have not even the faintest resemblance to the qualities of the things which present themselves to us through them, as I shall presently show. Only what really belongs to sensation must first be clearly distinguished from what is added to it by the intellect in perception. In the beginning this is not easy, because we are so accustomed to pass from the sensation at once to its cause, that the cause presents itself to us without our noticing the sensation apart from it, by which, as it were, the premisses are supplied to this conclusion drawn by the Understanding.
Thus touch and sight have each their own special advantages, to begin with; therefore they assist each other mutually. Sight needs no contact, nor even proximity ; its field is unbounded and extends to the stars. It is moreover sensitive to the most delicate degrees of light, shade, colour, and transparency ; so that it supplies the Understanding with a quantity of nicely defined data, out of which, by dint of practice, it becomes able to construct the shape, size, distance, and nature of bodies, and represents them at once perceptibly. On the other hand, touch certainly depends upon contact ; still its data are so varied and so trustworthy, that it is the most searching of all the senses. Even perception by sight may, in the last resort, be referred to touch ; nay, sight may be looked upon as an imperfect touch extending to a great distance, which uses the rays of light as long feelers ; and it is just because it is limited to those qualities which have light for their medium and is therefore one-sided, that it is so liable to deception ; whereas touch supplies the data for cognising size, shape, hardness, softness, roughness, temperature, &c. &c., quite immediately. In this it is assisted, partly by the shape and mobility of our arms, hands, and fingers, from whose position in feeling objects the Understanding derives its data for constructing bodies in Space, partly by
