Habitat selection in reintroduced bird populations: a case study of Stewart Island robins and South Island saddlebacks on Ulva Island

An understanding of the mechanisms influencing habitat selection in reintroduced bird populations is fundamental for successful translocation programmes. Plant species composition, abundance, structure and food availability are likely to influence animal movement and habitat choice, but few studies have evaluated their combined effect on habitat selection of translocated birds. Stewart Island robins (Petroica australis rakiura) and South Island saddlebacks (Philesturnus carunculatus carunculatus) are two threatened New?Zealand bird species that have been reintroduced to Ulva Island (Stewart Island). We hypothesised that their initial settlement patterns were driven by habitat quality. We tested this hypothesis by comparing habitat components between occupied and unoccupied habitats as the population grew after initial tanslocation. We also modelled probabilities of site selection as a function of the composition and structure of vegetation, availability of food (invertebrate composition) and nesting resources (cavity type). Founding pairs of both species first established territories in coastal habitat in the western part of the island, which is characterised by structurally complex broadleaved vegetation. Birds also selected sites with a greater abundance and diversity of food resources. Thus in the early stages of population establishment robins and saddlebacks appear to select high quality habitat that offers enhanced cover and foraging opportunities.     

Multi-scale habitat models for reintroduced bird populations: a case study of South Island saddlebacks on Motuara Island

Understanding resource selection by animals is important when considering habitat suitability at proposed release sites within threatened species recovery programmes. Multi-scale investigatory approaches are increasingly encouraged, as the patchy distribution of suitable habitats in fragmented landscapes often determines species presence and survival. Habitat models applied to a threatened New Zealand forest passerine, the South Island saddleback (Philesturnus carunculatus carunculatus), reintroduced to Ulva Island (Stewart Island) found that at landscape scale breeding pairs? preferences for sites near the coast were driven by micro-scale vegetation structure. We tested these results by examining models of breeding site selection by a reintroduced saddleback population on Motuara Island (Marlborough Sounds) at two scales: (1) micro-scale, for habitat characteristics that may drive breeding site selection, and (2) landscape scale, for variations in micro-scale habitat characteristics that may influence site colonisation in breeding pairs. Results indicated that birds on Motuara Island responded similarly to those on Ulva Island, i.e. birds primarily settled at the margins of coastal scrub and forest and later cohorts moved into larger stands of coastal forest where they established breeding territories. Plant species composition was also important in providing breeding saddleback pairs with adequate food supply and nesting support. However, Motuara Island birds differed in their partitioning of habitat use: preferred habitats were used for nesting while birds were foraging outside territorial boundaries or in shared sites. These differences may be explained because Motuara has a more homogeneous distribution of microscale habitats throughout the landscape and a highly bird-populated environment. These results show that resource distribution and abundance across the landscape needs to be accounted for in the modelling of density?bird?habitat relationships. In the search for future release sites, food (invertebrates and fruiting tree species) should be abundant close to available nesting sites, or evenly spread and available throughout the landscape.     

Variation in breeding success among reintroduced island populations of South Island Saddlebacks Philesturnus carunculatus carunculatus

South Island Saddlebacks Philesturnus carunculatus carunculatus were once found throughout the South Island of New Zealand, but by the early 1960s were confined to the island of Big South Cape, in the extreme south of the country. All subsequent reintroduced populations of South Island Saddlebacks are derived from 36 surviving birds from this relict population. The aim of this study was to compare the breeding success of three recently reintroduced populations of Saddlebacks relative to their distance from, and habitat similarity to, the relict population. The three study islands show a latitudinal cline with Ulva, Breaksea and Motuara Islands located 60, 190 and 810 km north of Big South Cape, respectively. Saddlebacks on Ulva and Breaksea appeared to prefer to establish breeding territories in coastal scrub, the dominant habitat feature of Big South Cape. The area of coastal scrub habitat was much smaller on Motuara, where breeding territories were instead scattered through broadleaf forest habitat. Nesting success, calculated using Mayfield's method, was significantly greater on Ulva (73%) than on Breaksea (32%) or Motuara (19%) owing primarily to higher egg fertility and hatching success. Although egg failure rates were highest on Motuara, the island least similar to Big South Cape, they were also relatively high on Breaksea where the habitat was similar to Ulva and Big South Cape. Therefore, the results only partially support the hypothesis that nesting success should decrease with increasing habitat difference associated with increasing latitudinal distance from the source population. The data from this 1-year study lay the groundwork for examining further hypotheses on the effects of reintroducing endangered species outside their contemporary range, but within their historical range.     

