180 Troiisari iuiis. 



IG legs, but tliose on segments 7 and 8 do not seem to be fully developed at 

 this stage. 



On the 20th December the larvae were about | in. in length, pale greenish- 

 brown in colour, with dark reddish-brown lateral line. At this stage they 

 unfortunately died. They were feeding very sparingly on the leaves of 

 what I thought at the time was toetoe. but which I afterwards found was 

 pampas-grass. 



In May, 1910, I took a number of these larvae on the toetoe at Makara ; 

 they were from 1 in. to 1| in. in length. The larva feeds on the leaves by 

 night, retreating during the day to the crevices at the base of the leaves, 

 where they are well protected from enemies. 



The full-grown larva is dull brownish-green in colour, sometimes tinged 

 with reddish-brown, especially on posterior segments. The dorsal and sub- 

 dorsal lines are very narrow, but fairly well marked ; dull white in colour, 

 faintly edged with red or reddish brown. The lateral line is somewhat 

 indistinct, white in colour On it are situated the spiracles, which are dull 

 cream-colour edged with black. The lateral line is often edged with small 

 brown blotches situated above the spiracles, and on the anterior segments 

 these blotches are sometimes joined to form a broad, faintly marked upward 

 edging to the lateral line. The integument, especially on the dorsal surface, 

 has a number of fine white branching veins, and on each segment are a 

 number of minute black dots from which spring short brown bristles. The 

 prolegs are of the same colour as the body, edged with a large number of 

 dark-brown hooks. The head is horny, amber in colour, mottled and netted 

 with brown. The number of legs is 16. Length when full grown, 1| in. 



The larva now makes its way into the flower-stem preparatory to 

 pupating. It enters the stem, and eats its way through the soft interior, 

 forming a chamber 2 in. or 3 in. long between two joints. It now loses 

 its green colour, and changes to a pale dull brownish-yellow, the dorsal 

 surface often strongly tinged with pink. This pink tinge becomes very 

 marked as the time of pupation approaches. The larva spends some weeks 

 in the stem, and before changing to a pupa cuts a neat round hole through 

 the stem, near the top of its chamber, leaving only a very thin .film of the 

 outermost layer intact. It then retires to the bottom of the chamber, and 

 in a few da}'s changes to a pupa, which rests on the old larval skin, head 

 upwards. 



The pupa is very robust, and is at first light Ijrown in colour, but soon 

 becomes very dark brown and shiny. 



After about six weeks the imago emerges, and, breaking its way through 

 the thin film of leaf covering the exit from its chamber, crawls out and clings 

 to the stem till its wings have expanded and hardened sufficiently for it to 

 fiy. The emergence usually takes place betAveen 7 and 9 o'clock in the 

 evening. 



The perfect insect is about from October to January. It is very sluggish, 

 and I have never seen one in a state of nature, though I have spent a good 

 deal of time in a locality where the larva is fairly common. 



The larva of this insect is sometimes attacked by a dipterous parasite. 



