752 BULLETIN 82, UNITED STATES NATIONAL MUSEUM. 



yellow tint, and no bands appeared. Hydrochloric acid, as well as acetic acid, also 

 destroyed the color of this solution. 



The absolute alcohol solution was of a fine rose color ; a deep layer transmitted 

 the red rays and a little of the green, but in the deepest layers there was not a 

 trace of a chlorophyll band in the red. A shallower depth showed an ill-defined 

 shading in the green, best marked toward its blue end. Ammonia discharged most 

 of the color, leaving a faint reddish tint and some ill-defined shading at the blue 

 end of the green. Hydrochloric acid destroyed the red color, changing it to yellow, 

 and the solution now showed no bands. 



The ether solution did not show a trace of a chlorophyll band in the red ; 

 it was of a pale yellow tint, and in a thin layer a faint, ill-defined band was seen 

 at the blue end on the green. On evaporation it left a yellow residue, which be- 

 came a transient dirty green with sulphuric acid, and became more yellow with 

 iodine in potassic iodide. 



Therefore it would appear that the ether removed a yellow lipochrome and 

 left the rose-red pigment untouched. 



So far, then, it may be concluded that the intrinsic coloring matter of Antedon 

 bifida is not identical with Moseley's antedonin, which will be described for pur- 

 poses of comparison farther on, and that no chlorophyll is present. The absence 

 of symbiotic algae, threfore, may be safely inferred. 



Knikenberg had previously examined the pigment of Antedon adriatica, and 

 had referred to a band in the red evidently belonging to chlorophyll. He called 

 the red pigment comatulin, and noticed that it, as well as the pigment of yellow 

 and brown specimens, is soluble in sea water and in diluted alcohol, but insoluble 

 in glycerine, ether, chloroform, benzol, carbon disulphide, turpentine, and almond 

 oil, only going over into these solvents when water is present. 



MacMunn remarks that it would appear that Krukenberg did not take the 

 precaution of removing the contents of the stomach, which, when squeezed out, show 

 the dominant and other chlorophyll bands well marked. He did not dry his speci- 

 mens completely, which may account for the discrepancy in his results as compared 

 with those of Krukenberg. 



According to Krukenberg the red, brown, and yellow pigments from corre- 

 spondingly colored specimens of comatulids are all related to each other, and none 

 of them yield absorption bands. He states that an extract of red comatulin be- 

 comes more yellowish by dilution, while an extract of yellow comatulin becomes 

 more red by concentration. If a slightly purplish red solution of comatulin in 

 dilute alcohol be evaporated on a water bath and the residue be exposed to the 

 light, the cherry red color changes first into an insignificant brownish yellow, 

 and this behaves exactly like the piirment obtained from brownish yellow species; 

 later it gets lighter in color until finally it becomes bleached. 



Diluted aqueous solutions of comatulin also undergo a similar change, but alco- 

 holic solutions are more stable. 



He also says that aqueous and alcoholic solutions are changed to orange or 

 brownish yellow by hydrochloric and acetic acids, and that the coloring matter 



