LYCAEDA ARION. 309 



consequently the markings show more clearly out from the ground- 

 colour. The hind margin of the hindwings is more frequently obsolete 

 (ab. caeruleo-marginata, n.ab.) than in any other race (unless possibly 

 in var. amurensis, of which w 7 e have seen too few specimens to enable 

 us to judge, though in one example this obsolescence is complete and 

 in another almost so.) The Cotswold variation has been exhaustively 

 worked out by le Chamberlain (Ent., xli., p. 202). though one or two 

 •of his names are synonyms. This race is certainly duller than the 

 Cornish as a whole, though some specimens of his ab. imperialu would 

 seem to be quite light ; this fine ab. also occurs among the Cornish 

 examples. Very small specimens seem commoner here than elsewhere. 

 Apart from this fact we find no difference between the Cotswold 

 examples and the few We have seen from Huntingdonshire, Bedford- 

 shire and Northampshire ; the Devonshire specimens rather resemble 

 the Cornish, though generally darker, and we should be disposed to 

 regard the British specimens as generally dividing into two races the 

 midland and the south-western, though the distinction is not sharply 

 defined, and specimens of either locality might have come (exception- 

 ally) from the other. Jn Brittany Oberthiir remarks that the specimens 

 are much like English ones, and it might have been added like both 

 forms of English ones ; elsewhere in France, as might be expected 

 from the great differences in climate and altitude, very different 

 forms are found, from the large, light form of the Riviera (var. Ugurica) 

 to the dark var. obscura of the Alps and Pyrenees, and the minute and 

 ■dark var. delphinatus of some parts of the Dauphine ; but in all these 

 •cases there is a remarkable mixture of specimens. From the Gironde 

 come pale specimens, with the black replaced by rusty brown. In 

 Switzerland there is almost as wide a range, except that the lightest 

 race, var. arcina, is only a transition to var. li;/urica, and the mountain 

 var. obscura is rarely as small as var. delphinatus. In Germany and 

 Austria, again, the range is very wide, but none of the German 

 forms are as light as those from the Riviera, though the var. laranda 

 from the South Tyrol is a light form, and the examples from the 

 Harz Mountains do not seem to be as dark as those from the Alps ; 

 the lightest German specimens we have seen are those from Frankfort, 

 these are very close to the Cornish race, but somewhat larger. Northern 

 specimens are, as a rule, somewhat dark in colour, and we should 

 suspect them of mostly belonging to le Chamberlain's ab. cotswoldensis. 

 The range of variation in Italy is as great as, or greater than, in 

 Switzerland, though outside the Alpine area the species is scarce and 

 local. The var. Ugurica was, as the name implies, originally described 

 from the Italian Riviera, though it extends along the French coast 

 and even into the Alps at Entrevaux and the Pyrenees at Vernet. 

 Very varied forms come from the Southern slopes of the Alps, even 

 including the almost entirely black ab. aldrovandus, the original speci- 

 men of which was taken at the foot of Vesuvius. We have little 

 information about the forms from the Apennines, where it is scarce, 

 but Wheeler reports (Ent. flee, xxii., p. 281) that at Palena and 

 lioccaraso, at about 4,000ft., in the Abruzzi, the specimens are rather 

 small and dark, but not approaching the blackness of var. obscura ; 

 these again are no doubt to be referred to ab. cotsicoldeiisis. Mrs. 

 Nicholl's specimens from Bosnia include a wide range of variation, 

 there is a $ obviously of the ab. unicolor, a £ referable to ab. aldro- 



