366 



It maybe well right hereto look a little more closely into the minuter 

 characteristics of the Yucca tiower at this stage of its development, 

 that we may understand more fully the action and influence of the moth. 

 In my first article, published some twenty years ago, announcing the 

 discovery of Pronuba and its action on Yucca pollination, I was strongly 

 inclined to the idea that the act of j)ollination had some compensating 

 inducement to the moth, aside from the impelling instinct of perpetua- 

 tion of the species. At that time it was supposed that the stigmatic 

 liquor was nectarian, and the conclusion was justifiable that the moth, 



Fig. 65.— Nectar apparatus of Yucca: a. longitudinal section of pistil, with duct (d) and gland (g)\ 6, 

 cross section about middle, showing same parts: o, still more enlarged cross section of nectar appa- 

 ratus; e, structure of septal gland— after Trelease; A, longitudinal section of top of pistil, showing 

 stigmatic tube (s) ovarian cell (oc). ovule (o), funiculus (/). placenta (p), and tibro-vascular tissue (/d). 



attracted to it for feeding purposes, would incidentally induce pollina 

 tion. On this view of the matter it did not require a great stretch of 

 the imagination to conceive that the pollen might also incidentally ac- 

 cunnilate in the spines, and that the vigorous action of the head that 

 had been noticed might even be considered as an effort to get rid of the 

 incumbrance while feeding. In those days I was more imbued with the 

 common notion that lower creatures are impelled for the most part un- 

 consciously to their acts. Twenty years of study and experience have 

 only served to prove the acts of Pronuba the moie unselfish and with- 



