A recent paper in Science (Adler et al. 2011, “Productivity is a Poor Predictor of Plant Species Richness,” link) takes a new look at the long-accepted (or at least long-discussed) relationship between plant species richness and productivity. The authors conducted a huge number of small-scale experiments (covering nearly every continent, some more thoroughly than others) and found very little support for a relationship between species richness and productivity within each site.
To me, this finding sparks more questions than it answers. Why have previous studies often found positive or hump-shaped relationships between SR and productivity? Is it an issue of sampling method? Scale? In particular, why do we consistently see positive relationships when we experimentally manipulate species richness à la Tilman et al., but not in some observational experiments? I think we need to re-evaluate the origins of the theory of SR-P relationships and come up with clear hypotheses about when we would and would not expect to see these relationships. For example, one of the mechanisms of positive SR-P relationships is thought to be coexistence mediated by the presence of predators. In this case, we might not expect to see SR-P relationships in highly invaded communities (since exotic species often lack natural predators).
In any case, my guess is that this matter has not been laid to rest. I look forward to seeing the rest of the results from this research group!