( cxxxiv ) 



narrow and straight-sided, showing no sign of spatulation. 

 In every species of Gmi.oris known to me, but especially in 

 G. brassicae, the scent-scales are abundant. 



The plumule of Ganoris crncivora is much like that of G. 

 rapae ; those of G. oleracea, G. melete and G. ajaka are more of 

 the najri tj'^pe. G. oleracea is remarkable in having the cornua 

 ending in sharp spines, which give a perfectly distinctive 

 appearance to the plumule of that species. A certain approach 

 to this condition is seen in G. ajaka, where many of the 

 plumules have a rudimentary spine on one or both cornua. 

 Forms included under the name of G. jnelete have scent-scales 

 showing differences which may be specific. One such form 

 has accessory discs Avhich are enormously large in proportion 

 to the laminae. 



With respect to their scent-scales, the members of the genus 

 Synchloe show a strong resemblance to many forms of Euchloe. 

 The laminae, for instance, in S. c.hloridice are strikingly like 

 those of Uuchloe creitsa, exhibiting a similar expansion towards 

 the blunted distal margin. The discs in both genera are 

 small, and circular or oval. Sfjnchloe heUica and aS*. johnstonii 

 have plume-scales of the like character, while those of S. 

 f/lauco7iome differ chiefly by the sharpness of their apex. The 

 laminae in S. dapUdice vary considerably in breadth ; they 

 bear much resemblance to those of E. auHonia and E. beleviia, 

 having like them a somewhat short apex. In both genera, 

 Euchloe and Syncldoe, the laminae are chai'acterised by a well- 

 marked longitudinal ribbing, whicli is in obvious relation with 

 the fimbriae. 



The genera Jkdtia and Phtdia appear to be devoid of plume- 

 scales. I'heir ordinary scales are peculiar, being unusually 

 short and broad ; in which respect they are in some degree 

 resembled by those of the genus TatochUa. 



The scent-scales of this latter genus are not altogether dis- 

 similar from those of iSijnchloe, though generally very much 

 larger. They show interesting specific differences ; the laminsi, 

 which is basally squared in T. stigmadice and rounded in T. 

 xantliodice, being shaped like the blade of a ma.son's trowel in 

 T. theodice, and having convex sides in 7\ autodice. 



Tn Pontia soracta the plume-scales, which are very numerous, 



