1 and (zero) agree with. I and 0, 2 is red, 3, 6, and 9 more or less greenish, 4 and 5 bluish, following the letter V, 7 orange, and 8 light yellow. I have also tried the Greek alphabet, with a view to testing its possible color associations. I find, however, that I can not separate these letters from their Latin cognates. Theta seems to me as greenish as tau; chi (X) vacillates between C and X, and psi (Ψ} is like the ps of which it is composed.
I have also made attempts to find the color relations in the Chinese alphabet, but without much success. I have no childish associations with these letters, and I can imagine color only in those which in some way suggest by their form the letters in the Latin alphabet. Thus T (tree) seems greenish like T, and xx (woman) seems to follow Z or X.
I find that with other persons who confess to similar color associations there is a decided lack of agreement as to the impressions produced by most of the different letters.
My friend, Prof. Edward Spencer, has given me a chromatic alphabet, arranged as follows:
| Shining black, | I, E, H, R, T, | Chocolate, | G. |
| 1, 3, 4, 6, 7, 9. | Light gray, | O, N, X, 5. | |
| Dull black, | F, J, K. | Pale, | D, Z, V. |
| Brown, | B, M, Q, W. | White, | O. |
| Golden, | C, 8. | Water-color, | U, Y. |
| Orange, | S. | Without color, | L, P. |
In this category the letters for the most part represent gradations from jet-black to white. We may, however, trace some relation between the supposed colors and either the forms or the sounds of the letters. Except in the color of the vowels I, O, U, and the isolated and emphatic position of S, there is little in common with the list above given by me.
In my own case, although I have no recollections to justify the theory, I feel sure that these associations are due to the bringing together of a childish classification of letters, with childish categories of color. I was, more than most children, interested in the individuality of the letters. I liked to assort them, to play with them, and transpose them to form other words. In like manner I was interested in colors. I had a childish liking for blue above the other colors, as also a meaningless preference for V and D over the other letters. I can, therefore, see how V and D should be associated with blueness. Other letters of pronounced qualities, as R, L, X, Z, came to head other categories, and the letters which. I regarded as indifferent took their places next to those which in form or sound or otherwise appeared to the child similar to these.
Dr. Gustaf Karsten, of the University of Indiana, a philologist interested from boyhood in phonetics, recognizes color in the various vowel sounds, but none in the letters themselves. Thus a (in
