156 [July, 



Abdominal, a very narrow fronl one, say width equal lo 1, then a broad one 

 carrying tubercles 1 and 3, width equal to 7, another, width equal to about 3, carry- 

 ing tubercle 2, and a posterior one nearly as wide ; these widths are only approxi- 

 mate, as they vary much according to the movements of the larva. These all end 

 at the lateral flange below the spiracle. In some attitudes this flange is invisible, 

 in others it stands up almost like a separate roll laid on. Relow this again there 

 are various rolls or flanges varying with the attitudes of the larva. 



Prolegs have a rather thick base, followed by a short cylinder and ending in a 

 complete circle of hooks, which is a little weaker (almost interrupted) on the outer 

 edge. Ihe hooks are in two rows, or rather in one row witli alternate hooks of 

 different sizes, the inner and larger being twice as long as Ihe others, about 22 to 24 

 in number. On the inner edge where they are strongest they might be described 

 as in three rows or of three sizes, total about 56. 



The anal prolegs have only the anterior margin armed with crotchets, in two 

 sizes, 40 crotchets. 



The anal comb has five large prongs each ending in a double point ; there is 

 one smaller spine at either side, and a very minute one beyond, making really nine. 

 The ends look broken or bifid. 



The anal plate is nearly circular and carries four hairs on either side. 



The head of the larva is darkest when young, in first skin black 

 to the naked eye, about third skin it is really pale. The approximate 

 width of the head at each instar is — 



1st 0'2 mm. 



2nd 0'4 mm. 



3rd 0'6 mm. 



4th 0-8 mm. 



5th 1-20 mm. 



6th 2-0 mm. 



in the 7th or sestivating instar it is just the same, or very triflingly 

 less. It seems the same in all particulars, except in being devoid of 

 chitinous brown colour or nearly so except the jaws ; it is remarkable 

 that the jaws are so well developed, as they do nothing but eat the 

 moulted larva skin, leaving the head. It is probably an instar that 

 in nearly related species is still an active feeding one. 



It may be observed that some of my larvse hatched at Eeigate, 

 and kept indoors, were already full-fed in the middle of January, 

 nearly three months earlier than they would have been at Hyeres. 

 Mr, Powell tells me at this date (January 0th, '05) plants of Asphode- 

 lus have made hardly any progress in bis garden at Hyeres. 



Pupa. 



The pupa is light brown in colour and rather slender. It is much the same as 

 a whole group of pupa? found in the genera Tortrix, Cacoecia and Pandemis, but 

 most resembles those of the latter genus. Other species in Tortrix and Cacoecia 



