EFFECTS OF FOOD ON COLOUR. 67 



proved which, fundamentally considered, is not so. Wallace, for 

 instance, relates that a Brazilian parrot Chrysalis f estiva can 

 be made to change the green in its feathers to yellow or red 

 if it is fed on the fat of certain fishes allied to the shad a method 

 largely adopted by the Indians. The same traveller further 

 asserts that the splendid Indian bird, Lori Rajah, is said to 

 preserve its gorgeous colouring by a peculiar mode of feeding. 

 The bullfinch is said to turn black when fed on hemp-seeds ; 

 recently a splendid orange-coloured variety of the canary has 

 been introduced into commerce, and it is said it is produced by 

 feeding ordinary specimens of the bird on Spanish pepper. The 

 statement is well known that butterflies, and more particularly 

 species of the genus Euprepia, assume an abnormal colouring 

 when the caterpillars are fed on leaves which they are not 

 accustomed to ; thus Euprepia caja becomes quite brown when 

 the larvae are fed entirely on walnut- leaves. This assertion, 

 however, has been frequently contradicted, and no systematic 

 and experimental investigations, as directed expressly to this end, 

 have ever been made to my knowledge, for the independent 

 experiments in feeding made accidentally or by a happy chance by 

 different entomologists cannot in fact be regarded as physiologi- 

 cal experiments. Still less can the statements made by travellers, 

 as by Wallace, count as such, since they rest entirely on hearsay 

 from wild Indians, and not on the results of their own investi- 

 gations. Of course I am far from asserting that no such direct 

 modifying influence of food on the colour of animals exists, or 

 that it is improbable ; I only would point out that up to the 

 present time we know nothing exact on this point, and that 

 nothing is actually proved beyond the possibility or probability 

 of such an influence affecting the skin-pigment of various 

 animals. As to the nature of this chemico-pbysiological process, 

 which is what is truly worth knowing in the matter, so far as 

 I know, not even a hypothetical view has as yet been ex- 

 pressed. 



A few experiments are better established which prove that 

 certain structural relations may be entirely changed by the 

 direct influence of food. The English anatomist Hunter pur- 

 posely i:ed a sea-gull Larus tridactylus for a whole year on 



