Science
Nov 27 1998: 1658-1659.Michael E. Soulé and L. Scott Mills
*[HN1]By the age of 16 every kid knows about the cultural taboos of incest--and that it is genetically contraindicated to marry your cousin, let alone a sibling. Avoidance of inbreeding is not limited to human beings, however [HN2], [HN3], [HN4]. Many plants (1) and animals (2) evaluate relatedness (3) and avoid matings with close relatives. In a small isolated population, inbreeding occurs because of a limited number of mates to choose from, not from preferential mating among kin [HN5], [HN6]. This can lead to increased homozygosity [HN7], [HN8], and, therefore, to homogeneity of the genes affecting the immune response (4) and to increased expression of recessive deleterious genes [HN9] that reduce survival, fertility, and physiological vigor (5).
Conservation geneticists have argued that in small populations the extinction probability should increase over time because these genetic effects magnify the extrinsic sources of jeopardy, including disease, inclement environmental conditions, and random demographic events. On page 1695 of this issue Westemeier et al. (6) [HN10] provide one of the first extensively documented examples of these complex interactions pushing a formerly large population toward extinction.
CREDIT: S. SUTER
Westemeier et al. (6) monitored greater prairie chickens [HN11], [HN12], [HN13], [HN14] in Illinois for 35 years, noting a steep population decline as habitat was lost, as the population became isolated during the 1970s, and as the Illinois population reached a demographic low of less than 50 birds by the early 1990s. At the same time, adjacent populations in Kansas, Minnesota, Missouri, and Nebraska have remained comparatively large and widespread [HN15], [HN16], [HN17], [HN18], [HN19]. Concurrent with the population decline, egg fertility and hatch success in Illinois prairie chickens also declined and was lower than in the neighboring large populations. The decline in egg hatching success in fully incubated clutches was correlated with a decrease in genetic variation, both for the Illinois birds when compared with the larger, nearby populations (7), and for the present population compared with historical samples collected before the demographic contraction (8).
Loss of genetic variation in small isolated populations is inevitable, as is an increase in the inbreeding coefficient [HN20] of surviving individuals. Nevertheless, critics have pointed out that the theoretical inexorability of inbreeding in small isolated populations does not necessarily translate into inbreeding depression [HN21],or an increase in the likelihood of extinction (9). For instance, such a small population is likely to be in dire straits already because of exposure to chance environmental events (droughts, storms, disease), or simply because of demographic accidents, including those that might skew the sex ratio. Although modeling results have demonstrated how genetic and nongenetic random fluctuations are mutually reinforcing (10)--emphasizing how genetics interact with extrinsic factors--a polarization of opinion favoring one or the other of these two factors has persisted for more than a decade.
In part the controversy has raged because of the difficulty in untangling a complex web of causation. Some of the inherent complexity arises from the interaction of several loops of positive feedback (see the figure above). For example, as a population becomes isolated as a result of habitat fragmentation, it becomes increasingly inbred, which can lead to inbreeding depression--manifest as decreased reproduction and survival--resulting in fewer reproductive adults in the next generation (smaller effective population size), thereby amplifying the consequences of a population "bouncing downward" due to bad weather or the randomness associated with year-to-year fluctuations in breeding success. Such mutual reinforcement can lead to a higher rate of genetic fixation and inbreeding in each succeeding generation, and to iterative declines in reproductive output and survival. The expected decline to oblivion for small populations has been called the "extinction vortex" (11) [HN22], [HN23].
The Westemeier et al. study (6) is one of the first examples that brings together several of these components, including the likely role of inbreeding depression. Although any large-scale field study must wrestle with the difficulty of excluding other hypotheses that could cause demographic fitness changes, several lines of evidence support the inference that in this case genetic effects exacerbated the decline of the Illinois prairie chickens, as opposed to being merely correlated with it. First, the decline in demographic rates and population size occurred despite aggressive efforts (and some success) in the 1960s and 1970s to control predators and increase the quality and quantity of habitat; this suggests that the extinction vortex was set in motion by the isolation of this population during the 1970s. Second, translocations to Illinois of prairie chickens from large populations in neighboring states since 1992 have increased hatching success, without any obvious concurrent changes in environmental variables (unfortunately, the authors do not have data from Illinois ¥ Illinois crosses to control for extrinsic factors affecting hatching success). The prospect that translocations have reduced inbreeding depression in egg hatching success is especially noteworthy because this single demographic rate surpasses all others in its effect on population growth rate in greater prairie chickens (12).
Although this study is unique in its comprehensive scope of factors and its temporal and spatial scales, it is consistent with a number of studies implying an important effect of inbreeding depression on population persistence. For example, inbreeding depression arising from small population size has been demonstrated in the wild for plants, vertebrates, and invertebrates (13). Furthermore, reduced genetic variation has been shown to reduce population growth (14) and increase probability of extinction (15).
Many questions remain, of course. Although we know that the costs of inbreeding range widely across species and time (16), we do not know much about how different demographic rates (for example, fecundity, egg survival, juvenile survival, adult survival) are affected by inbreeding for most species. We also do not know whether certain populations might become "adapted to inbreeding" as natural selection removes, or "purges," deleterious alleles (17).
Although it is becoming increasingly difficult to ignore the relevance of genetics to the extinction vortex, this understanding will not, by itself, address the main threat to biodiversity. This is because the most sinister actor in the extinction melodrama is neither genetics nor random population fluctuations; it is the loss of habitat and habitat quality, accompanied by overexploitation of biological resources, the increasing number of exotic species, pollution, and climate change. This said, researchers and managers can hardly justify a dichotomy that emphasizes one factor to the exclusion of all others (18). Small populations on the verge of extinction may well require aggressive acts of genetic intervention, including artificial gene flow and captive breeding.
References and Notes
M. E. Soulé is with The Wildlands Project [HN27], Post Office Box 2010, Hotchkiss, CO 81419, USA. E-mail: Soule@co.tds.net. Mills is in the Wildlife Biology Program [HN28], School of Forestry, University of Montana, Missoula, MT 59812, USA. E-mail: smills@forestry.umt.edu