170 THE AMERICAN NATURALIST [Vol. XLIV 



more probable that the female is attracted by maleness 

 alone and not by beauty. The immediate effect of court- 

 ship on the female is the stimulation of her sexual desire 

 by recognition of a male, not first the arousing of an 

 esthetic sense and second the arousing of sexual feelings. 

 On my interpretation the case in Astia may be explained 

 simply and without assumption: the niger-iovm is se- 

 lected by the female because he is more different in color 

 and structure from her than is the other male form, con- 

 sequently is more quickly recognized as a male; this is 

 what determines her, and not color or ornaments. The 

 significance of peculiarities in color, structure and move- 

 ments of the male lies in insuring quick sex-recognition, 

 not in arousing esthetic feelings. The case of Astia is 

 the only one known in which the female seems to exert a 

 choice between males, dimorphism of males being rare in 

 spiders. 



We conclude, accordingly, that the male in visual court- 

 ship is not actuated by a conscious effort to exhibit his 

 peculiar beauties, and that the female does not select 

 males by an esthetic sense. Courtship by the male re- 

 sults simply because fear is mingled with his desire; and 

 probably the female will accept the first male who courts 

 her, and makes himself recognized as a male, at the time 

 when she is physiologically desirous. Sexual selection 

 in the meaning of Darwin, accordingly, and in opposition 

 to the views of the Peckhams, has probably played no part 

 in the evolution of the secondary sexual differences of 

 spiders. 



V. The Nature and Use of the Secondary Sexual 

 Differences 



We are not concerned here with the question of the first 

 origin of secondary sexual differences, t. e., whether they 

 have arisen as gradual fluctuations, as Darwin held for 

 the most part, or from acquired traits becoming inherited 

 on the view of Cunningham (1900), or from sudden mu- 

 tations as Morgan (1903) has suggested, for there is no 

 observational evidence from which we can reason. We 



