442 Mr. G. A. K. Marshall on 



and experiments s|tecially undertaken in order to test the 

 suggestion would be of much interest. That the hypothesis 

 is im[)robable is further shown by a long series of experi- 

 ments (hitherto only ])ub!ished in abstract in the Report 

 of the British Association, Manchester, 18S7, p. 7G3) 

 which I conducted in 1887 with lizards and the highly 

 insectivorous marmoset. Large numbers of the imai^ines 

 of Vanessa io and V. urticcV were made use of, and I came 

 to the decided conclusion that both were somewhat \\n- 

 jialatable. They were certainly only eaten when tiie 

 insect enemies under observation were hiuigry. Now the 

 strongly cryptic under-side of both species associated with 

 a fairly-conspicuous upper-side renders them in every 

 way comparable to the dry phases of Precis. The results 

 of my experiments suggest that if Vanessa urlinv appeared 

 on the wing in the teeming organic environment of Africa 

 in the wet season — with far more enemies but an even 

 greater preponderance of palatable insects — it would be to 

 its advantage with its present degree of unpalatability to 

 acquire a conspicuous under-side coloration, and thus to 

 ensure easy recognition and rejection with comparatively 

 little loss of life by experimental trials. 



The considerations set forth above suggest what will 

 probably hereafter be proved to be true, that a degree of 

 unpalatability associated with a conspicuous appearance 

 in the tropics will often appear associated Avith a cryptic 

 appearance in the Holarctic Belt as well as in those areas 

 of the tropics in wdiich for special reasons the amount 

 and variety of insect life is greatly restricted. 



It is suggested on pages 475 to 477, that this is the 

 interpretation of the loss of much of the aposematic ap- 

 pearance of Limnas rhri/sijypiis, var. lingii, on desert areas 

 in the tropics. 



To return to the seasonally dimorphic Ethiopian species 

 of tlie genus Precis, if the two phases have been proiluced, 

 as is here contended, by natural selection working in 

 opposite directions because opposite kinds of ada})tation 

 are advantageous in the very different organic environ- 

 ments of the wet and dry seasons, the (juestions as to the 

 Avay in which the change is actually determined, and as 

 to the existence of any kmd of susceptibility to external 

 influences connected with the seasons, are still unanswered. 

 The considerable amount of labour devoted by Mr. Marshall 

 to the solution of this ]iroblem has up to the present 



