Endangered species have small and/or declining populations, so inbreeding and loss of genetic diversity are unavoidable in them. Since inbreeding reduces reproduction and survival rates, and loss of genetic diversity reduces the ability of populations to evolve to cope with environmental change, Frankel and Soulé [2] and others suggested that genetic factors would contribute to extinction risk in threatened species.

However, this view was challenged in the late 1980s and the contribution of genetic factors to the fate of endangered species was generally considered to be minor. Lande [3] suggested that demographic and environmental stochasticity and catastrophes would cause extinction before genetic deterioration became a serious threat to wild populations. A healthy controversy has persisted [1]. However, there is now a compelling body of both theoretical and empirical evidence indicating that genetic changes in small populations are intimately involved with their fate. Specifically:

• • Inbreeding causes extinctions in deliberately inbred captive populations.
• • Inbreeding has contributed to extinctions in some natural populations and there is circumstantial evidence to implicate it in many other cases.
• • Computer projections based on real life histories, including demographic, environmental, and catastrophic factors, indicate that inbreeding will cause elevated extinction risks in realistic situations faced by natural populations.
• • Many surviving populations have now been shown to be genetically compromised (reduced genetic diversity and inbred).
• • Loss of genetic diversity increases the susceptibility of populations to extinction.

Inbreeding has been known to reduce reproduction and survival (inbreeding depression) since Darwin's classic work [4]. For example, inbred individuals showed higher juvenile mortality than outbred individuals in 41 of 44 captive mammal populations studied by Ralls and Ballou [5]. On average, brother-sister mating resulted in a 33% reduction in juvenile survival. By extrapolation, it was anticipated that inbreeding would increase the risk of extinction in wild populations.

There is now clear evidence that inbreeding adversely affects most wild populations. Crnokrak and Roff [6] reviewed 157 valid data sets, including 34 species, for inbreeding depression in natural situations. In 141 cases (90%) inbred individuals had poorer attributes than comparable outbreds (i.e. they showed inbreeding depression), two were equal and only 14 were in the opposite direction. Results were very similar across birds, mammals, poikilotherms and plants. Further, significant inbreeding depression has been reported in at least another 15 taxa [1].

3.1 Relationship between inbreeding and extinction

3.2 Do taxonomic groups differ in susceptibility to inbreeding depression?

Inbreeding and loss of genetic diversity has been shown to increase the risk of extinction for two populations in nature. Inbreeding was a significant predictor of extinction risk for butterfly populations in Finland after the effects of all other ecological and demographic variables had been removed [14]. Further, experimental populations of the Clarkia pulchella plant founded with a low level of genetic diversity (and high inbreeding) exhibited 75% extinction rates over three generations in the wild, while populations with low inbreeding showed only a 21% extinction rate [15]. However, it was not clear whether these were general results, or exceptions.

Computer projections incorporating factual life history information are often used to assess the combined impact of all deterministic and stochastic factors on the probability of extinction of populations. Mills and Smouse [16] using computer simulations, found that inbreeding generally increases the risk of extinction, especially in species with low reproductive rates. These simulations encompassed only a 20 year time frame and they were criticised for not accounting for purging of deleterious alleles [17].

Brook et al. [18] conducted computer projections for 20 outbreeding bird, mammal, and invertebrate life cycles that allowed for the effects of purging. Median times to extinction were on average reduced by 24–31% when inbreeding depression of 3.14 lethal equivalents was applied to juvenile survival, compared to cases where inbreeding depression was omitted. These results underestimate the true impact of inbreeding depression, as it is approximately 12 lethal equivalents for populations in nature ([19], Frankham et al. unpublished data). A related computer projection for the rare European plant Gentiana pneumonanthe yielded similar conclusions [20]. These computer projections indicate that the results of Saccheri et al. [14] and Newman and Pilson [15] are not exceptions, but are likely to apply to the majority of species.

Declines in population size or extinction in the wild have been attributed, at least in part, to inbreeding in many populations including bighorn sheep, Florida panthers, Isle Royale gray wolves, greater prairie chickens, heath hens, middle spotted woodpeckers, adders, and many island species [1]. Further, inbreeding colonial spiders have a higher rate of colony extinction than non-inbreeding species.

6.1 Extinction proneness of islands populations

Lande [3] suggested that species would often be driven to extinction by demographic factors before genetic factors had time to impact, and many other authors have repeated this refrain. While Lande [17] has subsequently changed his views on the contribution of genetic factors to extinctions, this is due to his championing of ‘mutational meltdown’ and not due to a retraction of his 1988 views.

If the Lande scenario is common then threatened species should show little difference in genetic diversity, compared to related non-endangered species. The majority of threatened species do not fit the Lande scenario, as most have reduced genetic diversity compared to related threatened species [22–24]. Further, the magnitude of differences is such that threatened species are likely to be suffering serious reductions in fitness, as proportionate loss of genetic diversity estimates the inbreeding coefficient. Genetic diversity has been shown to be related to fitness, as expected from the relationship between genetic diversity and inbreeding in random-mating populations [25]. Consequently, most threatened species are likely to have both reduced reproductive fitness due to inbreeding depression and reduced evolutionary potential.

