( xcvii ) 



the lamina where the footstalk is inserted, this giving rise to 

 the fan-shaped appearance already noted. Leaving the base, 

 the channels run more or less parallel with one another to- 

 wards the distal margin, passing on into the fimbriae, and 

 finally debouching into the atmosphere by means of the 

 terminal oi'ifices of those structures. This may be con- 

 sidered the normal arrangement, but it is liable to considerable 

 modification in certain groups. 



The articulation of the plume-scale with its socket in the 

 wing-membrane has been spoken of by some writers as a ball- 

 and-socket joint. This is incorrect ; for, as has been seen, the 

 articulating portion of the scale is not a ball or globule, but a 

 flattened disc. The socket for the reception of the disc is 

 usually easily distinguishable from the sockets for the attach- 

 ment of the ordinary wing-scales by its superior size and by 

 certain other features. The shape of the socket is more or less 

 adapted in conformity with that of the corresponding disc ; 

 it does not, however, receive the whole of that structure ; a 

 portion, which may be the larger, being left outside the socket. 



The distribution of the plume-scales on the surface of the 

 wing is liable to great variation as between different groups. 

 In some instances they are scattered at regular intervals over 

 the greater part of the upper surface of both fore- and hind- 

 wing. In others they are confined to certain areas, within 

 which they may be so numerous as to produce an appearance 

 easily recognisable by the naked eye. But these accumulations 

 of plume-scales in the Pierinae seldom, if ever, become so con- 

 spicuous as the " sexual brands " seen for example in Amauris 

 or Euploea. The scales composing the brands in Bisniorphia 

 and in certain species of Colias, Terias, Catopsilia and some 

 allied genei-a are of a different character from the plume-scales. 

 As I hope to deal with these Pierine brands on a future 

 occasion, I will here only remark that they are commonly, 

 though not invariably, so placed as to be closely covered up dui-- 

 ing rest, often as by a sliding lid. This is no doubt for the 

 purpose of economising the perfume ; and it may also here be 

 mentioned that in several of these cases I have detected a 

 special supply of tracheal branches distributed to the brand or 

 scent-patch, suggesting a means by which, when the covering 



PROC. ENT. SOC. LONU., Y. 1909. G 



