Thus much, then, for the rate of molecular vibration which goes on in nerve-centres. But the rate of such vibration which goes on in sensory and motor nerves may be very much more rapid. For while a nerve-centre is only able to _originate_ a vibration at the rate of about nine beats per second, a motor-nerve, as we have already seen, is able to _transmit_ a vibration of at least 1,000 beats per second; and a sensory nerve which at the surface of its expansion is able to respond differently to differences of musical pitch, of temperature, and even of colour, is probably able to vibrate very much more rapidly even than this. We are not, indeed, entitled to conclude that the nerves of special sense vibrate in actual unison, or synchronize, with these external sources of stimulation; but we are, I think, bound to conclude that they must vibrate in some numerical proportion to them (else we should not perceive objective differences in sound, temperature, or colour); and even this implies that they are probably able to vibrate at some enormous rate. With further reference to these molecular movements in sensory nerves, the following important observation has been made--viz. that there is a constant ratio between the amount of agitation produced in a sensory nerve, and the intensity of the corresponding sensation. This ratio is not a direct one. As Fechner states it, 'Sensation varies, not as the stimulus, but as the logarithm of the stimulus.' Thus, for instance, if 1,000 candles are all throwing their light upon the same screen, we should require ten more candles to be added before our eyes could perceive any difference in the amount of illumination. But if we begin with only 100 candles shining upon the screen, we should perceive an increase in the illumination by adding a single candle. And what is true of sight is equally true of all the other senses: if any stimulus is increased, the smallest increase of sensation first occurs when the stimulus rises one per cent, above its original intensity. Such being the law on the side of sensation, suppose that we place upon the optic nerve of an animal the wires proceeding from a delicate galvanometer, we find that every time we stimulate the eye with light, the needle of the galvanometer moves, showing electrical changes going on in the nerve, caused by the molecular agitations. Now these electrical changes are found to vary in intensity with the intensity of the light used as a stimulus, and they do so very nearly in accordance with the law of sensation just mentioned. So we say that in sensation the cerebral hemispheres are, as it were, acting the part of galvanometers in appreciating the amount of molecular change which is going on in sensory nerves; and that they record their readings in the mind as faithfully as a galvanometer records its readings on the dial. * * * * *