104 Mr. E. B. Poulton. On a special Colour-Relation [Feb. 10, 



used for this purpose, but it is improbable that such was its original 

 meaning, for the fact that the appearance can be called up by the 

 appropriate surroundings shows that it belongs to the highest class of 

 protective colours, as far removed as possible from conspicuous 

 warning colours, the object of which is to become as unlike their 

 surroundings as possible. The former suggestion no doubt contains 

 the true origin of the character, if we add to it the statement that the 

 appearance is not only unlike anything organic, but strongly resem- 

 bles many common mineral substances, especially the widespread 

 mineral mica. The darker pupae, on the other hand, resemble grey 

 and weathered rock surfaces, just as the brilliant varieties resemble 

 many exposed and recently fractured rocks. The shape of the Vanessa 

 pupa is eminently angular and mineral looking. It is probable that 

 the glittering form arose in a hot dry country, where exposed rocks 

 would not weather for a long period of time. Gilded pupae of 

 Vanessa are formed from larvae which contain parasitic larvae of 

 ichneumon flies, probably on account of the absence of pigment in 

 such diseased individuals, and such absence being correlated with the 

 gilded appearance, the latter is therefore formed. Vanessa Io has a 

 green variety of pupa which appears when the insect is attached to 

 its food-plant ; V. atalanta has not such a form, and spins a tent of 

 leaves when it pupates on the plant, while V. urticm has neither the 

 green variety nor the latter habit, and exhibits a strong disinclination 

 to pupate among vegetal surroundings. During the past summer I 

 only found three pupae of the species on the food-plant in the field, and 

 all were " ichneumoned " and were abnormally gilded. 



III. Experiments upon Vanessa atalanta. — A few larvae of this 

 species, kindly sent me by Mr. J. L. Surrage, were subjected to gilt and 

 to black surroundings, while a few others were left in bright light 

 among the leaves of the food-plant. The results harmonised very 

 completely with those obtained from V. urticai^ the first set of pupae 

 being uniformly golden, the second very dark and with hardly any or 

 none of the gilded appearance, while the third were intermediate 

 but nearer to the former. The length of Stage (iii) appeared to be 

 about the same as in V. urticai, as far as this could be ascertained 

 from the limited data. 



IV. Experiments upon Papilio machaon. — Mr. W. H. Harwood 

 supplied me with larvae of this species. The eight largest were 

 selected and placed in brown surroundings (twigs, &c), four of them 

 being blinded. The larvae were very quiet and did not appear to be 

 irritated by the process, which was repeated three times. The posi- 

 tion of the ocelli on a distinct black area rendered it easy to ensure 

 that they are all covered with varnish. Eight bright green pupae were 

 obtained, fixed to the brown stems or roof or lying free on the brown 

 floor. This result surprised me very much, for I knew that there 



