Comparative and experimental methodologies offer two complementary approaches to deducing both ontogenetic and phylogenetic mechanisms. Both kinds of studies analyze the effects of perturbations in the generation of these mechanisms, but in comparative studies the perturbation was introduced by the evolutionary process itself. At the methodological level, however, these approaches have quite separate traditions and tools. In large part, this is because the evolutionary experiment was completed long ago and many modifications have subsequently occurred. Furthermore, the unmodified “control” organism is typically missing – that is, has become extinct or subsequently modified independently of the “experimental” organism – and can only be inferred from analysis of multiple contemporary organisms. Thus, whereas the experimentalist may, for example, be satisfied to compare a single genetic mutant with its parental type, the comparative biologist frequently must deal with multiple species that differ in many loci, most of which have nothing to do with the phenotypic change of interest. Although this approach may seem indirect to an experimentalist, there really is no alternative when studying evolutionary processes.
Before the evolution of developmental processes can be inferred, it is necessary to place the species under study within its phylogenetic context, in order to establish the actual evolutionary sequence of change in ontogeny. This is analogous to the requirement that ontogenetic stages be temporally ordered before developmental mechanisms can be deduced. Of course, there are potential difficulties, if, for example, the developmental processes have evolved faster than speciation events have occurred or if species extinction has eliminated transitional forms. But, without phylogenetics, comparative biology becomes reduced to developmental taxonomy. For example, neotenic species display juvenilized phenotypes relative to their immediate ancestors.