tlie colours of certain Tjcpidoptcra. 317 



passing through a sheet of yehow ghiss. The larvfiG 

 were restmg on a piece of bhick net, and the dark and 

 light varieties were intermixed, so that examples of both 

 were in areas of all degrees of illumination (which differed 

 greatly on the two sides of the net). A collotype of the 

 negative is shown on Plate XV., tig. 3. 



Tlie larviie were painted by ]\Iiss Cundoll on July 27, 

 and reproductions of the drawings are shown on Plate 

 XIV., figs. 1 and 2. The larvii? are represented of the 

 natural size, and the colour-difference is very well shown, 

 although the attitudes of the resting larva3 are not quite 

 natural. They were probably temporary attitudes as- 

 sumed after disturbance. 



The colours of the cocoons were also tested in these 

 experiments. At first sight the power of adjustment to 

 the surface of attachment seemed to be undoubtedly 

 present, but when I examined the cocoons in 1H89, 

 I soon found that the appearance was due to adventitious 

 material being woven into the fabric. The resemblance 

 to surroundings is extremely perfect, and so well packed 

 and so small are the foreign particles that the light 

 brown silk does not in the least interfere with conceal- 

 ment on a dark surface. This is shown in Plate XIV., 

 fig. 2, where a cocoon is represented on the right side of 

 the base of the twig of Qnercus cerris. When the cocoons 

 were spun on muslin, the larvfie had but little power of 

 gnawing off fragments, and these being few and thinly 

 scattered, the appearance of the cocoon was made up l)y 

 the light brown silk. Their power of dealing with paper 

 was superior to that of dealing with muslin, but far less 

 than when supplied with bark, wdiich is probably the 

 natural surface on which they spin. 



16. EXPEEIMENTS IN 1886, 1887, xVND 1888 UPON 



KUMIA CRATiEGATA. 



1886. 



These experiments have a personal interest to me, 

 inasmuch as they first indicated that the power of 

 individual colour-adaptation was widely present among 

 lepidoptorous larvie, and was best studied among the 

 most perfectly concealed forms, rather than among the 

 SpJiiin/idce. I have already said that I owe the suggestion 

 to test this species to Lord Walsingham. 



A captured female laid the eggs which furnished the 

 material for this experiment. 



