NOTES FROM THE PYRENEES. 53 



themselves easily to description ; there is, however, at least one differ- 

 ence that is fairly entitled to specific rank. On the 2nd abdominal 

 segment there is, in M. liinaedacti/la, a strong hook, rising up above 

 the anterior hair, and giving the armature of this segment much the 

 same character as that on the 4th and following segments. In M. 

 tuttodactjila there is no such hook, the armature is wanting much as in 

 the 1st abdominal segment. In one specimen I found a slight pro- 

 jection here, and, in a few J/, lunaedactijla, the hook is a little less 

 developed, but there is still a gap between those specimens that most 

 approach each other. The pupa of M. liniaedacti/la is often green, I do 

 not think I found one of M. ttittodactijla of that colour, on the other hand, 

 they varied from pale grey to absolutely black, not a few being of that tint. 

 I do not remember ever to have seen a black pupa of M. lunaedactyla, 

 though some are fairly dark. It may be noted that the pupa of 

 Stant/eia (Tutt) siceliota, on Ononis natriv, varies from pale to quite 

 black. [By the way, I can find " -S'frt?i//<?m"^' nowhere but in Tutt's 

 British lyepidujitera, vol. v., p. 492, where it occurs in a quotation from 

 me, but placed there by Mr. Tutt, not by me.] I quite agree, however, 

 that siceliota is abundantly distinct genericaliy from paludum. 



Plate VI. 

 Photographs of Two Specimens of Ancillaky Appendages of Marasmarcha 



tuttodactyla x 45. 



Fig. 1. — On slide (and compressed) laterally. 



Fig. 2. — Opened out, and dorsal portions separate. In fig. 1, the two hairs 

 are seen coiled round on one clasp ; on the other, the short, straight hairs, directed 

 across (apparently, in the flattened specimen) to those of the other side, are not very 

 apparent at first view ; the vacancy in the area, that is occupied in the other clasp 

 by the circling hairs, is very obvious, as it is also in fig. 2, in which the two short 

 hairs are lying along the axis of the clasp. 



Plate VII. 



Fig. 1. — Ancillary appendages of Marasmarcha agrorum x 18. 



Fig. 2. — Ancillary appendages of M. tuttodactyla x 18. Allowing for the 

 slightly different attitudes of the specimens, the differences between figs. 1 and 2 

 are slight; the more robust shaft of the clasp in M. agrorum is, apparently, a real 

 difference. 21. tuttodactyla is seen better in pi. vi. 



Fig. 3. — Ancillary appendages of Marasmarcha faunaxlS. These are 

 symmetrical, but smaller and more delicate in structure than those of M. lunae- 

 dactyla (pi. viii., fig. 3) ; the two hairs are not held down in a hollow, into an 

 exact circle as in that species. 



Plate VIII. 



Fig. 1. — Ancillary appendages of Marasmarcha asiaticaxlS (forwarded by 

 Staudinger as Plutyptilia asiatica). One clasp is folded over, so as to make the 

 very bold double hairs appear to curve in different directions in the two clasps ; 

 they are, however, quite symmetrical. 



Fig. 2. — Ancillary appendages of Marasmarcha colossa x 18. The double 

 hairs are asymmetrical ; the short hairs on left clasp well shown ; the spurs on 

 clasps appear to entitle this species to separate generic rank. 



Fig. 3. — Ancillary appendages of Marasmarcha lunaedactylaxlS. This fig, 

 shows the double circular hairs, symmetrically placed in each clasp ; the appen- 

 dages are decidedly larger and more robust than in M. tuttodactyla or M. fauna. 



* This is so. Stangeia was created for siceliota to separate it genericaliy from 

 both Buckleria and Trichoptilus. It was intended to deal with this in Nat. Hist. 

 Brit. Lep., vol. v., but was quite overlooked. It is a very distinct genus, with type 

 siceliota. — Ed. 



