( xcviii ) 



stx'uctui'e is removed, the escape and evaporation of the 

 odoriferous substance may be assisted. My observations do 

 not enable me to explain the exact mechanism of this process, 

 supposing it to take place. But the special distribution 

 of tracheae to these patches is a fact which can be verified 

 without difficulty, though I believe it is now noticed for 

 the first time. 



The plume-scales, on the other hand, even when collected 

 into more or less definite patches, are seldom provided with a 

 cover. The conjecture may be hazarded that their insertion 

 among other scales, usvially larger and longer than themselves, 

 ipay prevent a too rapid dispersal of their odour ; it may also 

 perhaps be the case that the di.sc so often possessed by the 

 plume-scales, but invariably absent from scent-scales of the 

 other type, may act as a reservoir ; the scent escaping a little 

 at a time through the comparatively narrow footstalk. The 

 sharp bend so often to be seen in the course of the footstalk 

 may impede the passage of the scent under ordinary circum- 

 stances, and if we may proceed a little further along the path of 

 conjecture, we may venture on the hypothesis that the peculiar 

 fluttering about the female, which is a noticeable feature in 

 the courtship of some Pierines, may have the effect of causing 

 a certain erection of the scales, with a conseqvient straighten- 

 ing of the footstalk, and more plentiful liberation of the 

 perfume. After this digression into the realms of guess-work 

 we will retv^rn to the region of sober fact. 



When definite " brands " are present, they are, of course, 

 available as an aid in distinguishing the sexes ; they have, in 

 fact, long been so employed. But I think it has not been 

 noticed, or at any rate recorded, that in very many instances 

 among the Pierines, even where the plume-scales are scattered 

 over the general surface and are not collected into definite 

 areas, their presence imparts such a peculiar roughened char- 

 acter to the texture of the wing as seen with the naked eye, 

 that the sexes can be distinguished at a glance by this feature 

 alone, without recourse to any other means of recognition. 



It was suggested many years ago that the characters of 

 wing-scales might be used for purposes of diagnosis. The 

 suggestion was dismissed by the late Professor West wood 



