Is it possible to admit that selection and heredity act equally upon that indefinable _something_ which is connected with the rudimentary intelligence and instincts of animals? With Darwin I unhesitatingly reply in the affirmative. With animals, as with man, all the individuals of the same species have not an equal amount of intelligence and do not invariably possess the same aptitudes; certain instincts, like certain forms, are capable of modification. Our domestic animals furnish a number of examples of these facts. The wild ancestors of our dogs were certainly not accustomed to point at game. When left to themselves and placed under new conditions of existence, animals sometimes change their manner of life entirely. Beavers, from being disturbed by hunters, have dispersed; they have now abandoned the construction of their lodges and dig out long burrows in the banks of rivers. The struggle for existence must have been favourable to the first discoverers of this new method of escaping from their persecutors, and natural selection, while preserving them and their descendants, has converted a sociable and constructive animal into a solitary and burrowing one. Up to this point it is evident that I agree in all that Darwin has said on the struggle for existence and natural selection. I disagree with him when he attributes to them the power of modifying organised beings indefinitely in a given direction, so that the direct descendants of one _species_ form _another species_ distinct from the first. IV. The fundamental cause of the disagreement arises evidently from the fact that Darwin had formed no clear conception of the sense which he attributed to the word _species_. I have been unable to find in any of his works a single precise statement on this point. The accusation is more severe from being brought with justice against an author who claims to have discovered the _origin of species_. More frequently Darwin seems to adhere to a purely morphological idea, which is also somewhat vague. He often opposes _species_ and _race_, which he also calls _variety_, but without ever stating clearly what he understands by one or the other. He endeavours, moreover, to bring them together as closely as possible, though occasionally recognising some of the points which separate them. “The species,” he says in drawing one of his conclusions, “must be treated as an artificial combination which is necessary for convenience.” His disciples have followed him faithfully in this direction, and those who use the most explicit language on this subject, join their master in declaring that a species is only a kind of conventional group similar to those which are used in classification. As for races, they are only species undergoing transmutation. Now from what he has already learnt, short though the study has been, the reader knows, I hope, to which view he should adhere, and understands to what confusions such a vague kind of theory must lead.