of that kind of fruit, and compares its archetype with the two individual nuts? Assuredly he does not; but the sight of one of these nuts recalls to his memory the sensations which he received from the other, and his eyes, being modified after a certain manner, give information to the palate of the modification it is about to receive. Every general idea is purely intellectual; if the imagination meddles with it ever so little, the idea immediately becomes particular. If you endeavour to trace in your mind the image of a tree in general, you never attain to your end. In spite of all you can do, you will have to see it as great or little, bare or leafy, light or dark, and were you capable of seeing nothing in it but what is common to all trees, it would no longer be like a tree at all. Purely abstract beings are perceivable in the same manner, or are only conceivable by the help of language. The definition of a triangle alone gives you a true idea of it: the moment you imagine a triangle in your mind, it is some particular triangle and not another, and you cannot avoid giving it sensible lines and a coloured area. We must then make use of propositions and of language in order to form general ideas. For no sooner does the imagination cease to operate than the understanding proceeds only by the help of words. If then the first inventors of speech could give names only to ideas they already had, it follows that the first substantives could be nothing more than proper names.
But when our new grammarians, by means of which I have no conception, began to extend their ideas and generalise their terms, the ignorance of the inventors must have confined this method within very narrow limits; and, as they had at first gone too far in multiplying the names of individuals, from ignorance of their genus and species, they made afterwards too few of these, from not having considered beings in all their specific differences. It would indeed have needed more knowledge and experience than they could have, and more pains and inquiry than they would have bestowed, to carry these distinctions to their proper length. If, even to-day, we are continually discovering new species, which have hitherto escaped observation, let us reflect how many of them must have escaped men who judged things merely from their first appearance! It is superfluous to add that the primitive classes and the most general notions must