Moderate inbreeding depression in a reintroduced population of North Island robins

Reintroductions of threatened species are increasingly common in conservation. The translocation of a small subset of individuals from a genetically diverse source population could potentially lead to substantial inbreeding depression due to the high genetic load of the parent population. We analysed 12 years of data from the reintroduced population of North Island robins Petroica longipes on Tiritiri Matangi Island, New Zealand, to determine the frequency of inbreeding and magnitude of inbreeding depression. The initial breeding population consisted of 12 females and 21 males, which came from a large mainland population of robins. The frequency of mating between relatives ( f >0; 39%, n =82 pairs) and close relatives ( f =0.25; 6.1%) and the average level of inbreeding ( f =0.027) were within the range reported for other small island populations of birds. The average level of inbreeding fluctuated from year to year depending on the frequency of close inbreeding (e.g. sib–sib pairs). We found evidence for inbreeding depression in juvenile survival, with survival probability estimated to decline from 31% among non-inbred birds ( f =0) to 11% in highly inbred juveniles ( f =0.25). The estimated number of lethal equivalents based on this relationship (4.14) was moderate compared with values reported for other island populations of passerines. Given that significant loss of fitness was only evident in highly inbred individuals, and such individuals were relatively rare once the population expanded above 30 pairs, we conclude that inbreeding depression should have little influence on this robin population. Although the future fitness consequences of any loss of genetic variation due to inbreeding are uncertain, the immediate impact of inbreeding depression is likely to be low in any reintroduced population that expands relatively quickly after establishment.     

Complex effects of habitat loss on Golden Eagles Aquila chrysaetos

The development of commercial forests presents potential threats to large raptors that rely on prey caught in open country. We examined the effect of afforestation of breeding habitat used by a population of Golden Eagles Aquila chrysaetos in Scotland where, over the last 50 years, extensive stands of exotic conifers have been planted. Using data for 31 years on territory occupancy and breeding success, together with spatiotemporally dynamic mapping of forest cover and predicted areas of territory‐use in a Geographical Information System, we examined relationships between forest cover and Eagle ecology at landscape and individual territory scales. Several territories were abandoned during the earliest phases of forest planting, but relatively few were apparently lost to later plantings. Territories with poorer breeding productivity appeared to be more vulnerable to abandonment than territories with better breeding productivity. At the landscape scale, temporal differences in breeding productivity were negatively related to the extent of forest cover, although productivity of individual territories showed no clear relationship with forest cover. Several territories with less than a 5% increase in forest cover experienced reduced productivity; however, territories least constrained by neighbouring pairs of Eagles showed an increase in productivity. Territories experiencing the greatest increases in forest cover showed a greater use of spatially separated nest‐sites by occupying pairs. Hence, pairs that were less constrained by neighbours appeared to compensate for loss of open habitat by shifting their territory‐use, whereas pairs that were more constrained could not compensate for open habitat loss and suffered reduced productivity (and, probably in some cases, abandoned the territory). We suggest that simple guidelines based on the extent and locations of habitat loss are inadequate when predicting effects on large territorial raptors such as Golden Eagles. Consideration should also be given to the ‘quality’ of a territory or occupying pair, as well as the extent to which territory‐use is constrained by neighbouring pairs or other ‘unsuitable habitat’ which may have been affected by previous episodes of open habitat loss.     

Invertebrate survey of coastal habitats and podocarp forest on Ulva Island,Rakiura National Park,New Zealand

Abstract

Inventory of the invertebrate fauna is important to establish taxonomic diversity, abundance and distribution, and hence the conservation of indigenous biodiversity. Invertebrate assemblages have been documented in some broadleaf‐podocarp forests and grassland habitats in New Zealand, but not in dense stands of coastal forest or in mature podocarp forest. This survey aimed to provide a taxonomic inventory of terrestrial invertebrates and their habitat associations on Ulva Island (Rakiura National Park, Stewart Island), an off‐shore sanctuary of significant conservation value in New Zealand. We systematically documented the invertebrate assemblages collected in ground litter and on tree trunks on the island. The invertebrate specimens identified represented 4 phyla, 6 classes, 25 orders and 62 species. The invertebrate fauna reported in this survey was distinct from those of lowland shrubland and broadleaved‐Nothofagus forests on the mainland, but shared species with that reported from another similar off‐shore island, Codfish Island (Whenua Hou).     

