THE PROBLEM OF COLORED AUDITION. 815 



other persons the impression is not visual, but may belong to a 

 different sense. It may be sonorous ; in some persons the sight of 

 colors gives a musical impression ; or it may be tactile, and sight 

 and hearing may be accompanied by mechanical sensations. In 

 short, all imaginable combinations of different sensations may be 

 realized. 



In colored audition the impressions of color are almost exclu- 

 sively provoked by speech ; the sounds and noises of Nature pro- 

 ducing the same effect only by a kind of analogy with the human 

 voice. Speech gives him who hears it an impression of color only 

 when the emission is full; a murmur has not the effect of the 

 singing voice or of a reading in public ; the height of the tone in- 

 fluences the shadings ; barytone and bass voices excite dark sensa- 

 tions and high voices light ones. On a closer examination of the 

 source of the phenomenon it is found that the color, while it may 

 borrow a general tint from the timbre of the voice, and con- 

 sequently from the individuality of the speaker, depends more 

 especially upon the words that are pronounced ; each word has its 

 peculiar color, or we might rather say colors, for some words have 

 five or six; pushing the analysis further, we perceive that the 

 color of words depends on that of the component letters, and that 

 it is therefore the alphabet which is colored ; and, finally, that the 

 consonants have only pale and washed-out tints, and the colora- 

 tion of language is derived directly from the vowels. With a few 

 exceptions this is true for all the subjects. 



By a curious complication produced by education, the appear- 

 ance of colors takes place in some persons not only when they 

 hear the word pronounced or when they think of it, but even when 

 they see it written. There are also persons who do not perceive 

 the color except while they are reading. Many facts, however, 

 seem to prove that reading is generally of no effect except as a 

 suggestion of the spoken word, and therefore constitutes a kind of 

 audition. 



The observations on the colors of the vowels in detail are irreg- 

 ular and contradictory. Thus, a, red to one, is black to another, 

 white to a third, yellow to a fourth, and so on ; the whole spec- 

 trum passes through it ; but as the number of colors and of letters 

 is limited, we can, by analyzing a hundred observations, meet two 

 or three among them that will agree. Sometimes agreement is 

 manifested between members of the same family, or between per- 

 sons who live together ; but waiving the instances afforded by 

 chance, by heredity, and by suggestion, it remains evident that 

 disagreement is the general rule ; and from this curious practical 

 effects follow. Two persons having colored audition, when 

 brought together, are not able to understand one another ; each is 

 greatly surprised at the colors which the other perceives, and we 



