Rarity predisposes populations or entire species to extinction via proximate causes. Usually, population abundances and species geographic ranges are not independent entities (Lawton, 1993). Other things being equal, species that are most at risk from proximate causes are those with small geographic ranges because they will, on average, also be locally rare, even in areas where they occur (Lawton, 1994). This kind of double jeopardy can be strong when populations and ranges are artificially reduced and threatened by ultimate causes of extinction.
Reductions in range, via habitat destruction or extirpation by hunting, may eventually lead to a reduction in population density in surviving populations, even when protected. Another pattern that may influence extinction risk is a reduction in the average sizes of species’ ranges within comparable taxa, as one move from higher to lower latitudes. Stevens (1989) listed such reductions in range size towards the tropics for trees, molluscs, fish, amphibians, reptiles, and mammals. It seems intriguing that latitude alone explains over 45% of variation in population densities in a major compendium of data assembled by Currie and Fritz (1993). Average population densities increase linearly from the equator towards the poles in invertebrates, ectothermic vertebrates, mammals and birds.
At any one latitude, densities range over at least four orders of magnitude within these groups, but the overall trend is clearly consistent with a decline in range sizes towards the tropics (Lawton, 1994). Both the trends summarized above may make tropical taxa more prone to extinction from such human activities as habitat destruction, compared with equivalent temperate taxa. Pronounced persecution or habitat destruction can leave isolated populations of high conservation importance in marginal habitats. If overall population numbers decline because of falling birth rates or rising death rates, ranges would be expected to contract anyway, even in the absence of habitat destruction.
A recent review by Kunin and Gaston (1993) suggests that locally rare and geographically restricted species in many taxa, e.g. mosses, vascular plants, protozoa, insects and mammals, may have characteristics that differ from commoner relatives, including lower levels of self-incompatibility, and poorer dispersal abilities. Some of these traits can contribute to or exacerbate rarity whereas others promote the persistence of rare populations. According to Lawton (1994), extant patterns of species’ distributions— where populations persist after human impacts are not immutable echoes of former ranges. Nor do all species react in the same way to similar abuses. Current ranges reflect biology and history, with some taxa apparently more prone to local, regional and ultimately global extinction than others.
In general, populations of most large-bodied species fluctuate less on an annual basis than smaller-bodied taxa (Pimm, 1991) making large-bodied species less likely to fluctuate unexpectedly to extinction in the short term (Lawton, 1994). A consequence of such a generalization is that small, isolated populations of insects (i.e., endangered butterflies) are more likely to fluctuate to extinction than equally isolated populations of birds of similar average population size, even when afforded maximum protection.