In Chapters 2 and 3, I set out to suggest that the theory that evolution has occurred is the explanans of the phenomenon of natural classification (the explanandum). We also saw that the hierarchical pattern of classification, apart from being the normal way in which classifications of objects are presented, is the result, at least historically, of an accepted method. That method, described in Chapter 6, is logical division. But since the time of Darwin and Wallace, the resulting hierarchy has been regarded as the natural pattern. Most methods of classifying organisms are therefore designed to produce a divergent hierarchy not just for convenience but because it is presumed to correspond to a real phenomenon in nature.

But classifications depend on characters possessed by the objects to be classified. Thus if the natural arrangement of organism is a divergent hierarchy, there must be an underlying divergent hierarchy of characters, and there should be something more to a character correctly uniting two species than a coincident resemblance. Behind the idea that one is dealing with the “same” character in two species, or higher taxa, is the concept of homology. Thus as natural classification is logically prior to phylogeny, homology is perhaps to be regarded as logically prior to natural classification.

It is not the case, however, that the concept of homology arose before that of natural classification in the history of systematics.