292 Dr. T. A. Chapman on What the 



taking it to be in the fourth instar, and that it must be 

 in the normal hibernating instar, the third adopted by so 

 many other species. 



Mr. Frohawk's discovery solved a most remarkable 

 problem, that our previous ignorance practically prevented 

 our realising as existing, viz. How many more instars had 

 the larva when going into hibernation to pass through to 

 become full grown ? and the startling answer was. None ; 

 it was already in its last instar. 



It further compelled us to regard as even more difficult 

 and insistent, the question of how this minute larva, one- 

 eighth of an inch long, grew to its mature dimensions of 

 well over half an inch long and correspondingly thick; in 

 fact, a larva suitable to produce a butterfly as large as 

 L. arion. 



Did it feed up in autumn, in winter or in spring, or in 

 two or all of these seasons ? What was its food ? 



That it was somehow associated with, ants seemed the 

 most acceptable basis for a hypothesis, but still there was 

 something to be said in favour of a vegetable diet. These 

 questions have been discussed by various people in various 

 journals, etc., at considerable length, with the object of 

 thinking out the best way of investigating the problem. 



It would not be profitable to discuss in detail the various 

 ideas suggested, since it happens that by a combination 

 of perseverance, and good luck almost comparable with 

 Mr. Frohawk's, I am able to give, not by any means the 

 whole history, but certainly its chief fact and keynote. 



On May 14, 1915, on pulling up plants over a nest of 

 Myrmica scabrinodis var. sabideti and disturbing the soil 

 at a point close to overhanging heather, etc., a larva of 

 L. avion was found; it seemed to be amongst loose earth 

 that the ants had worked over, and if not actually in the 

 ants' nest was within less than an inch of ground actually 

 occupied by the ants. 



Unfortunately in the rough process necessary in dis- 

 turbing plants and soil the larva suffered an injury. 



Its length was ITO mm. and thickness about 3'0 mm. 

 Its colour a pale earthy flesh-colour, no trace of green 

 anywhere, and the impression it gave was, that it must 

 be a concealed feeder. Mr. Frohawk, as well as I, con- 

 sidered that it was by no means full grown. There were 

 visible some dark contents of the posterior extremity of 

 the alimentary canal, shining through the ventral surface, 



