'Nullius in Bergmann' or the pluralistic approach to ecogeographical rules: a reply to Watt et al. (2010)

摘要

Bergmann's rule is one of the best known empirical generalizations in biogeography and remains a topic of much interest and debate. Watt et al. claimed in 2010 that the only definition for the rule should be the one devised by Bergmann in the 19th-century. Based on direct translations from the original German manuscript, they concluded that tests of the rule should be restricted to interspecific studies of body size variation in endotherms. Furthermore, they suggested that Bergmann's heat conservation mechanism is an integral part of the rule and, hence, a simple falsificationist test of this mechanism might be enough to validate the rule. Here I advocate on a pluralistic approach to study ecogeographical rules, in general, and Bergmann's rule in particular. Our perceptions on the status and validity of laws and rules depend on the narrowness of the epistemological scope we adopt. Laws can have exceptions, do not necessarily have to be explanatory and may not be predictive. Also, we should differentiate between correlative and causal laws. Bergmann's rule is a correlative law, not a causal law, as Watt et al. implicitly assumed. It is a broad generalization with no inherent mechanism, and subject to the scrutiny of empirical investigation. Because of the ecological and evolutionary contingencies there will hardly be any law in ecology that is universally true. We have to consider each of these contingencies and study Bergmann's rule in a diversity of systems, organisms and levels of biological organization to gain further insight into the processes underlying the geographic variation in body size. On the basis of the empirical evidence to date, we cannot entirely dismiss a thermoregulatory mechanism to explain body size clines in both ectotherms and endotherms and support a food availability mechanism instead, as Watt et al. suggested. Even so, a unifying explanation for Bergmann's rule still remains elusive.