SOCIETIES. 93 



develop pigment would tend to darken the general colour by throw- 

 ing up the dark ground colour ; and as Mr. Merrifield's specimens 

 exhibited at the same time, not only a failure to develop pigment, but 

 also a failure to produce the normal membranous structure of the wings, 

 it became apparent that the cause of one failure was the probable cause 

 of the other. He therefore assumed that the action of temperature 

 was indirect, and that the result vvas directly due to the insects being 

 unable to carry out their normal functions during development. With 

 regard to the moths it was different. They had essentially two colours 

 in their organism, so to speak. In Selenia illustraria these were purple 

 and reddish-ochreous, due to seasonal dimorphism, the pigment 

 ranging from reddish-ochreous, through red to purpl^ ; in Bombyx 

 quercus yellow and red-brown, due to sexual dimorphism, the 

 females being yellow and the males deep red-brown. In these 

 species, too, simple chemical experiments showed that the ground colour 

 was white. But here again the changes in colour, brought about by 

 temperature, were directly due to the normal elaboration or non- 

 elaboration of the pigment-forming material and- only indirectly with 

 the temperature that prevented this elaboration. It was also not so 

 intimately connected with failure in membranous development. The 

 genetic sequence of the colours in illustraria was yellow, red, purple. 

 .A. rapid development under high temperature produced the ochreous or 

 red form, whilst a longer and slower development was necessary for the 

 development of the purple coloration, so that a more complete usage of 

 the pigment-forming material could be made, This genetic sequence 

 in the development ot colour, as occurring in nature, he said he had 

 dealt with at length in the Introduction to vol. ii. of The British 

 Nocliice before referred to ; and he cordially supported the vote of thanks 

 proposed by INIr. Fenn to Mr. Merrifield for bis kindness in coming 

 forward to read the paper. Mr. Dobson stated that on one occasion 

 he had a large number of larvae of Nofodonta dicicEoides, these were 

 exposed to a very high temperature (naturally), and pupated under this 

 high temperature, the result was that the pupje were ill-developed, the 

 resulting moths malformed and cripples. Mr. Merrifield agreed with 

 many of Mr. Fenn's observations, and thought most of them consistent 

 with the results obtained in the course of his experiments as reported 

 by him. He would not expect to get such great differences in colouring 

 by the application of temperature alone as might be found in ordinary 

 collections where the different varieties were the result of a combination 

 of many influences and chiefly heredity. His object had been to test 

 the effect of temperature by itself, and with that object he had operated 

 on broods as uniform in all other respects as could be. It was almost 

 always absolutely necessary in experimentation to apply more or less 

 artificial conditions in order to isolate the insect experimented on from 

 the mixture of causes always found in the operation of nature. But 

 his moderately low temperatures were, in fact, often the natural outdoor 

 temperature, and though for "forcing" he had generally used a 

 temperature of about 80°, he had reason to think that there was little 

 or no difference in effect where the temperature was several degrees 

 lower, so as to approximate to what was often experienced for several 

 days together in summer. In any case there could be no doubt that 

 in the species principally operated on by him, temperature alone, 



