the colours of certain Lejndoptera. 4J:7 



remarkable likeness to the surface upon which they are 

 fixed, and I wished to ascertain whether this was entirely 

 due to the abundant admixture of adventitious fragments 

 gnawed off the surface, or in part to the colouring of the 

 silk. 



Six or more mature larvae were placed in glass 

 cylinders (1 in each) standing on sheets of glass ; 

 beneath the latter and round the lower half of the 

 cylinder white paper was fixed, so that the larvae were 

 in brilliantly illuminated white surroundings, and yet 

 were compelled to spin cocoons from the products of 

 the silk-glands alone. 



About 6 more were similarly placed, except that black 

 paper was used instead of white. 



Under these circumstances, most of the larvae spun 

 compact semi-transparent cocoons, the product of the 

 glands apparently forming a continuous sheet. Some 

 of them, however, failed to construct cocoons, and only 

 used the secretion to form a covering to the glass floor. 



The cocoons varied much in colour, being all shades 

 of brown, but there was no evidence whatever for the 

 existence of any sort of relation to the colour of the 

 environment. 



It is clear that the adjustment of colour which occurs 

 in nature is, like that of the cocoons of H. abruptaria 

 (see p. 317) entirely due to adventitious particles. 



Endromis versicolor. — A single mature larva was placed 

 among some shreds of white paper. It spun a cocoon 

 of normally dark brown silk. 



Tricliiura cratagi. — Some experiments similar to those 

 described below in the case of P. popidi were made upon 

 this species, with the same negative results. All the 

 cocoons were dark. I have mislaid the notes with the 

 exact numbers. 



Hemaris fuciformis 1889. — Six cocoons were spun in 

 two white paper boxes with clear glass covers, 5 of them 

 being among shreds of white paper. The silk of all 

 was more or less brown, but it varied through all shades 

 from dirty white up to dark brown. 



Four cocoons were spun among dark twigs in a box 

 lined with dark tissue paper, also with a clear glass 

 cover. These varied in tint in the same manner. 



It seems clear that the species is not capable of modi- 

 fying the colour of its cocoon into correspondence with 

 its environment. 



