The Multifaceted Nature of Biodiversity Conservation: Reply to Leroux and Schmiegelow JOHN F. LAMOREUX, ∗ †† JOHN C. MORRISON,† TAYLOR H. RICKETTS,† DAVID M. OLSON,† ∗∗ ERIC DINERSTEIN,† MEGHAN W. MCKNIGHT,‡ AND HERMAN H. SHUGART ∗ ∗ Department of Environmental Sciences, University of Virginia, 291 McCormick Road, Charlottesville, Virginia 22904, U.S.A. †World Wildlife Fund–US, 1250 24th Street NW, Washington, D.C. 20037, U.S.A. ‡Curriculum in Ecology, Miller Hall, University of North Carolina, Chapel Hill, North Carolina 27599, U.S.A. Leroux and Schmiegelow raise several key issues about our paper (Lamoreux et al. 2006) regarding global tests of concordance for species richness and endemism and their implications for conservation. Although we differ with Leroux and Schmiegelow over the importance of endemism, we agree with enough of their comments that we are left to wonder why these were offered as a critique of our work. Leroux and Schmiegelow partition our global data set of terrestrial vertebrates into three categories of overall richness and find that, within each category, the richness of the groups shows little overlap. They conclude that “partitioning the data into richness categories seriously challenges the notion that a given vertebrate class can be used reliably as a surrogate for others.” Although Leroux and Schmiegelow provide this analy- sis as a challenge to our work, we agree with their result: global correlations, especially those of richness, are of lit- tle practical use for conservation. And we said as much in the paper: “Global conservation priorities based on richness alone will overlook many endemic species.” We also noted that “although global correlations are sugges- tive of concordant diversity patterns, the question most relevant to conservation decisions is whether a specific set of ecoregions selected for one measure will represent nontarget species (Balmford 1998; Howard et al. 1998; Moore et al. 2003).” In our paper we addressed this question by selecting ecoregions based on the number of endemics they con- tain and asking how many total species are also repre- ††email lamoreux@virginia.edu ∗∗ Current address: Irvine Ranch Land Reserve Trust, 320 Commerce Drive Suite 150, Irvine, California 92602-1300, U.S.A. Paper submitted September 25, 2006; revised manuscript accepted September 26, 2006. sented. Leroux and Schmiegelow’s approach is the oppo- site. They took all of the ecoregions that contain no terres- trial vertebrate endemics (280 total) and found that these hold 38.9% of all species. Although this analysis appears meant to emphasize the importance of these ecoregions for conservation, it accomplishes the reverse. For one thing selecting 280 ecoregions at random represents sig- nificantly more species (mean 73.1 ± 3.7% of all species). Selecting 280 ecoregions on the basis of endemism, as we did, represents 94.1% of all species (significantly bet- ter than random), and more than 6000 of the species are found nowhere else. More important, the species in the ecoregions that Leroux and Schmiegelow highlight, by definition, occur elsewhere; representing them in ecore- gions that also contain endemics leads to efficient con- servation planning (Margules & Pressey 2000). Leroux and Schmiegelow’s final point is the one we agree with most of all: biodiversity is multifaceted. Ecoregions lacking endemic species in these four taxa are nonetheless important for the innumerable unique species from other taxa that live in them and for other conservation objectives, such as maintaining large wilder- ness areas and ecosystem services. Although we focused on endemism, we clearly acknowledge in the paper that “methods for setting conservation priorities are complex and should consider not just the number of endemics or total species present, but also degree of threat (Dobson et al. 1997; Myers et al. 2000), population viability (Groves 2003), ecological and evolutionary processes (Olson & Dinerstein 1998; Groves 2003), and economic costs and 269 Conservation Biology Volume 21, No. 1, 269–270 C 2007 Society for Conservation Biology DOI: 10.1111/j.1523-1739.2006.00629.x