Natural populations face continuous assaults from environmental changes including new diseases, pests, parasites, competitors and predators, pollution, climatic cycles such as the El Niño–La Niña cycles, and human-induced global climate change [1]. Species must evolve to cope with these new conditions or face extinction. To evolve, species require genetic diversity. Naturally outbreeding species with large populations normally possess large stores of genetic diversity that confer differences among individuals in their responses to such environmental changes [1].

Small populations typically have lower levels of genetic diversity than large populations [1]. There are compelling theoretical predictions that loss of genetic diversity will reduce the ability of populations to evolve in response to environmental change, and experimental evidence validates these predictions [1]. Consequently, we expect a relationship between loss of genetic diversity and extinction rate due to environmental change. However, there are only a few examples where extinctions of natural populations can be directly attributed to lack of genetic variation, as described below.

8.1 Relationship between loss of genetic diversity at self-incompatibility loci and extinction in plants

8.2 Relationship between loss of genetic diversity and susceptibility to disease, pests and parasites

The Lande [3] scenario has failed numerous tests so it must be rejected for the majority of species [1]. What assumptions were made in the Lande scenario that are incorrect? Lande [3] did not present an explicit quantitative model that can be addressed point by point in a quantitative manner. However, it seems from his work and other papers around that time that four factors were probably involved, the ratio of effective to census sizes (N_e/N), the extent of interactions among stochastic factors, the extent of inbreeding depression in the wild, and the effectiveness of purging.

Genetic impacts depend on the effective population size (N_e), so the ratio of effective to census size is critical in determining genetic impacts. Around the time of Lande's paper it was typical to talk of N_e/N ratios of 0.25–0.5 [35]. Subsequently, N_e/N ratios in unmanaged populations have been found to average approximately 0.1 [36], so genetic factors impact sooner than Lande [3] would have expected.

Fluctuations in population size and sex-ratio and variation in family size all occur due to demographic and environmental stochasticity and catastrophes and result in reduced N_e/N ratios. Consequently, there are interactions between stochastic factors that increase genetic impacts on population persistence [37,38].

It was common in the late 1980s and early 1990s for people to be sceptical about the extent of inbreeding depression in the wild. Considerable data now exists and points to much higher levels of inbreeding depression than found in captivity [1,6]. For example, Ralls and Ballou [12] found a median of 3.14 diploid lethal equivalents for captive mammals, but total inbreeding depression across the life cycle in the wild is approximately 12 lethal equivalents ([19], Frankham, O'Grady, Brook, Ballou and Tonkyn, unpublished data).

The final factor leading to an underestimation of the impact of inbreeding on population viability is the effectiveness of purging. At the time of Lande's paper, purging was considered to be effective in markedly reducing inbreeding depression. Subsequent modelling and empirical work indicates that purging effects are typically relatively small [1,39,40].

All the above points lead to greater impacts of inbreeding depression on population viability than would have been expected in 1988.

Recovery programs may not be successful if genetic factors are ignored. For example, the Illinois population of greater prairie chickens declined from millions to only 200 in 1962, and failed to recover following habitat restoration [41]. It showed clear evidence of inbreeding depression (reduced fertility and hatchability). However, when inbreeding effects were removed by crossing to unrelated birds from other states, the population recovered its fertility and hatchability and grew in numbers. In the case of the koala in southeastern Australia, reintroductions using a small island population with only 2–3 founders have resulted in a substantial reduction in genetic diversity, to a rise in inbreeding, to a decrease in sperm quality, to a marked increase in testicular aplasia [42,43].

The effects of loss of self-incompatibility alleles on population fitness in plants are likely to be a major factor in species persistence, but will not be addressed unless genetic factors are recognised [27–29].

Acute reductions of population fitness occur when diploid and tetraploid populations of a species are introduced into proximity with each other [29]. Sterile triploids have resulted, with consequently reproductive wastage.

Genetic considerations are of particular importance in management of fragmented populations. Small fragmented populations with limited gene flow will lose genetic diversity and become inbred and have elevated extinction risks [1,14,44]. Adequately genetic management of fragmented populations is rare, and is one of the greatest unaddressed issues in conservation biology.

Overall, there is little effective genetic management of wild populations of threatened species, but a substantial need for it [1]. By contrast, genetic management of captive populations is widely practiced and generally well done.

Inbreeding and loss of genetic diversity are of conservation concern as they increase the risk of extinction. Inbreeding increases the risk of extinction in captive populations, and there is now strong evidence that it is one of the factors causing extinctions of wild populations. Loss of genetic diversity reduces the ability of species to evolve to cope with environmental change. Inappropriate management and allocation of resources is likely to result if genetic factors are ignored in management of threatened species.