long ples wea ol Namalau i putum heg mo ol i mekem wan map. VPAI i lukum long trifala ples wea Namalau i putum heg. Tufala ples i stap long graon wea i go aot long solwota mo namba tri ples i stap long wan sanbij. I gat trifala defren grup blong ples wea ol i putum heg; niufala, klosap niufala mo olfala. Namba blong nuifala ples wea Namalau i putum heg insaed long wan hectare i 82.4 long pies A, 43.5 long ples B mo 10.5 long Buwoma sanbij. I no gat plenti save blong Namalau mo VPAI toktok plenti wetem ol man blong vilej long notwes mo wes Ambrym. Ol man is e namba blong Namalau i go daon bigwan mo ol i wari long fiuja long Namalau.

Anian veterinarians have much expertise which can be usefully applied to in situ avian conservation programmes. Their knowledge can assist conservation biologists in many of the manipulations which they are currently applying to endangered wild birds in their natural range. Possibilities are outlined and illustrated by examples from various field programmes.

The Sokoke Pipit Anthus sokokensis is endemic to coastal forests in Kenya and Tanzania (Keith et al. 1992). It has been listed as occurring in the Pugu Hills, Tanzania (Howell 1981, Collar and Stuart 1985, 1988), having been first recorded in 1938 (Moreau 1940). Two further records from this area are reported in Collar and Stuart (1985), in 1955 and 1981. Harvey and Howell (1987) noted that, over the period 1968–1978, many visits to the area by a variety of observers, some using mist-nets, failed to locate Sokoke Pipit.

Recent genetic analysis of New Zealand Cyanoramphus parakeets has direct implications for conservation management. A subspecies of Yellow-crowned Parakeet, Forbes's Parakeet C. auriceps forbesi, is genetically distinct from mainland Yellow-crowned Parakeets C. a. auriceps and should be elevated to species status. Even amongst mainland populations of Yellow-crowned Parakeets there was extensive genetic diversity. Interspecific hybridization is a major concern of Cyanoramphus conservation. The near-extinction of Forbes's Parakeets by genetic swamping through hybridization with Chatham Islands Red-crowned Parakeets C. novaezelandiae chathamensis has been averted by active management, involving culling of hybrid and Red-crowned Parakeets and habitat enhancement. Despite two decades of hybridization between these species a distinct Forbes genetic type still exists, probably due to a tendency for hybrids to backcross with Red-crowned Parakeets. Hybridization between Cyanoramphus species may also be a problem in other highly modified habitats, such as parts of the Auckland Islands. More information is needed to determine the extent of this problem. Captive populations include a significant proportion of interspecific hybrids and therefore should not be used as founder stock for releases into the wild, as has been done in the past. The Orange-fronted Parakeet should be reinstated as a separate species, Cyanoramphus malherbi, rather than a colour morph of Yellow-crowned Parakeet, at least until conclusive evidence is available to resolve the question of its taxonomic status. The conservation of this very rare species (and indeed all species of Cyanoramphus)- has been severely handicapped by lack of scientific information on which to base management decisions.