the colours of certain Lepidoptera. 



315 



Experiment I. 

 Dark Surroundings. 



June 6. — The lot of 9 : 

 2 had been sent away June 

 4 ; the remaining 7 were 

 spinning or just about to 

 spin. They remained ex- 

 tremely dark to the end. 



The lot of 7.-2 had been 

 sent away June 4 ; 2 were 

 spinning ; the remaining 3 

 were quite dark, like the 

 lot of 9. 



June 10. — Of those left, 

 1 spun and the remainder 

 were preserved. 



Experiment II, 

 Green Surroundings. 



June 6. — Lot of 9 : 1 of 

 the darker larvre spun up ; 

 no further notes of the 

 other. They remained com- 

 paratively dark to the end, 

 but not like the larvae of 

 Experiment I. 



Lot of 2. — These very 

 light larvae were both spin- 

 ning up. 



Experiment III. 



Green Surroundings 



in the dark. 



June 8. — 2 spun np ; 

 now that the larvae were 

 mature their tint was 

 unchanged. They all 

 remained much darker 

 than II., but far nearer 

 these than the larvie 

 of I. 



These results confirm those of the previous year, and 

 show the great susceptibility of the larvae. The effects 

 seem to have become prominent somewhat suddenly 

 after about 12 days' exposure to the conditions of experi- 

 ment. 



Experiment III. proved that the larvae are far more 

 strongly affected by dark surroundings in a strong light 

 than by darkness. This result has been confirmed in 

 other species, and may be considered as established (see 

 especially experiments on Amphidasis hetularia in 1892). 



By far the most important result, however, is found 

 in the fact that the susceptibility to green surroundings 

 was not diminished by the fact that the parent larvae 

 had been made dark by dark surroundings in the previous 

 year. The comparison between I. and II. leaves no 

 doubt on this point. The rather less complete results in 

 the lot of 9 in II. were probably due to the eftect of these 

 large larvae upon one another. When less crowded the 

 effects were more marked. 



There is evidence, then, so far as it can be relied on 

 in one generation, that these marked characters, acquired 

 in a normal manner, and very early in the life of the 

 parents, are not transmitted to their offspring, even in 

 the form of a tendency or bias in one direction rather 

 than another. 



2 b2 



