88 BRITISH LEPIDOPTERA. 



1. Some effect was produced in 8 days in young larvae of Geomctra papilionaria. 



2. Some effect was produced in 8 days in young larvas of Catocala electa. 



3. Much effect was produced in 12 days in young larvae of Crocallis eliiifinaria. 



4. Much effect was produced (about) 14 days in young larvae of TMelanippe montanata, 



5. Much effect was produced in 11 days in young larvae of Catocala elocata. 



6. Much effect was produced in 13 (or less) in young larvae of Hemerophila abniptariit. 



7. Much effect was produced in 17 days in young larvae of Rumia luteolatn. 



8. Much effect was produced in 8 days in, .young larvas of Amphidasys betidaria. 



He then summarises the main facts relating to his observations as 

 follows : (1) When carefully watched for, the changes are sometimes 

 seen to occur quite suddenly (C. elimjuaria and R. luteolata). (2) The 

 effects cannot be reversed by reversing the surroundings for a short time 

 (C. dinijuaria, H. abruptaria, A, betidaria). (3) When the conditions 

 are uniform the environment does not necessarily destroy individual 

 variability, but the most powerful forms of environment, when applied 

 to highly sensitive species, very nearly do away with it. If, however, 

 the environment be mixed, there does not appear to be any instinctive 

 knowledge leading the larvas to rest only on appropria;e objects. 

 Thus, if they have become green, and are beyond the power of change, 

 they will nevertheless rest on brown twigs in preference to green leaves. 

 The habit of these Geometrids is to rest upon twigs under any circum- 

 stance, and this is probably the reason why a small proportion of twigs 

 will produce a great effect. (4) Contact, or at least the closest 

 proximity, is required to effect the change. (5) Although larvae of 

 A. betidaria are so much more susceptible to brown surroundings when 

 these are mixed with green, there were no exceptions among 105 larvae 

 which, in 1889, became green among leaves and shoots. In the case 

 of larvae of Ji. luteolata and A. betidaria, there is direct evidence of the 

 power being efficient in concealing the wild larvae. (6) The larvae 

 are, probably, chiefly sensitive at the time when they quit the leaves, 

 and first begin to rest on the twigs. (7) Darkness does not produce 

 so great an effect as black surroundings in a strong light (A. betidaria, 

 R. luteolata, C. elinguaria). (8) Overcrowding tends to produce dark 

 larvae (A. betidaria, R. luteolata). (9) As might be expected, the 

 effects produced on the larvae do not influence the colour of the moths. 



There can be no doubt that these changes in the larval coloration 

 are such as would effectively bring about the concealment of the larvae. 

 In the majority of the larvae experimented upon, the only possible 

 change appears to be from dark brown to light brown, or to greenish- 

 brown. Larvas of the latter colour are, however, much less con- 

 spicuous on leaves than the darker forms, although they are not nearly 

 so well protected on the dark twigs. Poulton thinks that when the 

 larvae of any one of these species hatch upon a part of a tree where 

 there is a great abundance of young green shoots, their susceptibility 

 would certainly lead them in the direction of concealment. It by no 

 means follows that the power is useless in certain species, because it 

 leads to more perfect results in others. Concerning the latter, no one 

 who has once seen the larvae of A. betularia a,ud R. luteolata upon their 

 food-plants in the field, can have doubt about the meaning of the 

 changes in colour which they undergo. 



Eleven larvae of Rumia luteolata placed in green surroundings, on 

 July 7th, 1890, by Miss Gould, produced 1 brilliant green, 2 lighter 

 green, 6 duller shades of green (whilst two disappeared). Eleven other 

 larva) from same batch, among dark-coloured surroundings, produced 3 



