NOTES ON EREBIA GAVARNIENSIS, N.SP. 275 



fifteen years ago, yet, during the last few years, it has been (whenever 

 mentioned in the magazine articles) spoken of as var. caecilia. The 

 folio \ving extract is from Mr. Elwes' paper. 



" ab. caecilia, Hiibn., 213-14, Text, p. 85. Alps. 

 {Sii})ra fere vel rariiis tota )iii/)a.) 

 var. constans, caecilia, Dup., i., 49, 6, 7. Pyrenees. 

 (<? <'i 2 supra tota niijra, ? infra iiiiniis fusca nutata.)" 



He also states (p. 171), that he is not giving it a new name, not 

 being sure whether manto does not exist in the Pyrenees. Thus, while 

 suggesting separation, Mr. Elwes seems to class this race as a 

 Pyrenean fonu of caecilia, while in his definition he does not even 

 mention the extraordinary difference of colour in the 3' underside. 

 Seitz has made a curious mistake in applying the name c(mHtans, 

 Elwes, to a Pyrenean form of )nanto, since the latter 'gave no name 

 to it at all, but merely stated that while caecilia is an aberration in the 

 Alps., it is a constant variety in the Pyrenees. 



Dr. Chapman, without naming it, separated this race from manto, 

 on the strength of the genitalia, and though this is not always sufficient 

 proof by itself (as was shown not so long ago by certain species of 

 Melitaea), in this case there are so many other constant characteristics, 

 practically sufficient m themselves to give it specific rank, that one may 

 take it as conclusive. 



The following description of this Pyrenean Erehia, is made from 

 28 specimens in my collection ; sixteen $ s and twelve ? s, all taken in 

 the Val d'Ossue, Gavarnie (where it is exceedingly abundant), on 

 July 20th and 22nd, 1911. 



$ . — Slightly larger than E. manto, varying from 46mm. to 48ram. 

 {inantu 40mm. to 44mm.). 



Upperside : Ground colour velvety black, entirehj without markings. 



Underside : Same black ground colour as the upperside, but with- 

 out the velvety gloss. No trace whatever of the mahogany suffusion, 

 (which covers the whole surface of the underside of the wings in 

 manto), giving it the dull blackish-brown appearance, which is so 

 striking a characteristic in this species. Occasionally a verij small 

 rust-coloured spot at the apex of the forewings. Only two out of the 

 sixteen ^ s in my possession show this, and in these it is confined to 

 the underside. The fringes of both fore- and hindwings very much 

 less conspicuous than in manto, owing to their being of the same shade 

 as the ground colour of the wings. 



? . — Very constant in size, and larger than manto: 50mm. {nutnto 

 ? varying from 42mm. to 48mm.) 



Upperside : Usually entirely black, without bands or spots, as in 

 the (J ; but occasionally with two small black apical eye-spots present, 

 the spots in these instances are somewhat smaller than the corres- 

 ponding ones in manto. The ground colour on the whole darker. 



Underside : Frequently completely without markings, though never 

 so unicolorous as in the J ; more usually with one or two veri/ small 

 yellow spots on the hindwings ; no basal spots ; on the forewings a 

 square rust-coloured spot at the apex ; no eye-spots. The whole tone 

 of the ground colour much duller than in )nantu. 



In the neuration the following are the principal differences : — 



Pyrenean race, forewings : All the nervures slightly more curved 

 than in manto. Discoidal cell a little more than half the length of the 



