Oecologia (2007) 154:315–326 DOI 10.1007/s00442-007-0831-y 123 PLANT ANIMAL INTERACTIONS Test of local adaptation to biotic interactions and soil abiotic conditions in the ant-tended Chamaecrista fasciculata (Fabaceae) Luis Abdala-Roberts · Robert J. Marquis Received: 26 September 2006 / Accepted: 19 July 2007 / Published online: 18 August 2007  Springer-Verlag 2007 Abstract Few previous studies have assessed the role of herbivores and the third trophic level in the evolution of local adaptation in plants. The overall objectives of this study were to determine (1) whether local adaptation is present in the ant-defended plant, Chamaecrista fascicu- lata, and (2) the contribution of ant-plant-herbivore interac- tions and soil source to such adaptation. We used three C. fasciculata populations and performed both a Weld and a greenhouse experiment. The Wrst involved reciprocally transplanting C. fasciculata seedlings from each popula- tion-source to each site, and subsequently applying one of three treatments to one-third of the seedlings of each popu- lation-source at each site: control, reduced ant density and reduced folivory. The greenhouse experiment involved reciprocal transplants of population-sources with soil sources to test for a soil-source eVect on Xower production and local adaptation to soil conditions. Field results showed that ant and herbivore treatments reduced ant density (increasing folivory) and herbivore damage relative to controls, respectively; however, these manipulations did not impact C. fasciculata reproduction or the likelihood of survival. In contrast, greenhouse results showed that soil source signiWcantly aVected Xower production. Overall, plants in both experiments, regardless of population-source, always had higher reproductive output at one speciWc site. Native populations did not outperform nonnative ones, causing us to reject the hypothesis of local adaptation. The absence of treatment eVects on plant reproduction and the likelihood of survival suggest a limited eVect of ants and folivores on C. fasciculata Wtness and local adaptation dur- ing the study year. Temporally inconsistent eVects of biotic forces across years, coupled with the young age of popula- tions, relative proximity of populations and possible counter eVects of seed predators may reduce the likelihood of local adaptation in the populations studied. Keywords Ant defense · Chamaecrista fasciculata · Herbivory · Local adaptation · Spatial variation Introduction Spatial variation in environmental factors is universal in ecological systems. In response to such variation, plant populations can become genetically Wxed for a phenotype at each location in which they occur, resulting in local adaptation (Van Tienderen 1990; Stearns 1992; Alpert and Simms 2002), or exhibit a range of phenotypes across sites against a homogeneous genetic background, i.e., pheno- typic plasticity (Bradshaw 1965; Lortie and Aarssen 1996; Alpert and Simms 2002). Together, local adaptation and phenotypic plasticity across sites represent end points in a continuum of potential evolutionary responses. In general, low gene Xow (Kingsolver et al. 2002; Sultan and Spencer 2002) and divergent natural selection across sites by biotic (e.g., herbivory; Sork et al. 1993) and abiotic factors (e.g., abiotic soil factors; McGraw and Chapin 1989) will Communicated by Judith Bronstein. L. Abdala-Roberts · R. J. Marquis Department of Biology, University of Missouri, St. Louis, One University Boulevard, St. Louis, MO 63121-4499, USA L. Abdala-Roberts (&) Departamento de Ecología, Campus de Ciencias Biológicas y Agropecuarias, Universidad Autónoma de Yucatán, Apartado Postal 4-116, Itzimná, 97000 Mérida, Yucatán, Mexico e-mail: turcotoman@hotmail.com