LEPIDOPTEROUS PUPAE AND THEIR SURROUNDING SURFACES. 
347 
Thus the blinded larvae were, on the whole, more golden than the others, but of 
course no significance can be attributed to this. From the frequently recurring 
examination of the larvae it is possible to form a very accurate estimate of the 
duration of Stage III., although the whole period and the other stages cannot be 
similarly made out from the notes. The different estimates are as follows :—• 
lirs. min. 
In 2 larvas the duration of Stage III. is estimated at 
7 
1 
1 
1 
2 
2 
55 
55 
55 
55 
55 55 
5 5 5 5 
55 5 5 
5 5 5 5 
5 5 5 5 
5 5 5 5 
55 
55 
55 
55 
55 
55 
15 0 
16 0 
16* 0 
16 20 
17* 0 
18 0 
19 0 
16 larvae. 
Average .... 16 37 
Thus estimates were obtained for this stage in 16 larvae out of 24, and in many 
cases the estimates must have been very nearly accurate. Furthermore, the extremes 
only differ from each other by four hours. 
In addition to the above estimates, there is some insight into the other stages of a 
single larva in (/3) subdivision, and in this instance it is shown that Stage I. must 
have occupied at least three hours, and Stage II. about six hours. 
The experiment also shows the effect of white surroundings. 
B. The globes already described in Series Y. were made use of for this experiment} 
all the arrangements being identical, except that N os. 5 and 11 were omitted (as they 
still contained larvae of the previous series). The larvae in the globes were alternately 
blinded and normal, two larvae being placed in each globe, except two (numbered 
13 and 14), for it was thought that if allowance were made for their position this 
number might be safely included without introducing error. Notes were not taken 
with sufficient frequency to render it possible to estimate the length of the stages of 
the preparatory period, and therefore it is unnecessary to give more than the results 
of the experiments. The larvae were introduced into the globes on August 26, 
10.30-11 a.m. :— 
2 y 2 
