1902] A MORPHOLOGICAL STUDY OF ASCLEPIADACEAE 403 



radiating from the base of the ligule in Isoetes (12) have any 

 relation to those in the ovules of Asclepias? In any event, in 

 the ovule one would expect the formation of leptome rather than 

 of hadrome from such a cause. This throws us back upon Miss 

 Benson's (i) suggestion that this is an indication of a degenerate 

 chalazal vascular strand once extending into the nucellus. Judg- 

 I ing from the form and arrangement of the cells between the end 



of the vascular bundle, which appears later, and the sac, the 

 tracheid is just about where the former, extended, would meet 

 the latter. That a primitive feature so nearly lost, even among 

 the lower Archichlamydeae, should appear among the Asclepia- 

 daceae seems remarkable, and suggests a doubt as to its primi- 

 tive character. 



The fertilization of the Asclepiadaceae has long been a sub- 



ject of interest. 



) 



pollen tubes in the ovary and noticed that they were attached to 

 the ovules, but like all other botanists of his time he supposed 

 the tubes reached the ovary by way of the caudicules and the 

 head. According to Brongniart (2), Ehrenberg rediscovered the 

 tubes, but considered them permanent structures. He observed 

 that the tubes issue from the side of the pollinium, thus disprov- 

 ing the theory of fertilization through the caudicules, and leav- 

 ing the location of the stigma in doubt. Two years later Brong- 

 niart (2) properly located the stigma, and traced pollen tubes from 

 the pollinia into the style. Since he wrote before insect pollina- 

 tion had received much attention, he accounts for the develop- 

 ment of the pollen tubes within the anther by the passage of a 

 hypothetical liquid from the corpusculum into the pollinia by 

 way of the caudicules. Whether pollen tubes do sometimes form 

 within the anthers, or whether Brongniart was mistaken in the 

 location of the bursting pollinia is still a matter of doubt. Corry 

 (5) says that flowers artificially pollinated with their own pollen 

 remain sterile, although the tubes penetrate the style. He 

 observed one nucleus, probably the tube nucleus, enter the tube, 

 and traced the tubes to the ovary. Chauveaud (3) reports that 

 in Cynanchum he found pollen grains putting out tubes within 

 the anther. Gager (8) reports that in AXornutixXx^lvHo^ nucleus 



