Mammalian diversity in Neotropical lowland rainforests : a preliminary assessment. Bulletin of the AMNH ; no. 230

Supplemental Materials
Journal Title
Journal ISSN
Volume Title
[New York] : American Museum of Natural History
"Information about the magnitude and geographic distribution of mammalian diversity in Neotropical lowland rainforests is important for evaluating research and conservation priorities in Central and South America. Although relevant inventory data are rapidly accumulating in the literature, real site-to-site diversity differences are hard to identify because many confounding factors can affect the size and composition of faunal lists. Herein we assess the available information about Neotropical rainforest mammal diversity and suggest guidelines for future work by reviewing inventory methods, documenting and discussing faunal lists from ten localities, and summarizing geographic range data to predict diversity patterns that can be tested by field and museum research. All inventory methods are biased because each is suitable for collecting or observing only a fraction of the morphologically and behaviorally diverse mammalian fauna that inhabits Neotropical rainforests. Hence, many methods must be used in combination to census whole communities. Although no combination of methods can be guaranteed to produce complete inventories, the omission or nonintensive application of any of several essential methods probably guarantees incomplete results. We recommend nine methods that, used intensively and in combination, should maximize the efficiency of future inventory fieldwork. Ten rainforest mammal inventories selected as exemplars illustrate several common problems: sampling effort is highly variable from study to study, species accumulation curves are not asymptotic for any fauna, essential field methods were omitted in every case, and some localities were partially defaunated by hunters prior to inventory. Meaningful diversity comparisons are therefore impossible without a major investment in additional fieldwork at each site. Geographic range data provide an essential alternative source of diversity estimates. Comparisons of inventory results with geographic expectations (diversity predictions based on range data) suggest that all existing inventories are incomplete, that the degree of incompleteness is inversely correlated with inventory duration, and that special methods are required to add elusive species to faunal lists. The range data at hand also suggest several geographic patterns that should be tested with carefully focussed fieldwork. (1) Mammalian diversity in Amazonia is probably greatest in the western subregion (between the Rio Negro and the Rio Madeira, where over 200 species might be sympatric at some localities), least in the Guiana subregion (east of the Negro and north of the Amazon), and intermediate in southeastern Amazonia (east of the Madeira and south of the Amazon). (2) Geographic variation in Amazonian diversity chiefly involves marsupials, bats, primates, and rodents; by contrast, xenarthran, carnivore, and ungulate faunas are remarkably uniform across the entire region. (3) In Central American rainforests, a conspicuous and apparently monotonic diversity gradient extends from eastern Panama (where mammalian diversity is within the range of Amazonian values) to southern Mexico (where mammalian diversity may be less than anywhere else on the rainforested Neotropical mainland). Mammalian diversity in coastal Venezuelan and southeastern Brazilian rainforests is difficult to assess with existing literature and collection resources, but neither region is likely to be as diverse as Amazonia. Despite a few dissenting voices, the literature of New World mammalogy provides compelling evidence that mammalian diversity, as measured by sympatric species richness, is greatest in lowland tropical rainforests and decreases along gradients of increasing latitude, elevation, and aridity. Thus, the mammalian faunas of western Amazonia are the most diverse of any in the Americas and perhaps in the world. We briefly discuss the generality and causes of observed diversity patterns in terms of contemporary ecology and historical scenarios. Significant advances in understanding mammalian diversity patterns in Neotropical rainforests will require systematic revisions of many problematic genera and an aggressive program to inventory poorly sampled areas while opportunities to do so yet remain"--P. 3.
115 p. : ill. (some col.), maps ; 26 cm.
Includes bibliographical references (p. 71-86).