( cv ) 



(and var. calhmx), Satnrnia pavonia, Drepana falcataria, 

 Z. punctaria (both broods), S. bilunaria, S. tetralunaria, 

 S. lunaria (both seasonal forms of these three species), and 

 Ennomos autumnaria. With several other species I have 

 found no appreciable difference of this general character 

 caused by temperature, and in those I have mentioned it 

 varies very greatly in degree. In some of these, e.g. the Selenias, 

 especially S. tetralunaria, the difference is striking, being 

 equal to that between a dead leaf in a dry August and the same, 

 moist and darkened by a wet winter ; corresponding, in fact, 

 very closely to the difference in appearance between the summer 

 and spring phases of this insect, either of which can be con- 

 verted into a colouring approaching that of the other by the 

 appropriate temperature, a few days being sufficient for the 

 purpose where it is a case of high temperature which greatly 

 shortens the pupal stage. I do not hesitate to say that in 

 many species much the most conspicuous difference caused by 

 special temperature, experimentally applied, is that which is 

 caused by this "direct" effect during the latter part of the 

 pupal period, and a given temperature must have the same 

 effect in nature if the only difference between the two cases is 

 that in the one it is applied by man in a laboratory, and in 

 the other by nature in a field. 



In using the term " direct," I do so by way of dis- 

 tinguishing it from that previously spoken of as based on 

 hereditary predisposition. It is not at all the same thing as, 

 for example, the direct effect of heat in coagulating albumen, 

 or of light falling on a photographic plate, or developing green 

 chlorophyll, or the bright colour of a budding flower. The whole 

 effect on the colour of the imago is caused before its colour begins 

 to appear through the transparent skin of the pupa. Remove 

 the pupa into a normal temperature at this time, and the 

 colour comes out just the same as if the removal had not taken 

 place. It must therefore be caused through some vital 

 process * acted on by the temperature at a somewhat earlier 

 period, though in general later than that at which the pattern 

 is determined. 



* Of a nature, I suggest, resembling that which Professor Poulton has 

 shown to be concerned in the colouring of pupae by the surroundings 

 of the larva before pupation. 



