than descent from a common ancestor, a, different from the common ancestor b of D, E, and F, and all may be the descendants of G, in the way shown by the heavy lines, or A may be the descendant of I; B of K; C of G, etc. If we were to attempt to indicate all the possible ways in which the six living species, A, B, C, D, E, and F, may be related to the fossil M, the diagram would become a confused mass of lines, and we have pointed out enough to show that, in a very simple case, where there are only two living genera and only six species, the attempt to follow them back only two stages to a common ancestor leads to so many possible systems of relationship that there is a very great chance against the truthfulness of any particular one, and we may fairly ask whether the attempt to express the relationships of animals in a tree-like classification can have any scientific value if the chances against its correctness are so very great. At first sight it may seem as if no good could be expected from this sort of speculation, and we may feel inclined to condemn the construction of phylogenetic trees as unscientific; but a little examination will show that all the lines in the diagram agree in one important particular, and trace the recent animals, A, B, C, D, E, and F, back to a remote common ancestor with a general resemblance to M. This, after all, is the essential thing, the gist of the whole matter, for the precise line of descent has no more scientific interest than the exact pedigree of each person would have to the anthropologist of our illustration. Such an exact pedigree would have a certain value as a bit of specific information, but the general evidence is of such a character that it is more logical to accept the conclusion than it is to reject it, and it is as truly scientific as the conclusion of our anthropologist.
We find that living things are related to each other in a peculiar way, which can be explained upon the assumption that they are the modified descendants of more ancient generalized forms, with wider relationships, and this assumption can be readily expressed in the form of a phylogenetic tree. We find, too, that so far as the higher groups of vertebrates, the mammals, reptiles, and birds, are concerned—groups which are of comparatively recent appearance, like the last races of immigrants in our imaginary case—the fossil forms which we meet with are such as our assumption would lead us to expect. The presumption is, therefore, very great that the genetic relations of living things may be expressed with general accuracy by a phylogenetic tree, although the chances of minute accuracy of detail in favor of any particular tree which is drawn up from paleontological evidence are very slight. This lack of minute accuracy can not be urged as an objection to all attempts at following out, in a general way, the lines of evolution of our present groups of animals, according to the best evidence which is attainable, and we must remember that only a very small part of this evidence is furnished by paleontology. If no fossils were known, the facts of comparative
