We can most of us, perhaps, recall similar experiences, where colors and sounds, in themselves indifferent, took on either through analogy or association a decidedly repulsive character. How far, one wonders, does this process of transformation of things go in the case of imaginative children? There is some reason to say that it may go very far, and that, too, when there is no strong feeling at work cementing the combined elements. A child's feeling for likeness is commonly keen and subtle, and knowledge of the real relations of things has not yet come to check the impulse to this free, far-ranging kind of assimilation. Dickens was not, one feels sure, the only child who saw odd resemblances in letters, finding, for example, that the thick O and S of his primer stood out from the rest as the easy, good-natured ones. This sort of fanciful reading of character into things is of the very life of childhood. Before the qualities and the connections of objects are sufficiently known for them to be interesting in themselves, they can only acquire interest through the combining art of childish fancy. And the same is true of associated characters. A child's ear may not dislike a grating sound, a harsh noise, as our ear dislikes it, because of its immediate effect on the sensitive organ. En revanche it will like and dislike sounds for a hundred reasons unknown to us, just because the quick, strong fancy, adding its life to that of the senses, gives to impressions much of their significance and much of their value.
There is a new field of investigation which is illustrating in a curious way this wizard influence which childish imagination wields over the things of sense. It is well known that a certain number of people habitually color the sounds they hear, visualizing the sound of a vowel, or of a musical tone, as having its characteristic tint which they are able to describe accurately. This "colored hearing," as it is called, is always traced back to the dimly recalled age of childhood. Children are now beginning to be tested, and it is found that a good proportion possess the faculty. Thus in the researches on the Boston children already referred to it was found that out of fifty-three, twenty-one, or nearly one half, described the tones of certain instruments as colored. The particular color, as also the degree of its brightness, ascribed to an instrument, varied greatly among different children, so that, for example, one child visualized the tone of a fife as pale or bright, while another imaged it as dark.[1] It is highly probable that both analogy and association play a part here.[2] As was recently suggested to me by a correspondent, the classic instance of the anal-
