Phylogenetic repulsion in the assembly of Floridian oak communities J. Cavender-Bares1,2,3, D.D. Ackerly4, D.A. Baum5 and F.A. Bazzaz2 Running head: Phylogenetic repulsion of Floridian oaks 1Corresponding author: Smithsonian Environmental Research Center 647 Contees Wharf Road Edgewater, MD 21037 2Harvard University Department of Organismic and Evolutionary Biology Cambridge MA 02138 3Contact info March 1 – Aug. 31, 2003: Centre d’Ecologie Fonctionelle et Evolutive – CNRS 1919 route de Mende 34293 Montpellier cedex 05 Tel 33(0)467613289 Fax 33(0)467412138 [email protected] 4Department of Biological Sciences Stanford University Stanford CA 94305 5Department of Botany University of Wisconsin Madison WI 53706 Keywords Plant community structure, null models, phylogenetic repulsion, independent contrasts, niche overlap, Quercus, conserved and convergent traits, soil moisture2Abstract Plant species that occur together in communities and experience similar environmental conditions are likely to share phenotypic traits due to the process of environmental filtering. At the same time, species that are too similar are unlikely to co-occur due to competitive exclusion. Communities that are dominated by multiple species of a single genus are challenging to explain because they appear to present an exception to the principle of competitive exclusion. We examined the phylogenetic structure of oak dominated forest communities in north central Florida in relation to ecological and physiological traits of the seventeen species of oaks that occur in this region. Oak species co-occurring within communities are more distantly related than expected by chance, and the most closely related species occur in the same communities less often than expected, providing evidence for phylogenetic repulsion. Within individual clades, species tend to show spatial overdispersion among plots and greater niche partitioning across a soil moisture gradient than expected. However, cumulative distributions of species in the two major clades (i.e., red oaks and white+live oaks), representing the deepest phylogenetic node, show greater niche overlap than expected. Hence communities are more likely to include members of both the red oak and the white+live oak clades than expected by chance and the distributions of the two clades across a soil moisture gradient are significantly more similar than expected. This pattern of phylogenetic repulsion within clades can be explained because traits important for habitat specialization show high evolutionary convergence. At the same time, we hypothesize that certain conserved traits permit coexistence of distantly related congeners.3Introduction A critical challenge in community ecology is to understand how the ecophysiological attributes of species influence community assembly and how these attributes, and hence community composition, change over evolutionary time scales. There are two central processes involved in the assembly of communities 1) the ecological filtering of species that can exist within a community based on the abiotic environment (Weiher and Keddy 1995; Weiher et al. 1998) and 2) limitations to long-term coexistence including competitive exclusion (Gause 1934; Elton 1946; MacArthur and Levins 1967; Chesson 1991; Leibold 1998). These two processes make opposite predictions about the phenotypic and phylogenetic similarity and dissimilarity of co-occurring species (Tofts and Silvertown 2000; Webb 2000). If closely related species share similar physiological limitations and exhibit evolutionary niche conservatism, the former process, ecological filtering, will tend to cause closely related species to co-occur; this pattern can be termed phylogenetic attraction or underdispersion. In accord with this prediction, several previous studies have shown that related species are more likely to co-occur than expected by chance (Tofts and Silvertown 2000; Webb 2000). On the other hand, competitive exclusion should limit the coexistence of closely related species that are phenotypically similar, leading to the opposite pattern of phylogenetic repulsion, with co-occurring species being phylogenetically overdispersed. This latter pattern has been hypothesized but has yet to be demonstrated using phylogenetic analysis. Communities that are dominated by multiple species of a single genus are challenging to explain because they appear to present an exception to the principle of competitive exclusion. Most studies of this problem have focused on niche and character differentiation of sympatric congeneric species(Wilson and Lee 1994; Price et al. 2000), in an effort to find ecological and evolutionary mechanisms that may prevent competitive exclusion (Armbruster 1985; Schluter and McPhail 1992; Schluter 1996; Radtkey et al. 1997; Adams and Rohlf 2000). It is also important to look beyond the genus designation, and consider phylogenetic relationships at higher resolution. Phylogenetic repulsion or attraction may occur at any level of phylogenetic resolution (Silvertown et al. 2001),4reflecting patterns of trait evolution and the relative importance of ecological filtering and competitive exclusion (Webb et al. 2002). Oak-dominated forests provide an important model system to address these questions in plant communities. In North Central Florida, seventeen species of oaks co-occur. The high diversity of congeneric species can be explained in part by habitat differentiation at the landscape level with species occurring in three major community types (hammock, sandhill and scrub), differentiated by contrasting fire regimes and/or soil moisture and soil fertility regimes (Cavender-Bares et al. submitted). However, at small spatial scales, up to six species of oaks can still occur within a single 0.10 ha plot. In this study, we investigated the phylogenetic structure of co-occurring oak species in these communities and examined the phenotypic traits and habitat features likely to give rise to this structure. We estimated the phylogeny of the co-occurring oaks using molecular data, and we used environmental, ecophysiological, and morphological measurements in the field and greenhouse (reported in Cavender-Bares and Holbrook 2001; Cavender-Bares et al. submitted) to identify key functional traits and habitat features. We set out to test the hypothesis that co-occurring congeneric species would show patterns of either phylogenetic repulsion (due to factors
View Full Document
Unlocking...