revision of the genus Argynnis. 551 



closely resembling the average Pyrenean specimens, and 

 below intermediate between -pales and arsilache. Those 

 from Greece (var. grcsca, Stgr.) are described as paler 

 below and with the fringes variegated, but this is also 

 the case in some alpine and Himalayan specimens. The 

 majority of these last, however, which have been named 

 sipora and haralacha by Moore, are much nearer to 

 Swiss than they are to Central Asiatic ones, and certainly 

 cannot be separated. 



Staudinger considers arsilache as a var. only of pales, 

 saying that the Scandinavian form, which he names 

 lapponica, is intermediate ; but Zeller says that his 

 evidence, cf. Stett. Ent. Zeit., 1861, p. 347, and 1872, 

 p. 44, tells as much in favour of their distinction as of 

 their identity. Meyer-Dur also, in his excellent work on 

 the Swiss butterflies, strongly supports the idea that 

 they are diflerent, but Frey, who had probably a larger 

 material and experience, agrees with Staudinger. A. 

 pales, however, is everywhere the typical mountain form, 

 and arsilache the lowland one : out of fourteen pairs 

 from Denmark, Sweden, Norway, and North Eussia, all 

 but four are nearer to arsilache, those from the Dovrefeild 

 being the only exceptions. 



Considering, therefore, that in Europe we can usually 

 distinguish the two forms by their markings, and the 

 difference in their habitat, and that their larvse have 

 probably different food-plants, it will only be a question 

 of individual opinion whether they are distinct species. 

 It is curious that pales, which is found over such a wide 

 area, and is common wherever it occurs, should be 

 absent, and have no near ally in N. America. 



A. chariclea is a circumpolar species, but hitherto 

 found only in isolated localities in Lapland and Siberia, 

 whilst in Labrador, British America, and Greenland it 

 is more generally distributed. It varies considerably, 

 and in the extreme north is much darker in colour 

 above (var. obscurata, M'Lachl.) ; whilst in the Kocky 

 Mountains, and occasionally in Labrador, it assumes 

 the form hoiscluvaUi, which Edwards treats as distinct. 

 I cannot, however, see any good reason for this, as there 

 is no possible line of division between the two ; and 

 Geddes found both flying together at a high altitude in 

 the Eocky Mountains. It extends to a higher latitude 

 than any other butterfly, except A. p)olaris. 



TBANS. ENT. SOG. LOND. 1889. PART IV. (dEC.) 2 Q 



