Genetic constraints and community structure: Testing the concordance of trade-offs within and among species

Background/Question/Methods

The concept of trade-offs is central to many topics in ecology and evolutionary biology, from the evolution of life histories to mechanisms of species coexistence. Functional trade-offs observed among species are often assumed to reflect pervasive constraints on which phenotypes can be built with a finite amount of resources. If so, then among-species trade-offs should be concordant with trade-offs observed within species. On the other hand, the distribution of traits among species within a community may reflect historical contingencies and community assembly processes rather than within-species constraints. Here, we test whether a key functional trade-off between relative growth rate (RGR) and water-use efficiency (WUE) that has been previously described among winter annual species of the Sonoran Desert is apparent within four species that occupy different positions along the RGR-WUE trade-off spectrum. We sampled maternal seed families from multiple populations within each species and grew progeny in a greenhouse common garden for estimation of RGR and WUE. 

Results/Conclusions

We found significant genetic variation within species in RGR and WUE when these traits were examined singly. Univariate genetic variation was evident among families within populations and among populations that were sampled along gradients of precipitation and temperature. Species from the high-RGR end of the trade-off spectrum tended to harbor greater genetic variation among families than among populations, while the opposite was true for species from the high-WUE end of the trade-off spectrum. One high-WUE species displayed the expected negative correlation between RGR and WUE, both among populations and among families within populations. One high-RGR species had a positive correlation between RGR and WUE, suggesting genetic variation in resource acquisition rather than resource allocation. However, in most other instances a significant genetic correlation between RGR and WUE was not detectable within species. Taken together, our results are not consistent with a strong and pervasive genetic constraint on the evolution of a high-RGR/high-WUE phenotype within species.