are below-ground are those organisms which are buried in the past, and which we can study only through their fossil remains.
Most naturalists not only believe that, if we could trace back the history of life, we should find each group bearing evidence of wider and wider relationship as we receded from the present time, but they also believe that we should ultimately find that every form of life is related to every other in such a way as to show that, in the remote past, they all met in a single starting-point—the common ancestor of all living things.
When we come to examine the evidence for this theory of the common origin of all life, we find that it is almost entirely general in its character. There is a nearly complete absence of specific and definite proof. We find an abundance of fossil forms, which we may regard as connecting links between one great group of animals and another; but even in the mo it favorable cases the attempt to follow the history of any particular species back for a considerable length of time soon ends in a total failure, for we lose track of its line of descent entirely, and can go on only by substituting, for the species with which we started, the genus, family, or more comprehensive group to which we have traced it.
Once in a while we find, in the later geological formations, a fossil animal which exhibits such affinity to several closely related recent species, that there is a strong presumption that it is the common ancestor from which they have descended. We have enough evidence to enable us to trace the horse and its allies through several geological periods with considerable accuracy, and to reach a form which is widely different from the horse, and which shows relationship to quite different groups of recent mammals. There are a few other cases where the evidence is equally abundant; but more usually it fails completely, and, although the fossils of the later formations show a very close relationship to their living allies, the resemblance is not exact enough to prove that the fossil forms are the direct ancestors of the recent ones, rather than more distant relations, connected by some unknown fossil form.
In place of the exact evidence which would be necessary to prove that the nearest fossil allies of recent species are in the direct line of descent, we have only the vague general evidence which is furnished by those fossils which unite in themselves the characteristics of widely separated families, or classes, or orders of animals. While the attempt to trace any particular species of bird and any given species of reptile to a common ancestor would be hopeless, we do find fossil organisms in whose structure certain general characteristics of the class Birds are united to certain general characteristics of the class Reptiles, in the way which we might expect if those animals are the descendants of true reptiles and the ancestors of the true birds of the present day. There is no proof that this actually is the case, and it is perfectly