OF JOHNS HOPKINS UNIVERSITY.
WE will now examine our various sources of information, in order to see how far the evidence which they furnish can be used to establish phylogenies.
Comparative anatomy might not at first sight be expected to yield much of this kind of evidence, for the only animals which we can study thoroughly are those recent ones which have diverged very widely from their remote ancestors; and, while it is true that the study of the structure of living animals does not furnish direct evidence, the doctrine of homology supplies a means of sifting out, by general comparisons, what has been recently acquired from what is more deep seated, and of thus arranging animals in a series of groups of greater and greater extent and less and less contact. This classification of animals upon morphological grounds is essentially phylogenetic, for the difference between a system of converging lines and a system of more and more inclusive definitions is simply a difference in the manner of expression; nor can it be said that the one method assumes the disputed point, genetic relationship, any more than the other, for the naturalist who believes that classification is not simply a matter of convenience, but that there is one natural system, and that, according to this system, living things fall into a few great groups, each of which is characterized by certain general features, and that each of these groups is divided into smaller groups distinguished in a similar way, and these again into smaller groups, and so on, tacitly assumes that the natural system of classification or relationship is what we should expect it to be if the theory of descent with modification is true.
If there is a natural "systematic classification," it must be exactly the same as a phylogenetic tree, and the idea of descent is no more essential in the one case than it is in the other; they are simply different ways of expressing the same thing, the relationship of living things, and neither of them involves more than the other any particular interpretation of the word relationship; nor can it be said that, while the one method assumes that the larger trunks of the system have at some time been embodied in actual organisms, the second method allows us to believe that these groups are purely ideal, for, although this latter conception may have been defensible to some extent in the early days of morphological science, the progress of discovery has shown that even nowadays animals exist in which the characteristics of a branch