CHAPTEE III. 



TEKDBIIi-BEAiaSBS. 



Nature of tendrils — ^Biqhoniaoe^, various species of, and their different 

 modes of climbing — Tendrils which avoid the light and creep 

 into crevices — Development of adhesive discs — Excellent adapta- 

 tions for seizing different kinds of supports — Polemonuoe^; — 

 Cobsea scandens, much branched and hooked tendrils, their maimer 

 of action— Leguminos^ — CoMPoarra! — SxiLACEM—Bmilca) mpera, 

 its inefficient tendrils — FuMAKiACEa: — CorydalU daviculata, its 

 state intermediate between that of a leaf-cUmher and a tendril- 

 bearer. 



By tendrils I mean filamentary organs, sensitive to 

 contact and used exclusively for climbing. By this 

 definition, spines, Looks and rootlets, all of which are 

 used for climbing, are excluded. True tendrils are 

 formed by the modification of leaves with their petioles, 

 of flower-peduncles, branches,* and perhaps stipules. 



* Never having had the oppor- (1.) Plants supporting themselves 



tunity of examining tendrils simply by their branches stretched 



produced by the modification of out at right angles — for example, 



branches, I spoke doubtfully about Chiococca. (2 ) Plants clasping a 



them in this essay when ori- support with their unmodified 



ginally published. But since branches, as with Securidaea. 



then Fritz Muller has described (3.) Plants climbing by the ex- 



(Joumal of Linn. Soc. vol. ix. p. tremities of their branches which 



344) many striking cases in 8outh appear like tendrils, as is tlie case 



Brazil. In speaking of plants according to Endlicber with 



which climb by the aid of their ffelinus. (4.) Plants with tlie r 



branches, more or less modified, branches much modified and 



hestatesthat the following stages temporarily converted into ten- 



of development can be traced : drils, but which may be again 



