204 DIVERSITY OF THE SEXES 



is not impossible that this fold can be opened at 

 will by the insect, although I can hardly under- 

 stand how it can be done in the skippers ; and it 

 would then become conspicuous and probably an 

 attraction to the butterfly's mate. But what pos- 

 sible advantage can there be in partially or wholly 

 concealed androconia, scattered separately over 

 the wing ? In some blues they are exceedingly 

 scarce, numbering not more than one to a hundred 

 scales, and the exposed surface of this not one 

 tenth that of the scales about it. One might 

 search an hour with the microscope over an un- 

 ruffled wing and overlook it. Remember, too, 

 that it is the merest speck of dust in a dust-heap. 

 Does the sight of these creatures surpass our 

 power of vision with the microscope ? The theory 

 of sexual selection, proposed by Darwin, appears 

 to fail here, just where it should most aid us ; and 

 it seems as if we should be forced to conclude that 

 there is a principle underlying these phenomena 

 which may yet be found to cover both them and 

 those which have been used in support of the in- 

 genious hypothesis of Darwin a law of variation 

 for its own sake.* 



* Since this was written, Fritz Miiller has in various places de- 

 clared these androconia to be scent- scales (Duftsehuppen), having 

 an odor presumably attractive to their consorts. That the males 

 of many butterflies exhale odors has long been known ; it is very 

 marked in our Mountain Silver-spot (Argynriis Atlantis), which 



