ON THE DOCTRINE OF NATURAL KINDS. 437

plants." But he gives Horse, Animal, Sulphur, Phosphorus, Diamond, Gold, as examples of Kinds. Now are there absolute lines of demarcation, unfathomable chasms, between those classes? They greatly differ from each other, but not wholly. Is it not in reality a question of degree between their likenesses and those which connect Eose and Eubus ? Great degree, doubtless, but still only of degree. Is the criterion theoretically tenable ? OBJ. 4. But now let us turn to actualities. Are there in Nature classes clearly marked off from each other, classes to be sought for by us? "La methode naturelle," wrote Cuvier, " est 1'ideal auquel 1'histoire naturelle doit tendre ; car il est evident que si Ton y parvenait, Ton aurait 1'expression exacte et complete de la nature entiere." Such was the old view of a natural method, that it was nothing less than a reproduction of a certain orderly arrangement obtaining in the universe, waiting to be deciphered by man. When anomalies cropped up they were, if not too many and too weighty, relegated aside and labelled anomalies ; but if too overpowering, it was held that the right basis had not been chosen, or, to use a familiar phrase, a wrong key had been tried. But now there is a tendency to see that so-called anomalies are as legitimate facts as other characters. The old idea was a case of seeing double. There are no natural-made groups behind our groups. It is the endless seeking for this shadow in the stream that has so often misled us. I may quote some pertinent words of Prof. Newton of Cam- bridge : " The one merges insensibly in the other, as do the race, the species, the genus, and so on. There was a time, and that not long since, when each of these groups was looked upon as a concrete entity having an independent existence, and some men there are who still so regard them ; but whether that belief is destined to be perpetuated or restored may well be questioned. It would seem, rather, that each of these groups exists as a group but in the abstract." Prof. Asa Gray is equally emphatic : "The groups which we recognise and distinguish as Genera, Tribes, Orders, &c., are not always, and perhaps not generally, completely circum- scribed in nature, as we are obliged to assume them to be in our classifica- tion. This might be expected from the nature of the case. For the natural groups, of whatever grade, are not realities, but ideas. Their consideration involves questions, not of things between which absolute distinctions might be drawn, but of degrees of resemblance, which may be expected to present infinite gradations." Much more might be cited against the theory of " classes in Nature parted by impassable barriers," but I content myself with pointing out how entirely alien to this theory are the experiences of the constructors of natural methods. All who have worked in that field know that the individual is often indeterminable ; that the species cannot be fixed ; that the qualities of species do not remain constant ; that a regular progression cannot be obtained ;