308 JOURNAL, BOMBAY NATURAL HISTORY SOCIETY, 1890. 



chain of wonderful similitude from beginning to end. The hint above referred 

 to will be found, so far as I know, only in the Butterflies of India, where it is 

 stated that the chrysalis of D. chrysippus assumes two forms, the one being 

 green, its normal color, and the other a delicate pink, from which it is inferred 

 that the latter colour may be due to its imitating the tint of the buds and flowers 

 of the food plant, which are pink. 



Unless there be two wellc marked and constant varieties of these pupae, which 

 I doubt, the hypothesis is correct, but nowhere, except among the blossoms 

 themselves or in their immediate vicinity, can such a change take place, as certain 

 conditions seem essential for perfecting themselves in the colour of their surround- 

 ings. 



From the few experiments I have as yet made, and which I shall now record, I 

 believe that the faculty for protective colouring is strongly developed in all 

 chrysalides (I am not in a position to say, however, whether it is effected naturally 

 or wilfully, but believe the former to be the case), and that they possess the means 

 of toning themselves to their surroundings. Perhaps I may be too sweeping in 

 asserting that all chrysalides can do so, as in the present early stage of the 

 investigation, one cannot be too certain of results that are to be acquired only on 

 patient and careful research, but should any eventually be found constant to 

 their normal colour, I feel confident they will not prove the largest number. 



Thus far I have experimented on only two species of butterfly, common ones 

 luckily, but belonging at the same time to quite distinct families and the results 

 have interested and astonished me in no small degree. 



Danais chrysippus was one of my "subjects" and as stated above the colours of 

 its pupse are either green or pink, but following up a slight change of colour that 

 accident put into my hand I succeeded in getting them pure white, red, salmon, 

 Hack and blue. The two last colours, however, were not quite pronounced like the 

 rest, and were due, no doubt, to a slight error in operation, which I shall explain 

 further on. 



The other insect was P. erithonius, the eggs and larva; of which 1 got off a bael 

 tree, Mgle munnelos. (I do not know if this has been recorded as one of its food 

 plants, as writing away from all my books and papers, I have not the facility of 

 a reference, though I fancy it has been noted as such.) The pupas of this buttarfly 

 also are green, but in this instance even I succeeded in changing them to red, pink, 

 salmon, brown and other colours. Now these are curious and interesting results, no 

 doubt, and how were they arrived at? Simply by following certain rules that would 

 have come into full play in nature, and from which I conclude that the pupa; of 

 butterflies can be made to assume any tint, as it is a provision for protective 

 purposes. The subject, however, is not fully investigated as yet, and I therefore 

 throw out these few hints in the hope that other members of the Society may 

 follow them up, and place us in possession of some more interesting facts. 



A few hints seem necessary to carry out the investigation successfully, barring 

 which the experiments will result in failure. A strong light seems an indispen- 