Historic and contemporary levels of genetic variation in two New Zealand passerines with different histories of decline

We compared historic and contemporary genetic variation in two threatened New Zealand birds (saddlebacks and robins) with disparate bottleneck histories. Saddlebacks showed massive loss of genetic variation when extirpated from the mainland, but no significant loss of variation following a severe bottleneck in the 1960s when the last population was reduced from approximately 1000 to 36 birds. Low genetic variation was probably characteristic of this isolated island population: considerably more genetic variation would exist in saddlebacks today if a mainland population had survived. In contrast to saddlebacks, contemporary robin populations showed only a small decrease in genetic variation compared with historical populations. Genetic variation in robins was probably maintained because of their superior ability to disperse and coexist with introduced predators. These results demonstrate that contemporary genetic variation may depend more greatly on the nature of the source population and its genetic past than it does on recent bottlenecks.     

The risks, costs and benefits of using brodifacoum to eradicate rats from Kapiti Island, New Zealand

In 1996 an eradication operation against two species of rats (Rattus norvegicus and R. exulans was conducted on Kapiti Island (1965 ha) and its small offshore islands. Trials with non-toxic baits had been carried out to help determine the risks to non-target species, and research was undertaken to collect baseline data for measuring the response of vegetation, invertebrates, reptiles and birds to the removal of rats. Talon 7-20 bait (containing 0.002% brodifacoum) was distributed over Kapiti Island in September and October by helicopter and by hand, while bait stations were used on the offshore islands. Impacts on birds and reef fish were investigated. Although there were non-target bird deaths as a result of the poisoning operation, post-eradication monitoring indicated that the toxin had no deleterious effect on breeding and most losses would be rapidly made up by recruitment of new individuals into the breeding population. There was no evidence that reef fish were negatively affected. The successful removal of rats has apparently resulted in a significantly improved survival rate for stitchbirds (Notiomystis cincta) and saddlebacks (Philesturnus carunculatus). Benefits to other taxa are expected and will be documented as follow-up studies are completed.     

Distribution and nesting density of the Philippine Eagle Pithecophaga jefferyi on Mindanao Island,Philippines: what do we know after 100 years?

The Philippine Eagle Pithecophaga jefferyi , first discovered in 1896, is one of the world's most endangered eagles. It has been reported primarily from only four main islands of the Philippine archipelago. We have studied it extensively for the past three decades. Using data from 1991 to 1998 as best representing the current status of the species on the island of Mindanao, we estimated the mean nearest-neighbour distances between breeding pairs, with remarkably little variation, to be 12.74 km ( n  = 13 nests plus six pairs without located nests, se = ±0.86 km, range = 8.3–17.5 km). Forest cover within circular plots based on nearest-neighbour pairs, in conjunction with estimates of remaining suitable forest habitat (approximately 14 000 km²), yield estimates of the maximum number of breeding pairs on Mindanao ranging from 82 to 233, depending on how the forest cover is factored into the estimates.     

Feeding by Captive Rare Birds on Baits Used in Poisoning Operations for Control of Brushtail Possums

Non-toxic plain and cinnamon-flavoured carrots and cereal-based baits used in poisoning operations for control of the brushtail possum (Trichosurus vulpecula) were offered to seven species of captive rare birds at Mt Bruce National Wildlife Centre. Some individuals of all species ate plain baits. Antipodes Island parakeets (Cyanoramphus unicolor) preferred carrot to cereal- based baits, North Island kokako (Callaeas cinerea wilsoni) and North Island saddlebacks (Philesturnus carunculatus rufusater) preferred cereal-based baits to car-rots, but the other species showed no bait preference. Most baits eaten were greater than 2 g. Some individuals of all species also ate cinnamon-flavoured baits. However, cinnamon deterred North Island kaka (Nestor meriodionalis septentrionalis), Antipodes Island parakeets, and kokako from feeding on baits the first day offered, though not subsequently. Insufficient baits were eaten by North Island weka (Gallirallus australis greyi), red-crowned parakeets (Cyanoramphus novaezelandiae novaezelandiae), and Reischek's parakeets (Cyanoramphus n. hochstetteri) to determine whether they were also deterred by cinnamon. Only saddlebacks were definitely not deterred. All species except red-crowned and Reischek's parakeets probably ate sufficient to receive a lethal dose if the baits had been toxic. Baits may be made less acceptable to birds by increasing the strength or slowing the release of cinnamon, or by using a more repellent flavour. Because baits may always be acceptable to some birds, wildlife managers need to know the chances of wild rare birds feeding on baits before approving poisoning operations in areas where they occur.     

