Research into the interrelationships between genetic and species diversity has increased dramatically in the last few decades, due to a growing recognition of the ecological consequences of genetic variation and rapid evolution (Hughes et al. 2008). It is now well established that increased genetic variation in foundational species may in turn led to increased species diversity in communities, especially for other trophic levels (Crutsinger et al. 2006; Johnson et al. 2006; Whitham et al. 2006). The effect of species diversity on genetic variation and composition within constituent species is less clear. Three basic predictions have been made for this relationship: (i) Genetic and species diversity could be positively correlated, owing to extrinsic factors like time since disturbance or environmental heterogeneity, (ii) Genetic and species diversity could be negatively correlated, because increased species diversity leads to reduced niche breadth per species and thus fewer opportunities for divergent genotypes to coexist, (iii) Genetic and species diversity could be positively correlated because different species exert unique selection pressures on populations, thus creating more opportunities for different genotypes to coexist, and finally (iv) species diversity may negatively correlate with genetic diversity if adding species to a community reduces the population size of constituent species and thus increases genetic drift (Vellend & Geber 2005). Of course, these mechanisms are not mutually exclusive and may act in parallel in a given system. In natural communities where both the history of community assembly and the population founding events are unknown, it will be difficult, likely impossible, to convincingly distinguish between these alternatives. Therefore, carefully designed experiments will be vital towards advancing our understanding in this area. In this issue, Nestmann et al. provide some of the first experimental demonstrations of one of these mechanisms, showing that species diversity affected genetic composition via stochastic evolution rather than selection.
Nestmann et al. (2011) took advantage of the ambitious Jena Biodiversity experiment, in which plots in a central European grassland were artificially created with 1, 2, 4, 8 or 16 species, drawn randomly from a pool of 60 species (Fig. 1). This species richness gradient was crossed with a functional diversity gradient (1, 2, 3 or 4 functional groups – grasses, small herbs, tall herbs and legumes). In 2002, they sowed seeds from 15 genetically and phenotypically distinct varieties of Lolium perenne (perennial ryegrass) into each of the 78 plots with constant densities among varieties. After 4 years, they took bulked tissue samples from up to 100 L. perenne from each of these plots and genotyped the populations with 22 SNPs, using next-generation sequencing technology to determine gene frequencies in the bulk samples without necessitating genotyping individual plants. They then compared these final gene frequencies to the expected frequencies of the initially sowed population (determined by genotyping a population with equal representation of each of the 15 varieties). The genetic distance between the initial and final population for each plot provided a measure of genetic change, which could then be related to the experimental manipulations of species and functional diversity.
Experimentally manipulated species richness had a strong impact on the evolutionary change experienced by each L. perenne population, with richer plots leading to greater genetic divergence in the population. Interestingly, this evolution did not occur in a predictable direction – while the L. perenne populations in more species rich plots showed a greater genetic distance to the initial population, they also showed greater distance to each other. On the other hand, the populations in the low richness treatments remained relatively clustered in their genetic composition. This random divergence of populations is a classic signature of genetic drift. Indeed, the species richness effect could be almost entirely attributed to the final size of the L. perenne population in a plot; higher species richness led to smaller L. perenne populations, which led to greater stochastic changes in genetic composition. As reproduction occurred primarily vegetatively, and there was no genetic signature of distance to the field margin or other L. perenne populations, the authors could be fairly certain that these results were not driven by gene flow from nonexperimental genotypes.
However, there were also signals of selective differences driven by plant community composition, specifically the presence or absence of particular functional groups. As these effects were evident even after controlling for population size, this suggests these varieties varied in their ability to coexist with certain types of plant competitors. Unfortunately, the authors do not have information on how phenotypic traits of the populations have changed. An interesting next step will be to quantify the selection pressures exerted by these different plant functional groups in various contexts and determine whether the changes in genetic composition seen in this study reflect adaptive evolution in response to plant community structure.
The notion that reductions in population size will lead to increased stochastic (relative to deterministic) evolution is central to our understanding of population genetics (Falconer & MacKay 1996). Nevertheless, recognizing that population size can provide a mechanistic link by which increases in species diversity can lead to decreases in genetic diversity may have important consequences for our attempts to manage and conserve biodiversity broadly. A primary goal of conservation science is to preserve species diversity by preventing extinctions. However, preserving genetic variation within populations is also important, both for reducing the threat of inbreeding depression but also to ensure that populations have sufficient evolutionary potential to respond to future environmental changes (Crandall et al. 2000; Mace & Purvis 2008). Nestmann et al.’s results suggest that these two goals may come into conflict. In situations where increasing the number of species in a community is likely to reduce the population size of each species, land managers and policymakers must balance the goals of preserving species diversity per se with maintaining sufficient genetic diversity within species. Additionally, increases in species diversity may not always be desirable. While exotic invasions are often cited as among the most pressing threats to global biodiversity, at least for plant invaders the evidence suggests that most invasions add to the species richness of invaded communities with few extinctions of native competitors, at least to date (Sax & Gaines 2008). However, the worst exotic invaders can reach very high densities in their introduced range, which almost inevitably results in population declines of native species. It is worth recognizing that these competitively driven population declines will likely lead to increased genetic drift, loss of genetic variation and thus loss of evolutionary potential to respond to future threats like climate change, or even to adapt to current stresses like the aforementioned invader.