© 2007 The Authors DOI: 10.1111/j.1472-4642.2007.00376.x Journal compilation © 2007 Blackwell Publishing Ltd www.blackwellpublishing.com/ddi 725 Diversity and Distributions, (Diversity Distrib.) (2007) 13, 725–734 BIODIVERSITY RESEARCH ABSTRACT In order to assess the potential for molluscan conservation of a protected area con- sidered representative of regional megafauna, we sampled molluscs inside and out- side Lopé National Park in Gabon. In the northern part of Lopé National Park, 116 stations were surveyed and 71 species collected. Outside the park, 37 stations yielded 96 species, including 71 in Lastoursville, a small limestone area where molluscs are significantly more abundant than in other collecting sites. Lastoursville is among the richest sites known for molluscs in Africa. Overlap between sampling areas was limited, with 20.0% of the species found only in Lopé National Park, and 40.8% of the species found only outside. This suggests that Lopé National Park does not protect the whole molluscan diversity of central Gabon. Given the high levels of allopatric diversity of tropical land snails, conservation strategies cannot be the same for snails and for wide-ranging vertebrates. Protecting small areas with a high abundance and diversity of molluscs would be less expensive and as efficient for molluscan conser- vation as protecting large tracts of rainforest. Despite limited general knowledge of central African molluscs, robust estimates of site-specific diversity can be produced. Limestone areas harbour a remarkable biodiversity: sites such as Lastoursville would be ideal candidates for small protected areas dedicated to the conservation of land snails, and would complement the role of large protected areas. Keywords Mollusca, Gabon, umbrella species, rarity, limestone, allopatric diversity. INTRODUCTION Most species have a restricted range (Gaston, 1994): in this context, the clearing of tropical forests, which harbour a large proportion of the world’s species (WCMC, 2000), could lead to the extinction of many species. As most are still undescribed invertebrates (Hammond, 1995), their extinction would go unnoticed. It is therefore important to find strategies to protect this unknown biodiversity. One of these strategies relies on the umbrella species concept: protecting large animals implies protecting their habitat, and so protecting all the species, known or unknown, that live in this habitat, provided that they have less extensive spatial requirements than the umbrella species (Hunter, 1996). This strategy has been proven useful for species that rely on the same resources as the umbrella species (Martikainen et al., 1998; Suter et al., 2002), but its value in other cases is questionable. Considering mammals only, a study in Tanzania (Caro, 2003) showed that this concept is not always effective, as in certain circumstances background species are more abundant outside protected areas set up for umbrella species (background species are defined as species that live in the same geographical area as species that have been used to identify an area of conservation concern — Caro, 2003). A critical review of the literature to assess the usefulness of the umbrella species concept listed 18 studies that investigated the protection conferred by various umbrella taxa on various background taxa (Roberge & Angelstam, 2004). Most of these studies were performed in temperate regions, and none in tropical rainforests. Only six investigated the usefulness of this concept for the conservation of inver- tebrates, generally butterflies. The protection conferred was ineffective in three of the 18 studies, limited in six and effective in one, and a mixture of these in the other cases, depending on the taxa, the scale or the context. At a large scale, there is generally low congruence of species richness across taxa (Prendergast et al., 1993; Lombard, 1995; Kerr, 1997; Howard et al., 1998). The use of surrogate species to select areas for the conservation of poorly known taxa is thus not necessarily useful, and should be tested whenever possible (Caro & O’doherty, 1999; Simberloff, 1999; Fleishman et al., 2001). Moreover, most studies of the umbrella species concept are based on hypothetical reserves Muséum National d’Histoire Naturelle, Paris (France), Département Systématique et Evolution, USM 602, Case Postale No. 51, 57 rue Cuvier, 75231 Paris Cedex 05, France *Correspondence: Benoît Fontaine, Muséum National d’Histoire Naturelle, Paris (France), Département Systématique et Evolution, USM 602, Case Postale No. 51, 57 rue Cuvier, 75231 Paris Cedex 05, France. Tel.: 00 33 1 40 79 31 02; Fax: 00 33 1 40 79 57 71; E-mail: fontaine@mnhn.fr Blackwell Publishing Ltd Priority sites for conservation of land snails in Gabon: testing the umbrella species concept Benoît Fontaine*, Olivier Gargominy and Eike Neubert