IH THE BRITISH ISLANDS. 135 



stigmata, a bent one behind the posterior stigma, and another subapical 

 one marked, as in the others, with black dots, which are also occasion- 

 ally obsolete " ( British Moths,' pp. 135-136); whilst Guenee writes : 

 " It would be impossible to divide the very numerous but little 

 characterised varieties of this Tceniocampa. It is sometimes of a 

 yellow-ochreous, or very reddish and almost unicolorous, at other times, 

 of a testaceous-grey, very powdered. The black spots (on the subter- 

 minal) are sometimes in three pairs, sometimes reduced to the central 

 pair, at other times altogether absent. But these differences are never 

 simultaneous. It is therefore difficult to define the types of geminata 

 (us) and bimaculata(us) of English authors " (' Noctuelles,' vol. v., 

 p. 357). 



There appears to be no simple way of sub-dividing the varieties 

 except as follows : 



1. Ground colour greyish- white = var. pallida. 



2. Ground colour dark grey = var. grisea. 



3. Ground colour pale reddish-ochreous = munda, Esp. 



4. Ground colour reddish or ferruginous = var. rufa. 



Each of these groups can then be roughly sub-divided into the 

 following sub-varieties : 

 a. Without dots and indistinct transverse lines = immaculata, 



Stdgr. 



b. With two dots and indistinct transverse lines = bimacnlatus, Haw. 

 c. With two or more dots (generally six) and distinct transverse 



lines = geminatus, Haw. 



We should thus get immaculata-pallida, bimaculatus-pallida and 

 geminatus-pallida, as the sub-vars. of var. pallida, and so on for grisea 

 and rufa. Of munda, since sub-var. geminatus with six spots and 

 transverse lines represents the typical form, we should get immaculata 

 and bimaculatus as reddish varieties of the type. Here we get a case 

 where the different ground colours of a species present parallel varie- 

 ties in the different colours. Hiibner's fig. 166 (called by error lota) 

 agrees with Esper's type, but has four spots instead of six. I would 

 also draw attention to the fact that it must not be assumed that all 

 specimens of sub-vars. bimaculatus-pallida, -grisea and -rufa, have 

 suppressed transverse lines, central shade and stigmata, although it is 

 generally so in my very long series, nor that geminatus-pallida, -grisea 

 and -rufa have them distinct. In my series, some of the palest forms 

 with the transverse lines totally suppressed, have as many as six or 

 eight spots, and thus belong to the geminatus form. 



a. var. immaculata, Stdgr. This variety is thus described by 

 Dr. Staudinger : " Alis anterioribus maculis nigris ante marginibus 

 nullis" ('Catalog,' p. 114). Taking the ferruginous spotted form as 

 the type, the unspotted form, corresponding in colour &c. with the 

 type is Staudinger's immaculata. I have the pale reddish spotless 

 form from Hereford, Clevedon and Brecon. 



/3. var. bimaculatus, Haw. This red variety with one pair of 

 blackish spots in the centre of the subterminal line, is thus described 

 by Ha worth : " Alis pallide ferrugineis punctis duobus posticis con- 

 tiguis nigris." "Alas anticse subferruginea3 concolores stigmatibus 

 fere obliteratis, exterius reniforme et ad latus interius saturatius : 

 anterius annulare obsoletissimum margine solo vix conspicuum. Alae 



