of certain lepidopterous larva, and pupa. 53 



ourselves, to the effect that must be produced upon other 

 animals, and by observation we can often prove that the 

 inference is correct. I have before shown that if a 

 marking be repeated so as to form a regular series the 

 latter can be carried on by a mere suggestion of the 

 further repetition of the same marking without pro- 

 ducing any suspicion of imperfection. In this case we 

 cannot doubt that the advantage gained is by bringing 

 about a similar absence of suspicion in those animals 

 which are the natural enemies of the organism. But in 

 granting this we are committed to the belief that the 

 mental organisation of animals is similar to that of 

 man in this respect ; that both are equally deceived by 

 the suggested marking, because in both there is the 

 same subjective readiness (so to speak) to continue the 

 series. In the same way it is most probable that 

 animals, in searching after a prey that is protected by 

 colour, &c., are likely to continue finding those that they 

 first come across in any particular hunt. And this fact 

 would be of especial value as against those enemies that 

 seem to systematically work over a whole tree. 



To put the argument perfectly plainly, I think all will 

 admit that the larvae upon a tree stand a better chance 

 against their various enemies if they belong to two 

 differently coloured species (both well protected) than if 

 they are all the same. So dimorphism is an advantage 

 when the divergence in colour is quite complete. Again, 

 a large number of larvae of one colour attracts attention 

 and multiplies the chances of detection, and increases 

 the danger for all. This is especially true of the larvae 

 of small moths laying their eggs upon isolated trees, for 

 there are likely to be a great many larvae upon one tree 

 under these conditions. Without pressing the analogy 

 too far between man and the natural enemies of larv^, 

 I think we may admit that larval dimorphism may be of 

 direct value to a species in the manner indicated above. 

 If this is the case the dimorphism will continue as a 

 permanent condition unless there is a great difference 

 between the protective values of the two forms. 



I obtained twenty-eight pupae from eggs from a 

 female bred from the brown form and fertilised by 

 a male similarly bred. Of these twenty-eight exactly 

 half were brown and half green. In this case the 

 number of brown forms is much above the average, 

 and it is probable that the proportion could be increased 



