Outplanting and differential source population success in Lupinus guadalupensis

Abstract

The choice of appropriate source populations is crucial for the success of outplanting attempts, but this choice is often based on assumptions regarding patterns of adaptation and distribution of genetic variability in natural populations. Although local adaptation is often assumed to exist, few data exist to support this model on smaller geographic scales, particularly in rare plant species. My study investigated the pattern of adaptation in populations of an annual, island endemic plant, Lupinus guadalupensis, on San Clemente Island, California. A reciprocal transplant experiment with three populations provided no evidence for local adaptation, but two source populations performed significantly better than the other at all sites. Desiccation and herbivory are the major factors causing mortality and reducing fruit production. The young rosettes are the most vulnerable life-cycle stage. Differences among natural population sites in vegetation community, dominant species, soil type, and parent material did not affect populations of L. guadalupensis; significant differences were not observed in either total fruit production or in the majority of fitness components. Success of outplanting into three introduction sites varied from the death of all plants before flowering or before fruits ripened to fruit production exceeding that at natural population sites. The results suggest that large populations are better sources of seeds for outplanting in L. guadalupensis and that outplanting is most successful in sites where plants are least subject to desiccation.