86 BKITISH BUTTERFLIES. 



mesothorax to 8th abdominal, marked subspiracular flange, the 

 generalised nature of the secondary hair-clothing, the specialised 

 nature (or obsolescence) of the tubercles and tubercular setae, and the 

 hiding habits it affects, forming silken tubes in which to conceal itself 

 whilst feeding. Here again, however, there are considerable differences 

 in structural details, and these, apparently, of a most important nature, 

 although at present little understood. Some of these peculiarities are 

 (1) the spiracle-like tubercles known as lenticles, (2) the arrangement of 

 the hooklets of the prolegs, and (3) the "white scaly excrescences " 

 developed on the 7th and 8th abdominal segments in certain Urbicoline 

 (Pamphiline) larvae. Of the " lenticles," Chapman, in describing the larva 

 of Erynnis (Spilothyrus) lav at her ae, notes (Ent. Rec, xv., pp. 298-300) 

 that "these lenticles may represent the true tubercles, but by no means 

 certainly so. On the abdominal segments, on the 1st subsegment, 

 a little outside the mediodorsal black line, is a raised, dark chitinous 

 circle, without any hair, and looking almost exactly like a spiracle ; 

 this might be tubercle i. On the yellow lateral flange, at its most 

 prominent portion, a little in front of the middle of the segment, is 

 another exactly similar chitinous ring, but paler, i.e., brownish in 

 colour. These from their form, colour and position, might easily be 

 supposed to be the spiracles, but are perhaps tubercle iv. A single 

 similar circle is found on the left side of the 2nd abdominal 

 segment, above and in front of the spiracle, which might be iii, but 

 no other exists in this position on any other segment. The lenticles 

 that appear to represent i, and are on the 1st subsegment of the 

 abdominal segments, exist on the 2nd subsegment of the meso- and 

 metathorax." Reference to our detailed descriptions of the larvae of 

 the British species should be made, where other facts relating to these 

 peculiar structures may be noted. In the abdominal prolegs of the 

 larva there are some structural variations in the hooklets. In other 

 butterfly larvae the hooks exist, as in the Macro-Heterocera, only 

 along the inner margin of the originally circular pad. In the 

 Urbicolids the circle of hooks usually remains in the adult larvae, as 

 in the Micro-Heterocera. In some species, for example, Cyclopides 

 palaemon, there is a simple circle, or rather an oval, with a gap on 

 the inner side ; but in many of the species, chiefly among the 

 Urbicolids {comma, sylvanns, etc.), there is a complete circle, repeated 

 in three rows, the outer row having the smallest hooks. This latter 

 form is met with elsewhere only in Hepialid larvae, and is, therefore, 

 a very archaic form of proleg. Taking Erynnis lavatherae as a type of 

 the Hesperiine larva, Chapman notes that the inner posterior margin is 

 furnished with numerous hooks, apparently alternating in two sizes ; 

 round the anterior margin they become much more sparse, so that, 

 from the middle of the anterior to the middle of the outer margin 

 they can be easily accounted as being five, hence, on the outer and 

 back margins, for about one-fifth of the circumference, they are quite 

 wanting. The anal prolegs form a disc of three-quarters of a circle, 

 the chord of which forms its posterior border ; this border has no 

 hooks, the rest carries hooks, the outer half ton large ones and 

 intermediate smaller ones ; on the inner half they are much more 

 crowded and numerous." The most remarkable structures presented by 

 the larva 1 are the " white scaly excrescences " which are present in the 

 larvae of most Urbicolids (Pamphilids). These usually snow-white patches 



