164 Mr. Poulton's notes in 1885 iqjou 



The free anterior pointed end of the ovum is marked 

 off from the rest by a distinct Hne, and after development 

 begins it remains attached to the young larva as a black 

 and shining head-shield. Development began on August 

 11th, the very day on which the host ceased feeding and 

 darkened. On the next day the host spun a cocoon, 

 which was destroyed in order to watch the development 

 (subsequent attempts to spin being also frustrated). 

 There seems to be some significance in the development 

 of the parasites only beginning when the larva had 

 ceased feeding, for it is likely that a much earlier de- 

 velopment would irritate the larva and prevent it from 

 attaining the normal size. A somewhat earher develop- 

 ment has been previously recorded (DeGeer, &c.). I have 

 never seen the eggs attached to the larva before the last 

 stage, and I have not heard of any such instance, although 

 DeGeer implies that this may be the case when he states 

 that the eggs are so firmly implanted that the larva 

 cannot get rid of them by changing its skin. (This 

 statement needs confirmation.) There is another ad- 

 vantage gained by the habit of laying eggs on the large 

 larvae in the last stage, for they are then far less active, 

 and have often lost the means of defence especially 

 directed towards warding off the attacks of ichneumons, 

 the power of protruding the pink flagella. It seems 

 that the ichneumons have taken advantage of this weak- 

 ness, and it would be extremely interesting to observe 

 the varying success of attempts made upon many 

 individual larva? possessing the defensive power in 

 different degrees, or even to observe if those upon which 

 the eggs are found have always lost the power. It is 

 very likely that such observations would show that 

 natural selection sets with a strong current towards the 

 survival and ultimate predominance of those individuals 

 which retain the power ; for the ichneumons are abun- 

 dant, and are the chief enemies of the larva, while those 

 individuals which are detected and fail to drive oft" their 

 hymenopterous foes are doomed, for they can do nothing 

 in the way of self-protection when the eggs are once laid. 



The fixation of the larva by both extremities and its 

 early growth have been described especially by Newport. 

 It will be very interesting to make sections of the larval 

 skin with the eggs attached to show the method of 

 fixation (which is described by DeGeer as due to the 



