Now suppose the strings of the basilar membrane to be tuned to notes of all different pitches, within the range of {235} audible vibrations: then each string would be thrown into sympathetic vibration whenever waves of its own vibration rate reached it by way of the outer and middle ear; and the sense cells standing over the vibrating fibers would be shaken and excited. The theory is very attractive because it would account so nicely for the great number of elementary tone sensations (there are over 20,000 fibers or strings in the basilar membrane), as well as for various other facts of hearing--if we could only believe that the basilar membrane did vibrate in this simple manner, fiber by fiber. But (1) the fabric into which the strings of the membrane are woven would prevent their vibrating as freely and independently as the theory requires; (2) the strings do not differ in length a hundredth part of what they would need to differ in order to be tuned to all notes from the lowest to the highest, and there is no sign of differences in stretch or in loading of the strings to make up for their lack of difference in length; and (3) a little model of the basilar membrane, exposed to sound waves, is seen to be thrown into vibration, indeed, and into different forms of vibration for waves of different length, but not by any means into the simple sort of vibration demanded by the piano theory. This theory is accordingly too simple, but it probably points the way towards some truer, more complex, conception. The fact that there are many elementary sensations of hearing is the chief reason why the art of tones is so much more elaborate than the art of color; for while painting might dispute with music as to which were the more highly developed art, painting depends on form as well as color, and there is no art of pure color at all comparable with music, which makes use simply of tones (and noises) with their combinations and sequences. {236} Senses of Bodily Movement It is a remarkable fact that some parts of the inner ear are not connected with hearing at all, but with quite another sense, the existence of which was formerly unsuspected. The two groups of sense cells in the vestibule--the otolith organs--were formerly supposed to be the sense organ for noise; but noise now appears to be a compound of tones, and its organ, therefore, the cochlea. The _semicircular canals_, from their arrangement in three planes at right angles to each other, were once supposed to analyze the sound according to the direction from which it came; but no one could give anything but the vaguest idea of how they might do this, and besides the ear is now known to give practically no information regarding the direction of sound, except the one fact whether it comes from the right or left, which is given by the difference in the stimulation received by the two ears, and not by anything that exists in either ear taken alone.