Genetics and extinction of island endemics: the importance of historical perspectives


J. Groombridge, Durrell Institute of Conservation and Ecology, Marlowe Building, University of Kent, Canterbury, CT2 7NS, UK.

Identifying the main drivers affecting extinction of endemic island species remains a key issue in conservation biology. Jamieson (2006) revisits the issue of identifying the relative importance of genetic versus demographic factors in species extinctions, and the extent to which these factors affect endemic island species.

Genetic studies from experimental laboratory populations and meta-analyses derived from natural populations provide convincing evidence that genetic factors such as inbreeding and loss of genetic diversity play an important role in the increased susceptibility to extinction of island endemics (Frankham, 1998, 2005; Frankham, Ballou & Briscoe, 2002; Spielman, Brook & Frankham, 2004). In particular, these studies show that island endemics have lower levels of genetic diversity than mainland populations and that higher levels of inbreeding in island populations lead to lower fitness, therefore making island species more susceptible to extinction. This genetic approach to extinction contrasts with ecological work by Blackburn et al. (2004, 2005), who show that extinction probability in island birds can be explained by the number of introduced mammalian predator species, and that their impact is greater for endemic species.

Jamieson (2006) questions why these two very different explanations for the greater extinction probability of island endemics appear to be mutually exclusive; one reason may be the focus on entirely different processes. Studies of genetic impacts on extinction examine stochastic processes inherent in all small populations, such as loss of genetic diversity and inbreeding, while ecological studies of extinction tend to focus on deterministic causes of decline, such as habitat loss and predation from introduced species. More informatively, these two disciplines appear to draw evidence to support their claims from different time periods. Ecological explanations such as those put forward by Blackburn et al. (2004, 2005) focus on historical extinction events, while genetic evidence has focused primarily on currently threatened taxa (Spielman et al., 2004). While a difference in temporal scale might appear to confuse how we interpret extinctions of island endemics, this problem may also be indicative of where future efforts need to be invested.

Clearly, the challenge is to identify the relative contributions of genetic and non-genetic processes to extinction of island endemics. One of a number of problems, however, is that most island endemics have evolved in the absence of predators, and therefore are predicted to be unusually susceptible to introduced predators. In addition, it is difficult to generalize about the time it takes for island endemics to become extinct. For example, evidence suggests that extinction from genetic factors may be relatively slow, in comparison with a more rapid decline driven by introduced mammalian predators (Jamieson, Wallis & Briskie, 2006). Jamieson (2006) points to the influence of population growth rates, stating that time to extinction may well depend on which deterministic conditions underpin a population's decline; a gradual loss of habitat might be more likely to provide conditions for genetic deterioration rather than rapid predation of a population by introduced mammals. Jamieson (2006) concludes that Blackburn et al. (2004) were right to attribute historical extinctions of island avifauna to predation by introduced mammals. He re-affirms that genetic factors probably act too slowly in predator-induced extinctions, and should therefore not receive the same priority as control of introduced predators when managing island endemics.

At the heart of the issue is the need to understand the interplay between drivers of extinction across different timescales, and in this context, the debate over genetics and extinction of island endemics is perhaps not yet fully resolved. Blackburn et al. (2004) demonstrate that not only can the number of introduced mammalian species explain extinction probability for historical extinctions of oceanic island birds, with a proportionately greater impact on island endemics, but that this relationship does not hold for currently threatened taxa, implying a change in principal threat through time. This picture is further complicated when considering how different drivers of extinction might operate non-additively. For example, habitat conversion also appears to explain avian extinction rates on islands (Didham, Ewers & Gemmell, 2005). Other studies have focused on New Zealand avifauna to illustrate temporal change in principal threat; Duncan & Blackburn (2004) identified different causes for prehistoric and historic extinctions of island endemics, reflecting a more rapid impact of introduced predators on recent extinctions. Today, much of New Zealand's endangered avifauna relies on management of introduced mammalian species. Consequently, it is not surprising that genetic factors, perceived by some to have a more ‘gradual’ effect relative to deterministic causes of decline, might not receive equal priority in recovery plans (Craig et al., 2000; Jamieson et al., 2006). Indeed, biologists from New Zealand can be forgiven for appearing unconcerned about genetic factors, in part because their island species are considered by some to be less sensitive to genetic impacts due to a long evolutionary history of isolation (Craig, 1994). The problem here lies in the difficulty of using data from a single species to generalize about the genetic impacts of inbreeding depression in island endemics. For example, some species restored from just a single breeding pair appear to respond well to recovery efforts, such as the New Zealand Black Robin and Mauritius kestrel (Arden & Lambert, 1997; Groombridge et al., 2000), but that may not be representative of all island endemics. Indeed, Jamieson (2006) notes how Frankham (1998) uses contributory genetic data from Hawaiian forest birds to support his claim that inbreeding contributes to extinction risk in island endemics, despite the non-threatened status of those Hawaiian taxa.

To clarify the debate on the extinction of island endemics, more historical approaches are needed that use natural systems for which deterministic drivers of extinction (habitat loss and introduced mammal species) are well-documented, but that also allow for inclusion of historical genetic data obtained from museum specimens of extinct species alongside those of endangered and non-endangered status. Genetic studies across multiple species from archipelagos that have lengthy and well-documented histories of avian extinctions, such as Hawaii, might allow simultaneous assessment of potential genetic and ecological drivers of extinction. In the meantime, New Zealand continues to lead the field in recovering island endemics, lengthening their impressive list of successfully restored species. However, long-term monitoring of these populations will prove invaluable in determining whether the common New Zealand ‘cure’ of controlling introduced mammals is sufficient to produce viable populations for the future.