ABSTRACT
The chapter begins by explaining the evolutionary transitions from bacterium through to social insects, illustrating why multilevel selection explanations are indispensable. It then explains kin selection and group selection, before moving on to the varieties of multilevel selection. The multilevel conceptions of Lewontin and of Damuth and Heisler are explained and analyzed, as are the methods of contextual analysis. Lewontin’s research on the t allele in house mice is discussed in an extended fashion throughout the chapter to illustrate the limitations of these previous accounts. A new, three-part classification of multilevel selection models is proposed. They are divided into multilevel fitness models, multilevel force models, and multilevel trait models. These distinctions are explained and justified, and it is shown that Lewontin’s model of the evolution of the t allele is a multilevel trait model.
