174 NATURAL SCIENCE. March, 



to be due either to the special characters of the individual scales, or 

 to the relations of different sets of scales. Each scale consists of a 

 double membrane, the parts being separated by a space which is 

 continuous with the cavity of the stalk of the scale. The inner 

 membrane is transparent, colourless, and usually slightly folded ; the 

 outer shows considerable differentiation and usually bears on its outer 

 surface longitudinal rows of blunt projections, which are of much 

 importance in the production of structural colours. The two 

 membranes are connected by bridges of chitin. When pigment is 

 present, it may occur in the form of granules or may be simply 

 diffused through the chitin ; it is always absent from the inner 

 membrane, but it occurs freely in the bridges of chitin, in the stalk of 

 the scale and its surrounding follicle, and in the outer membrane. 



As to the exact nature of the pigments, our knowledge is still far 

 from complete. The best known is perhaps the yellow pigment of the 

 Pieridae, which for some time has occupied the attention of Mr. F. 

 Gowland Hopkins (4), (5), (6). This pigment is soluble in hot water, 

 but not in the usual organic solvents (alcohol, chloroform, benzol, 

 etc.); it gives the murexide reaction quite distinctly, is acid to litmus, 

 and undoubtedly a derivative of uric acid. In the wings of the 

 Pieridae, indeed, uric acid itself is said to occur as a white pigment. 

 According to Urech (16), however, the white scales of Pieris brassier 

 appear yellow by transmitted light, and when treated with hot water 

 yield a yellow solution which, besides yellow pigment, contains a white 

 "crumbly" substance. This observation makes it appear doubtful 

 how far even in this case white is an optical effect. White is certainly 

 in most cases in butterflies an optical colour ; in many cases it changes 

 according to the direction of the light. In these cases, Spuler is 

 inclined to think that it is produced by the relations of the layers of 

 chitin in the scales. The phenomenon is thus to be explained as due 

 to the interference colours of thin plates. It is interesting to compare 

 the result obtained by Urech in the case of Pieris brassier with the 

 observations made by Krukenberg on birds. Krukenberg (7) found 

 that even the apparently pure white feathers of a cockatoo yielded a 

 solution containing the yellow pigment characteristic of the family. 



The yellow pigments found in the other Lepidoptera are not as 

 yet well known ; they do not all give the murexide reaction, nor are 

 they in all cases soluble in water. 



Among the other pigments, the red presents some interesting 

 features. It has been noticed by several observers that in many cases 

 scales of a bright red or scarlet colour yield a yellow solution when 

 treated with hot water, but if the solution be evaporated to dryness a 

 red residue remains. Similarly, the red colour is turned yellow by the 

 application of acid, but the yellow colour may be restored by the 

 addition of ammonia. This is the so-called "reversion effect" of 

 Mr. Perry Coste (2). According to Hopkins, the red pigment is either 

 formed from the yellow by hydration, or by the association of the yellow 



