AGKIADES THETIS. 335 



found at the bases of the wings. Gillmer, discussing some Moiavian 

 examples, notes (Ent. Zeits. Guben, xix., p. 73), that the black- brown 

 2 s are often more or less scaled with blue, and the orange ante- 

 marginal lunules vary from a complete band across all the wings, to 

 almost entire obliteration on the forewings. Trautmann states (Int. 

 Ent. Zeit., ii., p. 162) that the 2 s vary much in the Jena district, the 

 reddish spots differing in size and brilliancy, in some reduced to nil, 

 in others very strongly developed. Verity observes (Bull. Ent. Soc. 

 ItaL, 1904, pp. 138-139) that, at Lucca, some of the $ s of the later 

 brood are more or less covered with blue. Staudinger says (Hor. Soc. 

 Ent. Ross., xiv., p. 244) that the specimens of this species, taken in 

 Amasia, in May-June 1875, are little different from the German ones, 

 the 2 s seldom tinged strongly with blue, only one being sufficiently 

 blue to reter to ab. ceronus ; Miss Fountaine adds that the 2 s were 

 quite typical in Amasia and at Brussa, in May, 1893, but Mann says 

 that the ab. ceronus also occurs at the latter place ; the only 2 , from 

 Brussa, in the British Museum coll., is brown, with slight traces of 

 blue on the hind wings, strongly developed orange-red lunules on 

 thehindwings and weak ones on the forewings (lunulata), whilst a $ from 

 Amasia shows blue scaling extending to the orange lunules on the 

 forewings, and along the inner margin of the hindwings, with bright 

 blue lunules edging the orange internally on all four wings, the orange 

 being more weakly developed on the fore- than on the hindwings (cuneo- 

 lunulata). ABorjom 2 is brown, the hindwings with strong, the forewings 

 with weak, orange lunules (lunulata). The French $ s are particularly 

 variable locally ; Warburg notes 2 s at Cannes as being very brilliant 

 violet-blue except on costa, and the orange spots of the hindwings very 

 bright ; Gurney reports that the 2 s taken at Digne in June, 1907, were 

 particularly blue, but a few we captured there ourselves in August, 1906, 

 had no trace of blue scaling, and were almost without orange on the upper- 

 side ; at Draguignan, in early May, 1905, most of the 2 s were dark 

 and tinged with blue, although some were brown and without a trace 

 of it. In our experience, the brown examples without blue on the 

 upperside, whether belonging to the early or late broods, appear usually 

 to be more plentifully supplied with orange above, and Hodgson notes 

 (Proc. Sth. Lond. Ent. Soc, 1908, p. 46) that the blue 2 s taken on the 

 North and South Downs, in 1907, had but little orange on the upper- 

 side. General statements concerning the seasonal difference between 

 the 2 s, i.e., those of the two broods of this species, should be accepted 

 with caution, but, on the whole, our experience leads us to believe that 

 the 2 s of the early brood are usually more blue-scaled than those of 

 the later brood, but they vary much in different seasons. In 1893, a 

 specially hot and forward summer in Britain, the 2 s at Cuxton, in 

 Kent, w r ere especially dark, and heavily scaled with blue ( = ab. caeru- 

 lescens) in June, whilst in August they were particularly brown, with 

 few or no blue scales, but with the orange specially well-developed (= 

 ab. marginata), a few only of the June captures being of this form, 

 but, in other years, in this locality, the autumn examples are as dark 

 and as heavily scaled with blue as the spring ones ; and this seems 

 especially to be the case if the second-brood is late in its time of 

 appearance. Butler has noted the difference in successive autumn 

 broods at Folkestone, and states that, in the autumn of 1902, the ? s 

 were much more dusted with blue than was the case in that of 1901. In 



