( xxxix ) 



are simply phases of the same species, the more markedly 

 uncinate form G. zaneka belonging to the dry season, and the 

 more deeply coloured form G. nip>alensis to the wet. 



" We have then, in passing from west to east, a chain of 

 forms as follows : — G. cleobuh (Canaries), G. maderensis 

 (Madeira), G. chopatra (Mediterranean), G. taurica (Levant), 

 G. nipalensis and G. zaneka (N. India), G. acuminata (China 

 and Burma), G. aspasia (E. Siberia), and G. maxima (Japan). 

 These appear to stand to one another in the relation of more 

 or less distinct geographical races ; in some cases, as in that 

 of G. cleobule, doubtless deserving the name of representative 

 species. In passing from G. cleobule to G. cleopatra the orange 

 flush diminishes in area but becomes heightened in colour ; 

 from G. cleopatra to G. taurica the area of the flush remains 

 the same, but its intensity is lessened. In each of the forms 

 ranging east of the habitat of G. taurica there are, as we 

 have seen, two phases, perhaps seasonal in significance ; one 

 of which phases, resembling G. cleopatra more closely in 

 contour, recalls it also by the reappearance, at least in the 

 far eastern forms, of an indication of the orange flush. 



" What, it may be asked, is the relation of G. rhavini to 

 the other forms of the genus 1 If after studying the Asiatic 

 forms in their 'seasonal' phases, we were suddenly con- 

 fronted with G. cleopatra and G. rhamni for the first time, I 

 believe we should be inclined to regard the two latter as also 

 phases of each other, parallel with G. nipalensis and G. zaneka, 

 or with the two forms of G. acuminata. But we know that 

 whatever may be the relation between G. rhamni and 

 G. cleopatra, it is not one of regular seasonal alternation. 

 Are they to be regarded as completely distinct 1 If so, this 

 would seem to carry with it a presumption that the supposed 

 seasonal forms of G. aspasia, G. maxima, etc., have been 

 wrongly associated ; in which case the strongly acuminate, 

 ' wet-season ' forms, with G. zanelca, must be looked upon as 

 geographical representatives of one species, viz., G. rhamni; 

 and the less acuminate, ' dry-season ' forms, with G. nipal- 

 ensis, as the same of another species, viz., G. cleopatra. 

 What evidence is there of a synepigonic kind 1 The state- 

 ment met with in popular books, that G. rhavini and G. 



