49 



lichen. I was not surprised to find the perfect insect resting on 

 them, but I was somewhat so at finding a recently vacated cocoon or 

 two ; there certainly was no vestige of a cocoon on these particular 

 posts a year before. I again visited the bridge wall, where I had taken 

 so many pupae the summer before, to see if my rough handling had in 

 any way affected the abundance of the species at that particular 

 spot, and was pleased to find that cocoons were there as thick as ever. 

 Two that I brought away produced, the one a fine full-sized imago, 

 the other one of the most diminutive that I have seen. So small was 

 the pupa that I fully expected it to produce B. per/a, but that wall 

 still remains without a record of that species. Biyophila per/a was 

 proportionately common, and, like B. mura/is, showed a considerable 

 amount of variation. 



The broods of Acidalia marginipu aetata appeared to overlap one 

 another more than usual ; this was hardly to be wondered at in so 

 favourable a season. On July 24th I found five examples, all a good 

 deal worn, and several in similar condition on August 6th ; while on 

 the 7 th, of some dozen examined none were in bad order, and the 

 majority were so fresh that they could not have been many hours 

 out of pupae. Unfortunately from this date fresh easterly breezes set 

 in, precluding any further opportunity of finding them in their usual 

 resting-places. 



It is many years since I have met with Me/anippe galiata so 

 commonly as I found it last year, nor do I remember to have 

 previously noticed so great a variation in the density of the colour of 

 the central band. From a worn specimen taken on July 25th, 

 evidently one of a second brood, I obtained ova which hatched on 

 August 2nd, and all the larvae but one pupated between the 18th and 

 20th of that month, the imagines emerging between September 5th 

 and 10th, this being evidently the third emergence of the year. Ova 

 were again obtained from a pair that emerged on the 7th, and they 

 hatched on the 20th, and like the former brood were kept out of 

 doors. About this time cool unsettled weather set in, and by 

 October 15th the larvae had grown so little, and appeared to be in 

 such an unsatisfactory condition, that to save their lives they were 

 brought into a warm room, and within a week had increased fully 

 three times in size. With the advent of milder weather at the end 

 of the month they were again put out of doors, and continued to 

 grow rapidly. The majority went down between the 7th and 1 2th of 

 November, the last on the 18th. The average time that the second 

 brood of larvae takes in feeding up in ordinary seasons appears to be 

 from four to five weeks. In the present case the second brood 

 occupied eighteen days from hatching to going down, but the time 

 taken by the third brood exceeded eight weeks, although their 

 growth was considerably hastened by their being brought into a 

 warm room at a critical time. 



What the effect of this third brood would be in a wild state it is 

 impossible to say, but there is some reason for believing that it would 



5 



