70 THE entomologist's RECORD. 



moult the orange dorsal band and yellowish tufts became less brilliant 

 in tone, the ground colour of the body less velvety, the body itself 

 harder in texture. 



When placed where the rays of the sun could reach them the 

 larVfB, otherwise of a sluggish disposition, became quite lively and 

 fed a trifle voraciously, but my subsequent observations led me to 

 conclude that the larvie were undoubtedly night feeders. The dates 

 upon which the larvie were afterwards inspected by me and the notes 

 made at such times are given below : — 



February 28rd. — ^The larva which had remained small in comparison 

 with the others, as well as another larva, were found to be dead, thus 

 reducing the number of survivors to 28. The larvae remained very 

 sluggish, but if roused would, like the larvae of Arctia caia, roll into a 

 ring and shortly afterwards crawl away at a rapid rate. 



March 2nd. — Removed the larv*, the majority of which were now 

 li inches in length, from the bottle they were in and placed them 

 upon a growing plant of dandelion enclosed in a good sized gauze- 

 covered cage standing in an airy room at the top of the house. The 

 bottom of the box up to the level of the flower-pot was filled with 

 cocoanut fibre and on top of this moss was placed. I now had a better 

 opportunity of seeing the larvae feed which they generally did at night- 

 time. Upon examining them by candle-light 1 found some of them 

 moving freely about whilst others were feeding. When aware of the 

 light those that were moving about ceased to do so, and some of 

 those that were feeding stopped feeding. 



March 9th. — Two more larv* died within the preceding seven days. 

 April 5th, three further deaths, leaving 21 larvte, and the survivors 

 continued to remain dormant during the greater part of the day, 

 resting either on the gauze or framework but always well towards the 

 ■ top of the cage. At night-time they would descend to their food-plant, 

 and, when it became quite dark, commence to feed. I noticed that they 

 had a marked preference for the unopened flower-buds of the dandelion, 

 into which they would bore for their food. The larvte by this date 

 had grown to Ih inches in length, but had not altered materially in 

 appearance. 



April 20th. — Found that three more of the larvte had died since last 

 inspection. I judged that by this time the larvai were practically full- 

 fed, and I, therefore, placed upon the moss a good supply of dead oak 

 and chestnut leaves, Mr. ]\Iontgomery having informed me that the 

 larvte required dead leaves in which to spin up. 



April 27th. — I could find seventeen only of the larvte on this date, tmd 

 concluded that the remaining one had gone down for pupation. I did 

 not ascertain that it had tictually done so for the reason thtit the larvte 

 seemed to be of such a delicate disposition that interference with them 

 at any critictil period generally meant disaster. Judged by subsequent 

 events I believe the larva referred to htid died. 



May 4th. — Found seventeen larvte still feeding. Noticed upon 

 examining them one night during the week that as many as four 

 were feeding on the same flower-head of dandelion, roughly speakmg 

 one from each quarter of the compass ; their heads being close to one 

 another. So slowly did they txppear to be feeding that I could not 

 observe, while watching them, any material diminution of the flower- 

 hetid upon which they were feeding. 



