Mammals, Biodiversity of overharvesting all have contributed to the high historic rates of mammalian extinction One would think that assessing the rate of extinction in such a well-studied group as mammals would be relatively easy Yet, despite relatively good taxonomy and detailed study of the question, there remains considerable debate among mammalogists and conservationists about the number, the precise identity, and the timing of disappearance of those species that may have gone extinct in historical times In the 1990s, two efforts to categorize historical extinctions came up with approximately the same number of species going extinct: 85 by one estimate and 88 by another What is disconcerting, however, is that the two lists contain only 57 species in common, resulting in, overall, 116 different species being listed by one or another of these studies This suggests that, even among experts, there is some debate about what actually constitutes an extinction How much does this matter when assessing patterns of modem loss of mammalian species? On the one hand, documented extinction rates are so far over background rates that definitive listing will not change the implications for conservation of mammals On the other hand, such lists are playing an increasingly important role both in setting priorities for conservation (the process of keeping species off the list) and in public debate, and an accurate, defensible list is essential A clear ability to agree on what is meant by extinction is critical both to conservation planning and to conservation science While all differences in lists of extinct species need to be reconciled, much can be learned by analyzing why one set of authors excludes species that other authors list as extinct (Table 4) The reason(s) why authors disagree can tell us much about the source of such errors, or differences of opinion, and also help us focus our efforts more clearly on those species where changes in status affect conservation action Most lists examine the extinction of species over a particular time frame For instance, one might examine mammalian extinctions since the beginning of the age of exploration starting in 1500 The first question one must ask is, did the species in question really exist at any time during the past 500 years, or was extinction during this time inferred incorrectly either through data in the literature or from misplaced stratigraphy of sub-fossil material? In Table 5, clearly a plurality of the 76 species excluded were in this category There is no dispute whether or not these 31 species went extinct, rather just a question of whether their extinction predated the period under study While inclusion or exclusion of these species will change the calculation of rates of extinction, such changes have little practical applications to conservation Similarly, removing or adding a species from a list for reasons of taxonomic uncertainty or taxonomic revision does not manifestly change conservation status of a species Take for instance the quagga Distinct from other zebras, with only vestigial stripes, the quagga, Equus quagga, was initially thought to be a distinct species of zebra extirpated from the southern tip of Africa at the turn of the twentieth century Molecular analysis of museum specimens showed unambiguously that the quagga was a subspecies of the common zebra, Equus burchelli – one less extinct species, to be sure – but such revision does not change the way we manage the existing Table 705 A review of species not accepted as ‘‘Extinct’’ Species not known to exist in past 500 years Species extant as another taxa Species extinct as another taxa Taxonomy unresolved Species extant 31 18 3 21 Source: Data from McPhee RDEand Flemming C (1999) Requiem: Aeternam: The last five hundred years of mammalian extinctions In: MacPhee RDE (ed.)Extinctions in Near Time, pp 333–371 New York: Kluwer Academic/Plenum Table Geographic patterns of mammalian extinction Type Loss(%) Islands Caribbean Pacific Indian Ocean All other Total 37.5 21.6 8.0 6.8 73.9 Continental Australia Africa Eurasia Americas Total 19.3 4.5 1.1 1.1 26.0 Source: Data from McPhee RDE and Flemming C (1999) Requiem: Aeternam: The last five hundred years of mammalian extinctions In: MacPhee RDE (ed.) Extinctions in Near Time, pp 333–371 New York: Kluwer Academic/Plenum populations of E burchelli Similarly, if two extinct species are found to be synonymous, as in the case of Johnson’s hutia (Plagiodontia ipaneum) and Plagiodontia velozi (no common name), management plans remain unaffected When a species thought to be extinct is determined to still be extant, more likely than not the size and status of the extant population will be either totally unknown or known to be critically endangered Hence, transition from extinct to extant, unlike other categories, has important conservation implications In Table 4, such cases represent about one-quarter of those species listed Such a transition should be a red flag indicating that further study, and conservation action, are likely needed Patterns of Modern Mammalian Extinctions Although the rate of extinction is extremely higher than the expected background rate of extinction, the time frame over which historical extinctions have occurred is so short that it is often difficult to discern patterns in data collected on mammalian extinction Nonetheless, both taxonomic and geographical patterns emerge, both of which may be informative if we try to project future patterns of extinction in mammals Perhaps the most striking pattern seen when one examines data on historical extinctions of mammals relates to the geographic distribution of extinct species In the past 500 years, the great majority of extinctions have been on islands, and