Habitat use by three rat species (

Ship rats (Rattus rattus) were removed from sites on Pearl Island, southern Stewart Island, in 2004 and 2005, to test whether they excluded Pacific rats (R. exulans) or Norway rats (R. norvegicus) or both from podocarp-broadleaf forest. As predators can influence habitat use in rodents, Pearl Island was selected because no mammalian predators of rodents are present. Rats were trapped in two other habitats to clarify rat distribution on the island and to obtain samples for stable isotope investigation of food partitioning within habitats. The experimental removal of ship rats failed, as Pacific rats were found to share forest and shrubland with ship rats. This result contrasted with the restricted distribution of Pacific rats on Stewart Island. Ship rats were ubiquitous, and appear to have been the dominant species in podocarp-broadleaf forest on Pearl Island. The largest species, the Norway rat, was trapped only on the foreshore of Pearl Island, but on Stewart Island it is more widespread. Ship rats and Norway rats were partitioning the coastal habitat by exploiting different food sources. Stable isotopic ratios (δ¹⁵N and δ¹³C) in muscle samples from Norway rats revealed a strong marine signature, suggesting intensive foraging in the intertidal zone. Ship rats trapped in the same habitat exhibited mixed terrestrial and marine sources in their diet. There was little obvious partitioning between ship rats and Pacific rats in forest, except a possible delay in breeding in Pacific rats relative to ship rats. Whether Norway rats select the intertidal zone to forage, or were excluded from forest by ship rats is unknown, but competitive exclusion is likely. Estimated densities of rats were low (2.1–5.1 rats ha^-1 in forest, 1.42 rats ha^-1 in shrubland) and similar to other New Zealand sites with low soil fertility. Further research will be required to elucidate the roles of food quality, habitat structure and predation in facilitating habitat selection in these species.     

Reproducing Bobcats to Cumberland Island, Georgia

Many felids are threatened by loss of habitat, lack of genetic diversity, and over-exploitation. The reintroduction of bobcats (Felis rufus) to Cumberland Island, Georgia provided an opportunity to reintroduce a mid-sized felid without the concern for species survival that is paramount with endangered species. We captured bobcats from the coastal plain region of Georgia, briefly held them in captivity, and released them on Cumberland Island. We describe and evaluate the protocols and techniques used to accomplish the reintroduction. Future reintroductions of felids should consider the problem of post-release dispersal, although our island was relatively isolated and inhibited dispersal. Also, any reintroduction effort should invest effort and resources into post-release monitoring of the population. Empirical knowledge about the effects of spatial distribution, genetics, population dynamics, especially mechanisms of population regulation, behavior, and environmental conditions on the viability of populations is critical to the conservation of endangered species. Future research of the bobcats on Cumberland Island will be able to address aspects of the population and genetic dynamics of a small, insular felid population.     

Habitat characteristics of Eurasian pine martens Martes martes in an insular Mediterranean environment

Eurasian pine martens are considered habitat specialists, associated primarily with mature stands of mesic mixed wood forest habitats, and avoid areas without overhead cover The species is found throughout the temperate and boreal regions of the continent but on the Mediterranean island of Minorca, introduced pine martens thrive in a competitor- and predator-free environment I test the prediction that because of evolved prey-capture and predator avoidance strategies Minorcan martens should select habitats most similar to temperate and northern parts of their range Scat index routes were used to quantify pine marten habitat selection Marten did not demonstrate any habitat type preferences although observed use of pine forests and coastal shrublands was slighly greater than expected Marten were indifferent to overhead cover whereas mesic sites and areas of tall high shrub density were favored Small mammal trap indices and preferred prey suggested that martens commonly used non-forested areas My results demonstrated that on Minorca pine martens were habitat generalists In the absence of predators open non-forested habitats were equally important to pine marten as were forested ones     

The plant communities and environmental gradients of Pitcairn Island: the significance of invasive species and the need for conservation management

