The fitness cost of generalization: present limitations and future possible solutions

One of the longest and liveliest debates in the evolutionary and ecological literature has centred on the existence and magnitude of constraints that can also be described by a proverb ‘jack‐of‐all‐trades is a master of none’. Often assumed, rarely tested, this proverb/assumption states that evolution of generalization necessarily entails a cost. The cost is expressed in terms of fitness loss elsewhere along an environmental gradient that leads to a genetic fitness trade‐off between a generalist and a specialist. Although there is a well‐developed body of knowledge that documents the cost of adaptation in general, the genetic fitness cost of generalization remains unclear. An empirical test of such cost is not a trivial task because it requires knowledge of a genotype’s fundamental ecological niche breadth to document the process of generalization. The estimation of genetic fitness correlation between environments, a commonly used method in the literature, has a limited explanatory power regarding the cost of generalization, and new approaches are needed to further clarify the existence as well as the nature/pattern of constraints in evolution of generalization and specialization. A new approach is proposed to examine experimentally the genetic fitness cost of generalization, which is based on statistical analysis of tolerance curve properties. The approach can be used to study natural populations of both unicellular and multicellular organisms. © 2007 The Linnean Society of London, Biological Journal of the Linnean Society , 2007, 90 , 583–590.