No. 540] 



STUDIES ON MELANIN 



745 



difficulty that they can he bleached": and on page 44 

 "one nses camel hair in the natural color or dyed dark 

 inasmuch as it can not be bleached." This is the only 

 connection in which the natural-occurring pigments are 

 mentioned. 



Tower states (1903, p. 63) that: 



The chemical nature of colors is a problem most difficult of investi- 

 gation, chiefly because such energetic measures are necessary to get the 

 color into solution that there is every reason to suspect that it is no 

 longer the same as that in the cuticula. With solutions of P. comutus 

 in HC1 or H 2 S0 4 , various tests were made, which established the fact 

 that it is much like some of the benzine derivatives, and the spectra of 

 this solution and of permanent brown are identical and closely similar 

 to that of Bismarck brown. There is, however, a great difference be- 

 tween the solution of cuticula color and Bismarck brown, the former 

 being a colored substance and the latter a coloring substance. 



Cuticula colors in acid solution are decolorized by reducing agents 

 such as tin and HC1 or strong alkalis. If, however, the decolorized 

 solution is treated with a mild oxidizing agent, part or perhaps all of 

 the color is restored. In the process of decolorization the solution 

 passes from deep brown to lighter shades, to yellow, and eventually to 

 a colorless solution. According to Steelier and others, fast brown is a 

 diazo compound belonging to the group of amidoazo and oxyazo com- 

 pounds, which form colored solids varying from yellow to deep brown. 

 These diazo, oxyazo, and amidoazo compounds are soluble in alcohol, 

 as is this cuticula color.- and when in solution are rendered colorless 

 by reducing agents, thus forming colorless azo, or hydrazo compounds, 

 which, by mild oxidation, may be reconverted into yellow- or brown- 

 colored diazo, oxyazo, or amidoazo compounds. There is thus a very 

 close agreement between the reaction of cuticula color and the diazo, 

 oxyazo, and amidoazo compounds to reducing and oxidizing agents. 

 In these and in other characters the cuticula colors resemble these ben- 

 zine derivatives, but not in any respect do they resemble other colored 

 substances or dyes known to organic chemistry. The existence of azo 

 compounds has been recognized by Bottler in the hairs of animals and 

 in sdk fibers, where they 1 unction as pigments and have some of the 

 structural peculiarities of cuticula color. I feel perfectly certain, 

 therefore, that these cuticula colors are azo compounds, a conclusion 

 based upon the following characters: (1) their colors; (2) solubility; 



2 On page 41 of the same work Tower states that these cuticula colors are 

 "insoluble in water, alcohol, ether, oils, weak acids or alkalis. Soluble in 



