colour of cocoons, pupce, and larva. 



211 



the action of light. A number of larvte were shut in 

 gilt boxes as described, and these were immediately 

 placed in a dark, closed cupboard, which was not opened 

 again until the larvse had pupated. With few exceptions 

 all these pupae belonged to the darkest class (see table). 

 Other larvie were put in a black box and similarly 

 treated, with the same result. 



The larvfe were collected from various places round 

 Cambridge, and belonged to some dozen or more batches 

 of larvffi, but I distributed the families among the boxes 

 so as to test the existence of any congenital differences 

 as regards pupal colour, but found none. 



Some interest attaches to the fact that the great pro- 

 portion of larvae collected by me were infested with 

 Tachincs. Probably, in round numbers, five or six 

 larvae died from Tachina for one that pupated, but those 

 that did pupate almost without exception emerged. 

 There is therefore no reason to suppose that either the 

 gilt pupje or the dark ones are diseased. 



Amount 



of 

 Gilding. 



Gold paper , 

 Hilver do. 

 White do. 

 Yellow do. 

 JJlack do. 

 Indian Ink 



Shut in the 

 dark: — 



Gold 2(3 



Black y 



Much Pigment. 



Some Pigment. 



Little oe no 



PiGMKNT. 



None. Some. I Much. 



14 

 S 

 2 

 5 



Taken together — gold, silver, yellow, and white papers gave 

 2 dark, 16 moderate, 23 light ; black paper and Indian ink gave 

 31 dark, 5 moderate, 3 light ; gilt paper, shut in the dark, gave 

 29 dark, 1 moderate ; black paper, shut in the dark, gave 9 dark. 



There are, of course, two things to be thought of : 

 first, the pigmentation ; secondly, the metallic colours. 

 As the table shows, both these qualities seem to be 

 affected by the surroundings. As Poulton has men- 

 tioned, the metallic appearance is an interference-colour, 

 disappearing when the pupae are dried, returniug when 



