191*.] o;u) 



equally common every yeai\ although those of alternate years were not 

 immediately descended from each other. 



In the case before us, how do the races of the alternate years 

 avoid invading the territory of each other. In ordinary cases, one is 

 inclined to suppose that some climatic or other accident has temporarily 

 decimated one of the alternate sets, and does so sufficiently often to 

 keep up the appearance of alternation, just as we know that ordinary 

 single-brooded forms are often wiped out in certain localities for often 

 a great number of consecutive years. 



When full-fed the larva of 0. pyrenaeUa fixes the case amongst 

 the upper parts of the herbage ; covers it with a coating of silk, which 

 forms a sort of cocoon with the case inside and secures it in its position. 

 The male examples then moult, extruding the cast skins from the free 

 end of the case. This cast skin usually falls to the ground, but 

 occasionally remains entangled in the opening or amongst the silk of 

 the cocoon. 



The head and plates of the larva are now of quite a delicate 

 structure compared with their previous strength and solidity. The 

 females do not do so, but moult to pupa from the ordinary larval skin. 



The males fix themselves up for pupation quite a week or ten days 

 before the females, possibly more. 



This extra prepupal moult of the males was first observed by 

 Zeller in 1847 and has since been shown to occur in a number of 

 Macropsycliids. There is some reason to suppose that it takes place 

 in all the Macropsycliids, in so far, that so far as I am aware, its 

 absence has not been shown in the case of any species. I think it has 

 not previously been demonstrated in the instance before us. 



It is very difficult to form any opinion as to the use of this extra 

 moult, and why it should occur in the males and not in the females. 

 There is no evidence that it occurs in the females except rarely as an 

 aberration. 



It is precisely similar to the extra moult that occurs in Graci- 

 lariids ( Phyllocnistis) and to that in Hastula hyerana. In the latter 

 case the larva in its prepupal instar does some actual spinning within 

 its cocoon, and spends several months aestivating thereafter, and we 

 may well suppose that the prepupal instar is fitted for this resting 

 condition in a special way, that is absent in the feeding larva. 



The larva3 of these Psychids seem to be identical in the two sexes 

 as regards skin panoply, though the male may be rather smaller. 



