Rules and exceptions in conservation genetics: genetic assessment of the endangered plant Cordylanthus palmatus and its implications for management planning Erica Fleishman a, *, Alan E. Launer a , Kathy Rehm Switky a,1 , Ulla Yandell b,2 , John Heywood c , Dennis D. Murphy a,3 a Center for Conservation Biology, Department of Biological Sciences, 371 Serra Mall, Stanford University, Stanford, CA 94305-5020, USA b Department of Biology/314, University of Nevada, Reno, NV 89557, USA c Biology Department, Southwest Missouri State University, 901 S. National Avenue, Spring®eld, MO 65804, USA Received 19 September 1999; received in revised form 14 May 2000; accepted 20 July 2000 Abstract We analyzed genetic variability among the four naturally-occurring populations of the endangered plant Cordylanthus palmatus to test whether a central tenet of conservation genetics Ð large populations are more genetically diverse than small populations Ð held true in our study system and to guide long-term conservation planning for the species. Genetic variability in C. palmatus was moderate at the species level but relatively low in several populations. About 2% of the measured genetic variation was attributable to variation between populations. Genetic variability in C. palmatus did not increase with population size. The two largest popu- lations were relatively invariate and genetically similar, and neither contained any unique alleles. # 2001 Elsevier Science Ltd. All rights reserved. Keywords: Allozymes; Plant conservation; Reintroduction; Spatial autocorrelation; Wetlands 1. Introduction Maintenance of genetic diversity is central to a fun- damental goal of conservation biology Ð preservation of the evolutionary potential of native taxa Ð and genetic variation is thought to increase the probability of population persistence. However, genetic considera- tions in conservation are often secondary to attempts to forestall population declines. Consequently, most con- servation eorts seek primarily to arrest habitat destruction, fragmentation, and degradation in order to minimize the detrimental eects of catastrophic events, environmental stochasticity, and, to a lesser extent, demographic stochasticity on species and populations (Lande, 1988; Holsinger and Gottlieb, 1991; Menges, 1991; Caro and Laurenson, 1994; Keith, 1998; Oostermei- jer et al., 1998). Moreover, genetic concerns often are over- shadowed by the struggle to amass resources sucient to conduct even basic ecological studies, monitoring, and adaptive management (Holsinger and Gottlieb, 1991; Avise, 1996). In addition, some species apparently can persist despite consistently small population sizes. Thus, although the importance of genetic stochasticity to population persistence is well-understood on theoretical grounds, measurement and management of genetic variability rarely assume key roles in conservation planning. In light of these constraints, use of genetic rules of thumb has practical justi®cation in many conservation situations. It is reasonable to hypothesize, for example, that genetic diversity is positively correlated with popula- tion size (Mee and Carroll, 1994), and that attempts to reintroduce populations are more likely to succeed if they draw upon several sources of founders or a genetically diverse population rather than a single invariant source 0006-3207/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved. PII: S0006-3207(00)00140-3 Biological Conservation 98 (2001) 45±53 www.elsevier.com/locate/biocon * Corresponding author. Tel.: +1-650-725-9914; fax: +1-650-723- 5920. E-mail address: e¯eish@leland.stanford.edu (E. Fleishman). 1 Present address: Community Impact, 3921 East Bayshore Road, Menlo Park, CA, USA. 2 Present address: 150 Gold Hill Avenue, Reno, NV 89506, USA. 3 Present address: Department of Biology/314, University of Nevada, Reno, NV 89557, USA.