Disintegration of the community?

Ricklefs' idea that you cannot predict geographical patterns of diversity from local diversity is true, but to my knowledge nobody claims such a thing. What is RIcklefs actually stating (but he dares not to do explicitly) is that species do not interact (the old Gleasonian idea). ritihs people obviously interact as little as possiblewith each other, but this is the exception, not the rule among the peoples of the world. In warm countries social interactions are tight, frequent and warm (although British of course believe this to be a sign of inferiority). And social prejudices strongly bias the interpretation of nature. Although dispersal at all scales (from the local dispersal to long range colinozation) plays a very important role, species obviously interact, as shown by both theory, practice and experiment. And, terribly, you cannot predict interactions from natural selections, since interactions are the cause and selection is the effect: this is stated very clearly in the famous chapeter of The Origin of Species where is said that competition is a conseuqnece not of abiotic factors, but of the coexistence of organism, from individuals to species.