No. 540] 



STUDIES ON MELANIN 



747 



and which reoxidize to the original color and compound. 

 but the azo compounds are distinguished from these 

 colors by the fact that no reoxidation takes place. 



4. I have been unable to find any data as to the crystal 

 form of cuticula pigments. All writers report that they 

 are non-crystalline. If Tower succeeded in preparing 

 crystals he lias succeeded where every one else has failed. 

 The azo compounds, on the other hand, are usually easily 

 obtained in crystal form. 



5. As to the value of the spectrum I will quote Tower's 

 own words (1903, p. 56). 



The study of animal and plant pigments has been carried on almost 

 entirely by means of spectrum analysis. . . . There is, however, very 

 grave doubt as to the value of such spectrum work. After spending 

 several months in the analysis of cuticula pigment of Coleoptera, Lepi- 

 dopterous larva\ and 1 1 ymenoptera, I discarded as worthless all of my 

 results, because in no case was I able to obtain a spectrum of sufficient 

 permanence, or one that hail absorption hands of enough constancy, to 

 be reliable. The same solution with the same apparatus and light will 

 vary in its spectrum from day to day. Moreover, any slight change 

 in acidity or alkalinity, or in dilution or concentration, etc., will pro- 

 duce a variation in the spectrum. 



6. In regard to the decomposition products of cuticula 

 pigments, I can find no data in Tower's paper, so that I 

 do not know in what respect they may resemble known 

 organic compounds. Enough has been shown, however, 

 to prove that Tower's statement (1906, p. 136), "we are 

 safe, then, in concluding that these cuticula pigments 

 are azo compounds, but whether they are azo, diazo, 

 oxyazo, or amidoazo 3 is not known," is entirely with- 

 out foundation. 



8 For those who are not familiar with this nomenclature I will give here an 

 example of the formula of each class: 



Azobenzene, C«H 6 — N=N — C„H 8 , 

 Diazobenzene, C e H s — N=N— OH. 

 Tower probably refers to Dittazo compounds, inasmuch as diazo compounds 

 are very unstable and are not coloring matters. 



Disazobenzene, C.II — N=N— CJI — N=N— C,H„ 

 Oxyazobenzene, C,H 5 — N=N— C„H,OH, 

 Aminoazobenzene, C.H S — N=N— C.H^NH,. 



