440 , BOTANICAL GAZETTE [may 



It is unnecessary to follow the relations of the vascular system 

 of the petiole and tendril of Smilax in detail, since this has already 

 been done by Queva. The general equivalence of the vascular 

 structure in these organs is shown in fig. 3 B. It may perhaps be 

 objected that the occurrence of a ring of bundles in the Smilax 

 tendril, instead of being an indication of homology with the petiole, 

 may merely represent that skeletal arrangement which best enables 

 the tendril to perform its special function. The weight of this 

 criticism, however, is lessened when we realize that in the leaf 

 tendrils of another member of the Liliaceae {Gloriosa superba L., 

 fig. 6), which approach radial symmetry in their external form quite 

 as closely as do the tendrils of Smilax, the plan of the vascular 

 system remains purely dorsiventral, although the amount of xylem 

 increases considerably in passing from the lamina to the • tendril 

 (figs. 7, 8). The leaf tendrils of Fritillaria verticillata Ledebour, 

 and of Polygonatum cirrhifolium Royle, resemble that of Gloriosa 

 in structure. 



Accordin 



lamina 



lamina 



upper region of the petiole. The thickened tip (fig. 5 ap) which 



some 



petiolar apex. 



xpanded 



interpretation 



petiole -\- pseudo-lamina. On the other hand, in Gloriosa the 



seems 



the 



Petiolar tendrils, with the blade entirely, or almost entirely, 

 aborted, are not unknown among dicotyledons, as, for example, the 

 first leaves of a species of Tropaeolum described by Darwin (3)- 

 In connection with Smilax, it is perhaps significant that a tendency 

 to torsion of the petiole or leaf-base is by no means rare among 

 monocotyledons. Cases are known, for instance, in the Com- 

 melinaceae, Amaryllidaceae, Zingiberaceae, and Gramineae, as 



the 



most 



in the extremest exam 



merely leads to resupination of the leaf: but in Gloriosa this 



