416 Mr. J. C. F. Fryer on 
plant with both green and brown stems, and were left 
until they had pupated. The results were as follows :— 
In red light. Five green pupae, one brown pupa (two 
(eo) ) 
green on green stems, three green on 
brown stems, one brown on brown stem). 
In green light. Five green pupae (four on green stems 
5 5 fama 
one on brown stem). One larva died 
. before pupation. 
In blue light. Five green pupae, one brown pupa (four 
o Hage 
green on walls of vessel, one green on 
brown stem, one brown on brown stem). 
EXPERIMENT V. 
A single wild pupa was discovered on a black tarred post 
in a very exposed situation, and it was noted as being 
exceptionally dark in colour. To test this point further 
a tage was prepared as in Experiment II, but the paper 
used was black instead of green; no food-plant was given. 
Six larvae pupated inside and all formed brown pupae, 
but the colour was entirely normal and there was no sign 
of darkening in response to the black surroundings. 
EXPERIMENT VI. 
An attempt was made to discover the exact period during 
which the colour of the pupa is determined. The larva 
as a rule remains on the food-plant until quite full-fed; 
then during the night it wanders until it finds a suitable 
spot, fixes itself there, and by morning has assumed the 
usual curved semilunar attitude, supported only by the 
silk thread behind the thorax and by the silken pad to 
which the terminal segment of the abdomen is affixed. In 
the morning it is still a clear translucent green, but towards 
the evening it becomes somewhat opaque and lighter in 
colour; during the night the larval skin 1s shed and by 
the next morning the pupa is fairly dry and hard. Any 
stimulus therefore due to daylight must be received during 
the day spent in the “semilunar” position. To test this 
more exactly full-fed larvae (number uncertain—between 
six and twelve) were allowed to fix themselves in green 
light and were left until mid-day, when they were trans- 
ferred to ample white hght amidst dark surroundings, 
conditions previously found favourable to the formation 
