The impact of taxonomic bias when comparing past and present species diversity

It is becoming increasingly important to understand how present diversity patterns compare with past ones, in order to understand the extent of change that present faunas exhibit with respect to past baselines for such parameters as extinction rate and magnitude, ecological structure, and ecosystem function. However, these comparisons have been difficult to quantify because the modern and paleontological records are inherently different. This study examines how those differences affect comparisons of fossil and modern mammalian species diversity in the United States and suggests how the data can be treated to minimize their biases. I first compare extant mammalian species diversity to a paleo-baseline constructed from fossils covering the past 30 million years. Species-area relationships show that, contrary to expectations, today's mammalian diversity appears to have increased since the Holocene (11,500 to-500 years ago). This bump in diversity is the result of an increase in small mammal species in the modern dataset, in particular those that are the most difficult to identify and diagnose in the fossil record (e.g., Geomyidae and Heteromyidae). This increase results from neontological classifications of small mammal species that employ methodologies and characters (notably soft-tissue and molecular information) that cannot be used with fossils. One way to correct for these differences would be to reevaluate neontological species using the same morphological characters and species concept commonly used by paleontologists. (c) 2012 Elsevier B.V. All rights reserved.