394 BOTANICAL GAZETTE [December 



their length, after thickenings had been formed, to at least fifteen 

 times. Whether this was due to the formation of new cells or 

 to the elongation of those present was not determined with cer- 

 tainty, but I am strongly inclined to believe that both things 

 occur. One finds scattered short cells with very close spirals 

 not yet well marked, beside them may be long cells with spirals 

 so long drawn out that their spiral nature is almost lost, the cell 

 walls perhaps even disappearing, the whole showing disorganiza- 

 tion. In fact, all the stages in the ontogeny of a vessel may be 

 found in a single longitudinal section of a forming loop. This 



makes it appear that the spirals form while the cells are not very 

 greatly elongated ; that the elongation of the cell takes place 

 by the widening of the part of the wall between the spirals, 

 due either to growth of the cell or to tension arising from the 

 growth of the surrounding tissues ; that when the cell has become 

 extremely elongated it collapses and disorganizes, while several 

 cells beside it have meanwhile formed spirals, thus preserving 

 the vascular continuity. Since there is only one bundle in a 

 stamen, the nutrient solutes would find it necessary to pass to 

 the anthers by way of the tips of the hoods. There is no direct 

 connection between anther and pedicel. The tissue of the inner 

 side of the filament forms aerenchyma, thus being able to keep 

 pace with the curving of the bundle without leaving any great 

 hollow. The mature hood is then largely composed of a tissue 

 resembling the spongy tissue of a leaf, but without chlorophyll 



The horn is carried up some distance by the growth of the 

 hood, but has no connection whatever with the fibrovascular 

 bundle {^fig. i8), as one is led to think from one of Schacht's 

 figures. It seems to be built by the division of the cells near 

 its apex, and the later transformation of its tissue into aeren- 

 chyma, for its whole interior is filled with very loose tissue 

 (/^- -^P)- I^ Acerates, as the name implies, there is no horn. 

 However, the bundle passes into the hood as in Asclepias, while 

 in Cynanchum (3) it simply curves into the base of it. The 

 direction of the bundle in relation to the hood may prove to be 

 of value in determining relations among the Asclepiadaceae. 



