504 BRITISH LEPIDOPTERA. 



1878. He notes it as somewhat smaller than L. lanestris, the forewings 

 shorter and broader in proportion, the hind wings more ample, the hair 

 of the body black-brownish grey. The forewings are blackish-brown and 

 whitish-grey mixed, the grey predominating, the blackish-brown colour 

 being largely confined to the central area and to the margins of the 

 white markings. The latter are much sharper than in the type, &c. 

 Frey observes : " The larva lives on Alnus viridis and different species 

 of Salix, in the Alps ; the pupa may not disclose its imago for many 

 years. There are fine examples in Zeller-Dolder's collection, which 

 do not differ generally from the type." Wocke notes the Alpine form as 

 occurring in the Silesian mountains. Curd observes that the larvae 

 feed also on Vaccinium uliginosum. Riihl reviews (Soc. Ent., vi., pp. 

 140 et seq.) the history of arbusculae as follows : The larva was dis- 

 covered by Freyer in July, 1842, who obtained at first only larvae, and 

 later cocoons, together with one single unemerged imago taken from 

 the pupa, as a result of twelve years' effort, and an attempt to breed 

 10,000-12,000 larvae. Bischoff was at last, in 1859, able to figure and 

 describe (J. -B.Nat. Ver. Augsburg ,xii.,ipj). 81 etseq., figs, a-h) an imago of 

 arbusculae, which had been bred by Hnatek, of Sils-Maria. Riihl goes 

 on to say that he himself had for years found nests of larvae of arbus- 

 culae, the latter usually feeding on Vaccinium. uliginosum , but failed to 

 breed them in the lowlands (although he transplanted the food-plant 

 into his garden) ; single larvae also were occasionally found on Alnus 

 and dwarf sallow, but he got no cocoons until 1889, when he bought 

 22, and the next year ten others, and from these last reared at length 

 a specimen of arbusculae in the spring of 1891. He discovered 

 thousands of larvae in 1892, and brought away about a hundred and 

 obtained cocoons. The larvae were never found below 6500ft., generally 

 from 7000ft. -8000ft.; they are very fond of moisture, the nests usually 

 being found close to the mountain streams that sprinkle the Vaccinium 

 with their spray, yet in the lowlands they require to be kept very dry if 

 they are to be reared with success, and the hotter the August sun the 

 more rapidly they come to maturity. The larvae will eat, sparingly, 

 Corylus, Salix fragilis, Sorbus, Rhamnus, but invariably die on this 

 diet, even if the food be changed two or three times a day. Salix 

 caprea is the most successful substitute food-plant, and when sleeved 

 thereon in the open the larvae will sometimes reach the cocoon (but 

 not the pupal) stage. The greatest care so far has resulted in failure. 

 Chapman notes that arbusculae (as represented in Constant's collection) 

 is a little larger than L. lanestris, the $ s especially darker, and the 

 markings in both sexes more pronounced, the lines being broader and 

 whiter. Wackerzapp observes {Stett. Ent. Zeit., li., p. 143) that the webs 

 of larvae, 1ft. -2ft. long, were found hanging from the birches on the 

 Simplon, some of the webs containing as many as 80 larvae wbich left 

 their dwellings at night to feed, returning to them by day. The imagines 

 emerged the following March (some not till the second year), and 

 differed from the German examples in having a whitish dusting over 

 the wings. He further notes (loc. cit., p. 214) that at Berisal, at the 

 end of July, 1884, full-grown larvae were to be found quite commonly 

 away from their nests, and feeding on low plants with which they are 

 not usually connected — Cynauchum vincetoxicum, &c. Standfuss ob- 

 serves that it appears to be a general thing for the pupae of this variety 

 to go over from four to six years before the emergence of the imago. 



