Background/Question/Methods
            Patterns of precipitation are likely to change significantly in the coming century, with important consequences for plant communities.  Both experimental and correlative studies offer useful predictions about how communities will respond, but little research has addressed the degree of concordance between these two approaches.  Here, we synthesize results from four experimental studies of water addition and contrast the results with a correlative analysis of community changes across a large natural gradient in precipitation.  We use plant functional traits as a basis for comparison among studies and sites, and because they provide a mechanistic basis for understanding community changes.
Results/Conclusions
             Experimental results suggested that increased precipitation should cause communities to shift towards species with smaller seed sizes and higher leaf N concentrations.  In contrast, the natural gradient analysis showed increasing seed size and decreasing leaf N with higher precipitation.  We suggest that this disagreement is due in part to the limited duration and spatial scale of experimental manipulations, which reduces the opportunities for community turnover to occur.  These results suggest that responses in natural communities to climate change are likely to be complex, and could involve transient dynamics that do not reflect the long-term shifts in community composition.