LEPIDOPTEROUS PUP^ AND THEIR SURROUNDING SURFACES. 
313 
species (which Mr. Wood expressly affirms were unknown to him), at once stated in 
the discussion which followed the paper that “ he had had thousands of pupse of 
Papilio machaon, and had often had the brown variety of pupa on a green ground¬ 
colour, whilst in some seasons he had obtained no brown specimens at all.” My own 
experiments also clearly show, as far as the numbers employed could do so, that this 
species has no trace of susceptibility to surrounding colours. Mr. A. G. Butler also 
stated in the discussion that “ he had obtained a red or rosy chrysalis of Pieris rapes, 
which had undergone its transformation in a piece of scarlet cloth ; and pupse upon 
glass were generally of a pale slate colour.” Concerning the former observation, 
P. rapes is very commonly tinged with pink, and I think it almost certain that if the 
pupa w 7 as not removed from the scarlet surroundings the reflected light would intensify 
the pink tinge of the ground colour, and would thus produce a simulated resemblance 
between the tw 7 o. I found that it was never safe to compare the colours of pupse 
until they had been removed from the coloured surface on which pupation had taken 
place, and were arranged side by side upon white paper. Furthermore, in my experi¬ 
ments a. dark red surface did not produce reddish, but dark, pupse in P. brassices, and 
(as will be shown below) I can quite confirm Mr. Bond’s criticism that reddish pupse 
of P. rapes are not found on red brick walls. At the same time Mr. Bond’s objection 
to the main position taken up by Mr. Wood —that the pupse of Pierides have some 
general resemblance with their surroundings—is without sufficient foundation, and 
the fact must now be generally admitted. His criticism that he has seen pupse of 
P. rapes with all shades of colour on the white painted woodwork of a greenhouse is 
valueless unless he can show that the dark forms are as common as when the pupse 
are found on a tarred fence. Mr. Bond’s observation that the variable pupse of 
Anthocaris cardamines are not sensitive is important, but the species needs experi¬ 
mental investigation in order to confirm his observations. 
The late Mr. Edward Newman also expresses an opinion adverse to Mr. Wood’s 
observations upon P. rapes in these words (‘British Butterflies,’ 1871, p. 162), “An 
ingenious—but, as I think, futile—attempt has been made to show that the colour 
of the chrysalis varies with the colour of the object to which it is attached.” 
Furthermore, Bev. J. Hellins (‘Larvae of British Butterflies,’ &c., by William 
Buckler (Ray Society), 1885, p. 155) states of the pupse of P. rapes, “The colour 
seems very varied, but, as all the varieties occurred side by side on the cover of the 
tin box in which my larvae were reared, I could not account for their difference.” 
And of the different forms of P. brassicee he also says, “ These varieties were 
developed side by side in the same cage.” It will be abundantly proved below that 
only certain colours affect each species which is sensitive to surrounding colours, and 
when other colours are made use of the pupse either assume the commonest form, or, 
if very variable, are free to obey the influence of their varied hereditary or individual 
tendencies. And the use of tin as a background for the Pieridse is especially likely 
to produce such results, for this surface is • extremely unlike any of their natural 
M'DCCCLX XXVII.-B. 2 S 
