I 



A NOTE ON HYBERNATION IN LKPIDOPTERA. 169 



habit and instinct, an individual of Ai/lais nrticae may thus be con- 

 strained to commence hybernation in June, and a larva must die rather 

 than hybernate, if unable to feed up to the proper stage. It is, no 

 doubt, a matter of natural selection that tlie habit and the conditions 

 (chiefly the seasons) shall be duly co-ordinated. The process or possi- 

 bility of hybernation no doubt originated (at some remote epoch) in the 

 quiescence of protoplasm at a low temperature and under starvation, 

 but it is now something much more definite than this. 



The faculty of, or capacity for, hybernation is rather a function of 

 the protoplasm than of the organism as a whole. That it is so follows, 

 I think, from these considerations ; as a function of protoplasm it is 

 not only of very ancient date, but must have taken a long period for 

 its evolution, but, as we see it now, and if we regard it as a function 

 of the organism, it is necessary to suppose that it can be developed or 

 lost at very short notice. For instance, the cases are abundant in w^hich 

 closely allied species hybernate at different stages. The habit of hyber- 

 nating as an egg can be comparatively quickly lost and replaced by hyber- 

 nation as a half-grown larva (say Affriatlt's tin'tis and A. coridon) or vice 

 versa, or the change may be from pupa to imago [Arasclmia levana and 

 Aftlais nrticae), etc. It would seem, therefore, that, though the definite 

 power to hybernate must have been of slow evolution, the decision as 

 to the stage in the insect's life at which this power shall be exercised 

 is capable of much more rapid variation. For this, possibly a very few 

 generations will suffice, the motive for the change being that the 

 natural stimulants to hybernation be brought to bear at a different 

 stage. The following observations on Leinptilus tephradacti/la show 

 that questions of food still have a determining power, if not on the 

 commencement of hybernation, at least on the date of its termina- 

 tion, and that not merely an increase of temperature is necessary to 

 this end, but, in some larvae, at least, the protoplasm does not give up 

 the hybernating attitude till fresh nutriment is supplied to it, but once 

 the attitude is j'ielded it cannot be resumed. This last remark, as 

 more than a statement of a special case, is flatly contradicted by an 

 observation on Psdnophoras brachi/ilacti/liis, of which several hybernated 

 larvjB, instead of going ahead with their brethren on the spring 

 awakening, fed a little, moulted once (one twice), and finished by 

 starting hybernating again, to appearance, in the same instar in which 

 they had already hybernated ; a very similar occurrence is not unusual 

 in Anthrocerids. 



Daring the winter of 1909-10 I had hybernating a number of larvae 

 of [j. tephrailacti/la on plants of Snliilafio cirt/aurea in three flower-pots. 

 They were, however, somewhat neglected, and, on examining them at 

 the end of March, I found that the plants in two pots were apparently 

 dead, that iu the third, however, in fair condition. Very few of the 

 larvae were, however, dead, but several on the living plant seemed to 

 be eating a little. Having to leave home, 1 put all the larvje on the 

 living plant, and left them to their fate till May 16th. I then found 

 that the living plant had been eaten away, a few larvjB still nibbling 

 at stumps. A few larvas were dead, from starvation almost certainly, 

 several hardly grown at all, but most had moulted at least once, and 

 were somewhat advanced. The living larvj^i were at all stages up to 

 half-grown in last skin. 



I then found that, in one of the pots, laid aside as having the 



