526 Notes in 1887 upon lepidopterous larve, cc. 
and associated borders were developed to a far greater 
extent than in any of the others, was now removed to a 
separate cylinder, in order to test its relation to the 
other larve in the next stage. This larva is represented 
in outline on Plate XV., fig. 8, x 5°8. The shagreen 
tubercles are omitted, except in profile: all the dark 
marks represent pigment. The dark mark between the 
2nd and 8rd thoracic segments probably belongs to the 
next stage, and is seen through the transparent cuticle. 
The oblique white lines above the subdorsal were less 
distinct in these light larve, but they could be made out 
by comparison with the more decided appearances in the 
other set. They were best seen by looking at the larve 
with the naked eye and from a little distance. 
The Sphinz-like attitude is rarely, if ever, assumed in 
either of the two first stages: the larve still stretched 
themselves along the midrib on the under sides of leaves, 
but especially along the leaf-stalks and stems of the food- 
plant. When disturbed they wriggled from side to side, 
and often fell from the food-plant. I have never before 
met with larve which manifested so little disposition to 
wander. On Sept. 28rd I noted that no larva had ever 
been found off the food-plant, and this in spite of 
continual disturbance during examination. 
The larve were fond of eating the leaf-stalks and 
stems as well as the leaves of the food-plant. Towards 
the end of the stage, as the skin became stretched, the 
larve began to assume a somewhat glistening appear- 
ance. During the resting-period before ecdysis the 
ground colour became lighter, and of a more trans- 
parent and yellowish green: the change is almost 
certainly due to the comparative absence of food from 
the digestive tract at this time. 
The mature larve were about 14 mm. in length when 
moderately extended, but the largest could stretch to 
15 or even 16mm. During the resting-period the 
length was very uniformly 14 mm. 
I now add data from which the length of the stage 
and of the resting-period can be calculated. The 
average length of the stage appears to be about eight 
days, and the variations are not more than one day on 
each side of this. 
