In 1894 William Bateson described a class of discontinuous variation that he considered to be especially valuable for the study of evolution. This variation involved the repetition of a set of features typical of one member of a meristic series (e.g., a vertebra or a segment) in a new location in the series. Bateson referred to this process, whereby one body part is transformed into the likeness of another, as homeosis (from Greek homoios= same and -osis= condition or process, Bateson, 1894, pp. 84-85). Because phenomena such as homeosis could accomplish "at one step" a change similar to the differences observed between species, Bateson held that it was the discontinuity of variation itself, rather than the action of natural selection on continuous variation, that gave rise to the discontinuity of species (Bateson, 1894, pp. 568-570).
More than a century later, homeotic mutants play an important role in contemporary ideas of how genetics, development, and evolution intersect, and in large part have inspired the research program known as evolutionary developmental biology, or more popularly, "evo-devo" (e.g., Stern, 2000). The developmental genetics of Ed Lewis on homeotic mutants in Drosophila established the role homeotic genes play in specifying segmental identity during ontogeny (Lewis, 1951, 1964, 1978). Application of molecular techniques to these genes led first to the discovery of the homeobox, a conserved stretch of DNA sequence that codes for the DNA-binding portion of a large class of transcription factors (McGinnis, Levine, Hafen, Kuriowa, & Gehring, 1984; Scott & Weiner, 1984), and later revealed gene expression patterns consistent with mutant phenotypes (for review, see Akam, 1987; Harding, Wedeen, McGinnis, & Levine, 1985). These studies, those that followed in vertebrates (Awgulewitsch, Utset, Hart, McGinnis, & Ruddle, 1986; Duboule & Dollé, 1989; Graham, Paplopulu, & Krumlauf, 1989; for review see McGinnis, 1994), and comparative studies of other arthropods (e.g., Averof & Patel, 1997) provided critical evidence for the commonality of the genetics of developmental programs across the bilaterian animals, and led to efforts at broadly integrating the genetics of homeosis with patterns of body plan evolution in the fossil record (e.g., Jacobs, 1990). In essence, homeosis has revealed commonalities in the way animals are structured and suggested ways in which those body-plan organizations evolve.
Yet, even before the advent of molecular biology, homeotic phenomena fueled attempts to integrate genetics, development, and evolution. Indeed, during the late 1920s through the 1950s, homeotic mutations in Drosophila were at the center of efforts to achieve conceptual or theoretical integration - to create an evolutionarydevelopmental synthesis. However, those who researched homeotic mutants were unable to convince their peers that such phenomena ought to play a defining role in the emerging evolutionary synthesis. Why not?
This chapter consists of three sections. The first will consider the history and historiography of developmental biology and the evolutionary synthesis, as well as the nature of the split between embryology and genetics. The second will consider early efforts to characterize homeotic mutants as developmental, evolutionary, and genetic phenomena. Although we will focus on homeotic mutants of Drosophila, we should note that in the early twentieth century homeotic mutants were widely recognized in arthropods more generally, as well as in plants (Sattler, 1988). The third section of this chapter will consider the reception of evolutionary interpretations of homeotic mutants by the primary architects of the evolutionary synthesis. We will argue, as have others, that the history of research on homeotic mutants reveals that the split between embryology and genetics was not absolute. Instead, a research program of developmental and physiological genetics grew around the analysis of homeotic mutants. …