According to the traditional Volterra-Gause viewpoint, the number of species in ecological communities (local richness) was thought to be controlled mainly by species interactions operating within the community itself. In order for this viewpoint to be correct, communities had to be close to equilibrium, species interactions had to be intense, and local richness had to be limited by the number of resources and physical factors. Here I argue that resource limitation and other local factors often fail to set upper limits on local richness. Revised theoretical models that add realism in the form of environmental complexity and non-equilibrium population densities of the constituent species do not predict such limits. In addition, critical evidence for such limits in real communities is often lacking. Local richness frequently correlates with the number of species in the surrounding region, and does not become independent of the latter in the richest regions as would be expected if the community became saturated. Similarly, communities in comparable environments but in regions supporting different numbers of species are not always convergent. In some cases, the effects of local environmental factors, which are known to have strong effects on local richness under some circumstances, can be severely constrained by history under other circumstances. Finally, evidence for classical interaction-imposed limits on local richness, such as density compensation and invasion resistance, does not consistently appear in many assemblages. Taken together, the evidence suggests that a more complete understanding of species richness in communities requires information on regional/historical as well as local processes. Hence, we need to broaden our perspective beyond the familiar Volterra-Gause equilibrium viewpoint and embrace community history as a relevant determinant of community structure and species composition.