Quantitative surveys of the vegetation of south-east Polynesian Islands are rarely undertaken owing to time and logistical restrictions; however they are fundamental in determining the conservation status of fragile island ecosystems. The aim of the research was to document quantitatively the vegetation of Pitcairn Island by investigating whether clearly definable plant communities existed on the island, and the underlying environmental gradients influencing these communities. Initially, 10 x 10 m quadrats were taken from all areas of the island, with environmental parameters recorded for each quadrat. The vegetation was then mapped from high altitude vantage points. Two-way indicator species analysis was used to identify distinct plant communities, and canonical correspondence analysis was used to determine the underlying environmental gradients. The vegetation consists of 14 plant communities: four coastal, six forest, two fernland and two scrub communities. Large areas are covered by non-native scrub vegetation, and by monospecific Syzygium jambos (rose-apple) plantations. Less than 30 % of the island is covered by native forest, and these areas are limited to remote valleys. Fernlands also cover large areas, including both eroding areas and ridge tops. Coastal vegetation comprises rock and cliff communities with limited strand vegetation. The major environmental gradient affecting the composition of the plant communities is altitude, but anthropogenic influences also have a large effect, owing to forest clearance and introduced species. The light environment is affected by the canopy species, and determines what ground flora can develop. Identification of distinct plant communities has allowed for a system of nature reserves to be suggested, which conserve all of these plant communities and a significant proportion of the threatened plant species.     

Genetic monogamy in two long‐lived New Zealand passerines

High rates of extra‐pair paternity (EPP) can be relatively common in passerines whereas low rates or absence of EPP are often associated with taxa that are long‐lived and exhibit obligatory paternal care. We examined EPP in an under‐represented category: passerine species with relatively long life spans (or low annual mortality rates). Specifically, we studied EPP in New Zealand saddlebacks Philesturnus carunculatus and robins Petroica australis, two species with unusually low annual mortality rates (6.5–11% and 10–20% respectively). No EPP (0%) was detected in saddlebacks (39 pairs, 202 offspring) and only one case of EPP (1.9%) was detected in robins (54 pairs, 198 offspring). Genetic monogamy in these passerine species supports the hypothesis that low annual mortality rates play an important role in explaining variation in rates of EPP across species.     

Multiscale Habitat Use and Selection in Cooperatively Breeding Micronesian Kingfishers

Abstract: Information about the interaction between behavior and landscape resources is key to directing conservation management for endangered species. We studied multi-scale occurrence, habitat use, and selection in a cooperatively breeding population of Micronesian kingfishers (Todiramphus cinnamominus) on the island of Pohnpei, Federated States of Micronesia. At the landscape level, point-transect surveys resulted in kingfisher detection frequencies that were higher than those reported in 1994, although they remained 15-40% lower than 1983 indices. Integration of spatially explicit vegetation information with survey results indicated that kingfisher detections were positively associated with the amount of wet forest and grass-urban vegetative cover, and they were negatively associated with agricultural forest, secondary vegetation, and upland forest cover types. We used radiotelemetry and remote sensing to evaluate habitat use by individual kingfishers at the home-range scale. A comparison of habitats in Micronesian kingfisher home ranges with those in randomly placed polygons illustrated that birds used more forested areas than were randomly available in the immediate surrounding area. Further, members of cooperatively breeding groups included more forest in their home ranges than birds in pair-breeding territories, and forested portions of study areas appeared to be saturated with territories. Together, these results suggested that forest habitats were limited for Micronesian kingfishers. Thus, protecting and managing forests is important for the restoration of Micronesian kingfishers to the island of Guam (United States Territory), where they are currently extirpated, as well as to maintaining kingfisher populations on the islands of Pohnpei and Palau. Results further indicated that limited forest resources may restrict dispersal opportunities and, therefore, play a role in delayed dispersal and cooperative behaviors in Micronesian kingfishers.     

Delayed plumage maturation increases overwinter survival in North Island robins

Many bird species show delayed plumage maturation (DPM), retaining sub-adult plumage until after their first breeding season. Most explanations assume that DPM increases fitness over the breeding season. However, unless birds undergo a full moult before breeding, DPM could also be an adaptation to increase survival over the previous winter. The winter adaptation hypothesis has never been tested owing to the difficulty of measuring overwinter survival. We experimentally tested this hypothesis in North Island robins (Petroica longipes) using a closed island population where we could accurately estimate survival. The experiment involved dyeing 41 juveniles to mimic adult males, and comparing their survival with 41 control juveniles treated with the same peroxide base minus the pigment. The population was monitored with a series of resighting surveys, and mark-recapture analysis used to estimate overwinter survival. Survival probability was estimated to be 10% for dyed birds versus 61% for control birds in 2001, and 29% for dyed birds versus 40% for control birds in the winter of 2002, supporting the winter adaptation hypothesis for DPM. Access to suitable habitat is the key factor limiting juvenile survival in this population, and the locations where dyed juveniles were sighted suggest that they were often excluded from suitable areas.     

