case, besides the general inference that rudiments point us to a remote ancestry, we have direct paleontological evidence that there have been a whole series of extinct horse-like animals, that began low down in the geological strata with five toes (on the fore-feet, one being rudimentary), which afterwards became reduced to four and then to three; after which the two lateral toes began to become rudimentary, as we now see them in oxen, and later on still more So. Lastly, as we come nearer to recent times, we find fossils of the existing horse, with the lateral toes shortened up to the condition of splint-bones. Thus we have some half-dozen different genera of horse, all standing in a linear series in time as in structure, between the earliest representative with the typical number of five toes, and the existing very aberrant form with only one toe.

It is sometimes said that a striking corrobora

tion of a scientific theory is furnished when it enables us correctly to predict discoveries. Such a corroboration is afforded in this instance; for Professor Huxley, speaking in 1870, said, "If the expectation raised by the splints of the horses that, in some ancestor of the horses, these splints would be found to be complete digits, has been verified, we are furnished with very strong reasons for looking for a

no less complete verification that the three-toed plagiolophus-like avus' of of the horse must have had a five-toed 'atavus' at some earlier period. No such five-toed 'atavus,' however, has yet made its appearance." But since then the "atavus has made its appearance, if not with five complete toes, at least with four complete and one rudimentary; and any day we may hear that Professor Marsh has found in still

earlier strata a more primitive form with all five toes complete.

I have no space to go into the evidence of similar "missing links" which have been recently supplied by palæontological researches in the case of several other groups of animals; but their consideration seems to me quite to justify a more recent utterance of Professor Huxley, who, in 1878, wrote in the Encyclopædia Britannica: "On the evidence of palæontology, the evolution of many existing forms of animal life from their predecessors is no longer an hypothesis, but an historical fact; it is only the nature of the physiological factors to which that evolution is due which is still open to discussion."

BUT this allusion to fossils leads me to the next division of my subject—the argument from geology. It is not, however, necessary to say much on this head, for the simple reason that the whole body of geological evidence is for the most part of one kind, which although of a very massive, is of a very simple character. That is to say, apart from the increasingly numerous cases, such as the one just mentioned, which geology supplies of extinct "intermediate links between particular species now living, the great weight of the geological evidence consists

in the general fact, that of all the thousands of specific forms of life which palæontology reveals to us as having lived on this planet in times past, there is no instance of a highly organised form occurring low down in the geological series.1 On the contrary, there is the best evidence to show that since the first dawn of life in the occurrence of the simplest organisms, until the meridian splendour of life as now we see it, gradual advance from the general to the special -from the low to the high, from the few and simple to the many and complex-has been the law of organic nature. And of course it is needless to say that this is precisely the law to which the process of descent with adaptive modification would of necessity give rise.

1 Some of the lower vertebrata (Elasmobranch and Ganoid fishes) occur, indeed, in early strata (upper Silurian); but still far from the earliest in which some of the invertebrata are found. The general statement in the text applies chiefly to the more highly organised forms of the vertebrate series.