LEPIDOPTEROUS PUPiE AND THEIR SURROUNDING SURFACES. 
387 
alternative hypothesis—that something in the skin is sensitive to colour influences— 
seems at first sight to be opposed by the stronger effects which followed the appli¬ 
cation of the stimulus to the smaller, anterior, skin area. But, in the first place, the 
difference between the sizes of the two areas was very small, for the anterior part is 
much swollen before pupation ; and, furthermore, I now feel sure that I over-compen¬ 
sated for the greater illumination of the upper chamber. I did not allow for the fact 
that the larva always spins a film of silk over the glass for a considerable distance 
round the boss from which it is suspended, and that the transparency is much 
impaired in this way. Finally, it has been shown that when in the last experiment 
(XIII., B., y) there was an equally large opening to the upper as to the lower chamber 
the gilt surface when above did actually produce greater results than when it was 
below, although the conclusion is only supported by a pair of pupae (the sole survivors 
in this experiment). Although there are so few, the fact that the only (5) out of 
83 pupae was thus produced is certainly important testimony. Finally, when we 
compare these results with those of the more perfect conflicting colour experiments 
conducted with the frames previously described (XII., D.), and which will be further 
mentioned below, we see that the above explanation of the effects produced in the 
compartmented tubes is probably correct; that such results show that the sensitive 
surface is not represented by a sense-organ in the head, or with an anterior position 
only ; wdiile, on the other hand, when all the conditions of experiment are considered, 
the results harmonise well with the converse theory of a general susceptibility of the 
larval skin to the influence of certain colours. 
C. Notwithstanding the order in which the above experiments are arranged, the 
use of compartmented tubes was an earlier contrivance than the frames mentioned in 
XII., D., and by the end of the last experiments I realised that the latter method 
was far better, because of the equality of illumination, whether the gilt area was 
above or below. I therefore made a larger and better frame, of which a drawing is 
shown in fig. 4 (^ the real size), while a section is shown in fig. 5 (of the actual size). 
Upon the frame were pinned all the larvae which were found suspended in the stock 
cylinders, for those covered by the tubes were previously placed during Stages I. 
and II. in special boxes or cylinders. Unfortunately, the larvae on the frames began 
to pupate almost at once, and out of the 56 larvae only seven can have been 
subjected to the colours for nearly the whole of Stage III., while eight others had 
been on the frame for about 12 hours (two-thirds of the stage). The experiment is 
shown below in a tabular form :— 
3 d 2 