Climatic preferences and frequent co-occurrence of boreal and temperate plants in Hokkaido Island,northern Japan

Effects of climatic factors on the forest plant distribution were examined in Hokkaido Island, northern Japan, where boreal and temperate plants frequently co-occur, and the relationship of floral make-up with phytogeography and life form was analysed. From the climatic preference of tree species consisting of mixed forest, the co-occurrence of boreal and temperate plants was found over a wider thermal range in regions of little rainfall. However, co-occurrence in a certain forest site was more notable in the understory than in the crown. A particular case of this involves southern evergreen shrubs and herbs in a boreal coniferous forest. Another is the presence of northern summergreen herbs in a temperate hardwood forest. It is speculated that the co-occurrence dates back at least to the late Quaternary, when a decrease in temperature associated with the glacial period forced understory plants to adapt their life form or leaf habit to snowcover and the light conditions of interior forests. Recent geographical evidence suggests that the island has been isolated from the southern island for at least 70000 years. Thus, the postglacial re-expansion of southern species does not appear to be due to immigration but rather due to their expansion from refuges on the island such as Oshima Peninsula for snow-tolerant species and Hidaka District for snow-intolerant species.     

Restoration of a tropical island: Cousine Island,Seychelles

Tropical island species and ecosystems are threatened worldwide as a result of increasing human pressure. Yet some of these islands also lend themselves to restoration, as they are physically defined units that can be given focused attention, as long as resources are available and clear conservation targets are set. Cousine Island, Seychelles, is a tropical island that has received such intensive restoration. From a highly degraded island in the 1960s, the island has now been restored to what is believed to be a semblance of the natural state. All alien vertebrates have been eradicated, as have 25 invasive alien plants. Cultivated plants are now confined to one small section of the island. Poaching of nesting marine turtles has been stopped, leading to an increase in turtle breeding numbers. The shearwater population has increased in size with poaching activities under control. The Sooty tern has also returned to the island to breed. The coastal plain has been restored with over 2,500 indigenous shrubs and trees, which have now grown into a forest carpet. There are strict quarantine procedures on the island, keeping it free of rats, mice, various alien invertebrates and potentially invasive alien plants. Three threatened Seychelles endemic land birds (Seychelles warbler, Seychelles magpie robin and Seychelles white-eye) have been introduced and are thriving, with these introductions contributing to both the magpie robin and the white-eye being downgraded from CR to EN (the warbler remains at VU). Ecotourism, and nature conservation for the local inhabitants, have been introduced in a way that does not reduce the improved compositional, structural and functional biodiversity of the island. The result of the restoration effort appears to be sustainable in the long term, although challenges still remain, especially with regards to adequate clean water and a non-polluting power supply on the island. Cousine is thus paving the way in the art and science of tropical island restoration as a legacy for future generations.     

Breeding of the Seychelles Magpie Robin Copsychus sechellarum and implications for its conservation

The total population of the Seychelles Magpie Robin Copsychus sechellarum declined from 38–41 birds in 12–13 territories in 1977–1978 to 17–21 birds in eight to nine territories in 1988–1990 and was entirely confined to Fregate Island (210 ha) in the Seychelles. After a successful cat eradication program in 1981–1982, recruitment improved, although the abandonment of agriculture had caused a reduction in the amount of feeding habitat and hence in the carrying capacity of the island. The population declined because of the failure of recruitment to compensate for the annual adult mortality. Foraging activity of the Magpie Robin was greater in high-quality territories (measured by soil invertebrates available), leading to increased reproductive success. Through supplementary feeding, five times as many recruits were produced. Of the 11.5 potential annual breeding recruits, 5.3 are required to compensate for adult mortality, and the other 6.2 recruits can be regarded as “surplus” contributing to an increase. Magpie Robins prefer to breed in rotten trees, which are a scarce resource. The greater the distance between the nest site and feeding areas, the less time was spent in incubation and nest guarding, resulting in greater egg loss. Because of lack of suitable areas for establishing territories, many young Magpie Robins became “floaters”. Nest disturbance, both by these floaters and by the introduced Indian Mynah Acridotheres tristis, had adverse effects on the breeding success of robins. A recovery plan has been designed to save the Magpie Robin. Territories have been improved for feeding (by tree planting) and for breeding (by providing nestboxes and reducing nest disturbances). Given the vulnerability of one small island, the presence of surplus birds (supported mainly by supplementary food) and the suitability of neighbouring Aride Island (68 ha), successful translocations to this island took place in 1992 and 1994. Given the presence now of 47 individuals on two islands, it is hoped that the species will pull back from the brink of extinction.