An ordinary watch is, as a rule, a good timekeeper, under a fixed, say a temperate meridian; but take it to the arctic circle, and it goes too fast; or to the torrid zone, and it goes too slow. But in a chronometer, delicate compensations are made, of a beautiful kind, so contrived as to counteract the thermal changes. Would it be logical to think of the more complex contrivance as devoid of design because it is self-acting?

I grant that there may not be absolute parity of reasoning in the human as compared with the natural instance; but I only desire the reasoning to apply to one essential point. If primarily the methods of nature, which by their rhythmic action automatically produce contrivances, living or otherwise, took their origin in mind, the contrivances so produced were so designed to be produced; and this is contrivance of an infinitely higher order than the self-compensating balance of the chronometer. It is contrivances brought about by arrangements that are infinitely complex, transcending all thought; and which include in their vast sweep, all time, all space, all matter, all motion, and all their relations, from the most inconceivably minute phenomenon to the most stupendous that ever has happened, or can happen, as a product of nature.

We may easily reduce this to a practical form by illustration. The arm or fore-limb of all mammals is constructed essentially on the same type. The forearm of the horse is most highly specialized. In the forearm there is, as in the human skeleton, the radius and the ulna. But the ulna has lost its function, and is fused with the radius. What we call the knee in the horse’s fore-limb is really the wrist. There are eight bones in the human wrist. The horse has only seven. Now, immediately succeeding the wrist bones in the human hand, are the five bones that form the palm; and these are followed by the five digits or fingers.

It appears, then, that, unlike the majority of mammal hands, the horse is peculiar. Instead of five bones corresponding to the palm bones, it has only one, which is called the ‘cannon bone;’ and this is followed by but one digit or finger with a huge nail. The horse then walks upon a single finger, viz. the third, on each fore foot and each hind foot. Why is this? and can we discover the history of this modification?

Examine the cannon bone of a horse with care; you will find fixed on either side of it two delicate bones, called ‘splints,’ with actually no mission in the economy of the extant horse.

Now go steadily back in geological time. There are three great geological epochs, the primary, the secondary, and the tertiary. The recent period called the quaternary is a sub-section of the last. Now in the quaternary and upper tertiary, fossil horses are found. They correspond with the horses we see. But in the middle of the tertiary epoch we come upon the horse, not only in Europe, but in India, and above all in America, with the splint bones, that we find in extant horses fixed to the cannon bone, as long as that bone itself, and provided with small spurious or useless hoofs, while the ulna can be traced for its whole length.

In the earliest miocene deposits the horse is found with these spurious or partially aborted toes, no longer useless, but having the full length of the middle toe; making the horse of that epoch a distinctly three-toed creature, and each toe operative; while the forearm is in a condition normal to mammals.

Yet another form, found in America, shows, not only the three distinct and useful toes, but upon the third toe a splint, suggesting a rudiment of the little or fifth finger; while in the oldest of the tertiary rocks has again been found a horse with the fourth finger complete. While finally, in the very earliest tertiary deposits is found a horse with a foot of four perfect digits, consisting of the second, the third, the fourth, and fifth fingers, with a rudiment or splint of the first finger distinctly visible.

Similar facts reveal themselves in relation to the hind feet and legs; and slow variations palpably affected the entire body of the successive types of horse; the teeth, for example, undergoing remarkable successive modifications: and whilst the Eohippus and the Orohippus, the last types of horse, discovered respectively in the oldest and later of the eocene beds, possess forty-four teeth, as in a large number of fossil and extant mammalian forms, the recent horse has the number reduced to forty, and their forms remarkably altered, as the final outcome of a succession of modifications.

We have here, then, a series of generic types; for the true relations of which to each other we are indebted to the insight of such workers as Huxley and Marsh and Lartet, and others. They do not afford us illustrations of the minute and scarcely observable modifications which the law of the origin of species involves; they present us rather with a series of family groups; but the relation between them is such as to leave the mind accustomed to biological investigations convinced, that, could we see all the forms that occurred between them, there could be no question as to the origin of these forms, from eohippus of the oldest eocene beds to equus of to-day. Unlike as in a general sense they are, they are progressive modifications, with higher and higher specializations, until, in the extant horse, the highest special modification is attained.