RESEARCH ARTICLE Effects of outcrossing in fragmented populations of the primarily selfing forest herb Geum urbanum K. Vandepitte O. Honnay H. Jacquemyn I. Rolda ´n-Ruiz Received: 23 October 2009 / Accepted: 20 May 2010 / Published online: 3 June 2010 Ó Springer Science+Business Media B.V. 2010 Abstract In fragmented landscapes, small populations may be subjected to inbreeding or genetic drift. Gene flow is expected to alleviate the burden of deleterious mutations in such populations. The beneficial effects of outcrossing may, however, depend on life history characteristics such as the species’ breeding system. Frequent selfing is expected to purge (sub)lethal alleles and mitigate inbreeding depression, at least if the load of mildly dele- terious mutations has not accumulated through genetic drift in populations with a small effective size. Gene-inflow from distant source populations can cause outbreeding depression due to genomic incompatibilities. We tested these predictions using highly fragmented populations of the self-compatible forest herb Geum urbanum. Assessment of mating system parameters using microsatellite markers inferred high selfing rates (92.5%), confirming the predominantly self-fertilizing character of the study species. We conducted experimental pollinations with self and outcross pollen collected from populations at different distances from the target populations. There were no significant signs of inbreeding depression, even in very small target populations. Except for a minor negative effect on the germination rate for the long-distance crosses, we found no effects of out- breeding on fitness estimates. Keywords Self-pollinating Á Autogamy Á SSR Á Fragmentation Á Forest herb Á Inbreeding Á Outbreeding Á Population size K. Vandepitte (&) Á O. Honnay Á H. Jacquemyn Laboratory of Plant Ecology, Biology Department, Katholieke Universiteit Leuven (KULeuven), Kasteelpark Arenberg 31, 3001 Heverlee, Belgium e-mail: katrien.vandepitte@bio.kuleuven.be I. Rolda ´n-Ruiz Plant Sciences Unit, Growth and Development, Institute for Agricultural and Fisheries Research (ILVO), Caritasstraat 21, 9090 Melle, Belgium 123 Evol Ecol (2010) 24:1353–1364 DOI 10.1007/s10682-010-9395-0