1886.] Larva of Smerinthus ocellatus and its Food-plants. 151 



me to use some harmless pigment instead of silver nitrate. The 

 lamp-black mixed in the usual way with Mc. Guilp. acted in every way 

 satisfactorily, drying very quickly and being perfectly innocuous, and 

 completely opaque. The possibility of future success was shown by 

 these experiments, for the death of the larvae was certainly not due to 

 the conditions under which they were placed, as was shown by 

 comparing them with the normal larvae. 



Further experiments were tried upon the larvae from this batch of 

 eggs, to ascertain if possible the exact periods of larval life which are 

 most sensitive to the influence of the food-plant, as gauged by the 

 persistence of effects after the change to another food-plant which 

 tends in an opposite direction. Most of the larvae were used in this 

 series of experiments. A large number (forty or fifty) were fed upon 

 ordinary apple, and about an equal number upon Salix rubra. At 

 the end of the first stage a certain number (six) were shifted from 

 apple to S. rubra, and an equal number from S. rubra to apple, and 

 so with each succeeding stage. Thus if the larvae had lived there would 

 have been the following groups when they were full fed : — 



1. Fed upon ordinary apple during stage 1, and S. rvbra, stages 2-5. 



2. „ „ „ stages 1-2, „ „ 3-5. 



3. >, j> jj ,j 1~3, „ 2—5. 



4. „ „ „ „ 1-4, „ stage 5. 



5. „ „ „ „ 1-5. 



And again, 



1. Fed upon S. rubra during stage 1, and upon ordinary apple, stages 2-5. 



2. „ „ „ stages 1-2, „ „ „ 3-5. 



3. „ „ j, 1-3, „ „ „ 4-5. 



4. „ ,, „ 1-4, „ stage 5. 



5. „ „ „ „ 1-5. 



I think there is no doubt that a careful comparison of these ten 

 groups (which would in all cases have been kept separate as soon as 

 their food was changed) would have very completely answered the 

 question of which the solution was sought in this series of expe- 

 riments. I have given an account of these experiments — although 

 they yielded no results owing to the unfortunate and altogether 

 exceptional season — because it is likely that future work on these 

 lines will be successful in throwing some light on this difficult subject, 

 and because it is to be hoped that others may be induced to assist 

 in these investigations. 



Series V. 



Eggs were laid by a female moth bred from a larva which had 

 been fed during 1884 for the whole period of larval life upon Salix 

 rubra, and which was rather on the yellow side of an intermediate 

 variety. (The larva was one of those mentioned in the paper quoted 



m 2 



