VIII] TENDRIL-CLIMBERS 109 



In the cases hitherto cited the tendrils are only incom- 

 pletely differentiated : one sees at a glance they are leaf- 

 stalks or prolongations of midribs, &c., and in Smilax the 

 two long thin tendrils flanking the base of the leaf-stalk 

 are stipules. 



But the best idea of the evolution of these leaf-tendrils 

 is obtained by comparing a series of Leguminous plants 

 Vicia, Pisum, Lathy rus, &c. where almost every stage 

 can be traced between cases of pinnate leaves with three or 

 more pairs of normal leaflets, beyond which the rachis is 

 produced as a tendril bearing one, two or more pairs of 

 tendrils in place of leaflets, e.g. Vicia sativa, Pisum sativum, 

 and cases where, as in Lathyrus Aphaca, all the leaflets 

 have disappeared and the rachis alone persists, transformed 

 into a long slender tendril bearing two large stipules 

 below. 



We have only to suppose the stipules to disappear 

 also, and the place of the leaf is taken by a simple tendril 

 only, a case realised in some Bignoniaceae, Cucurbitacese, 

 Cobcea scandens (Polemoniacese) and Mutisia (Composite). 



So far the tendrils considered are all more or less 

 obviously leaves, or parts of leaves, as to their morpho- 

 logical nature ; but it is a very common occurrence to find 

 true tendrils in such positions, or bearing such structures 

 (leaves, flowers, &c.) as to warrant our concluding that they 

 are altered branches in origin. One of the best studied 

 examples here is that of the common Grape Vine (Fig. 49), 

 or that of the Virginia Creeper (Fig. 50), where, partly from 

 their position opposite the leaves, and partly from the fact 

 that small scale-leaves are borne on them with branches in 

 their axils, their branch characters are clearly recognisable : 

 taking the internode below the tendril, and tracing it up 

 to the latter, we have the tendril terminating this segment 

 of the shoot, and the next internode above may be regarded 



