tion of an affective sensation in one part of theWhy affection is thought to be entirely unextended. body were a matter of gradual training. A certain time elapses before the child can touch with the finger the precise point where it has been pricked.—The fact is indisputable; but all that can be concluded from it is that some tentative essays are required to co-ordinate the painful impressions on the skin, which has received the prick, with the impressions of the muscular sense which guides the movement of arm and hand. Our internal affections, like our external perceptions, are of different kinds. These kinds, like those of perception, are discontinuous, separated by intervals which are filled up in the course of education. But it does not at all follow that there is not, for each affection, an immediate localization of a certain kind, a local colour which is proper to it. We may go further: if the affection has not this local colour at once, it will never have it. For all that education can do is to associate with the actual affective sensation the idea of a certain potential perception of sight and touch, so that a definite affection may evoke the image of a visual or tactile impression, equally definite. There must be, therefore, in this affection itself, something which distinguishes it from other affections of the same kind, and permits of its reference to this or that potential datum of sight or touch rather than to any other. But is not this equivalent to saying that affection possesses, from the outset, a certain determination of extensity?
