156 THE STRUCTURE OF FLOWERS. 



and never occurring without it, leaves no hereditary effect as 

 in the case of galls * and of the thickening of the tissues of 

 some climbers after they have caught and clung to a foreign 

 body, such as the petioles of Clematis,^ and the hooked 

 peduncles of Uncaria (Fig. 46). In other cases the effect 



has become hereditary, and may 

 then be regarded as a specific 

 character. These diff'erences are 

 well seen in the tendrils of Ampe- 

 lojpsis Jiederacea as compared 

 ■with those of A. Veitchii. In 

 the former there are no traces of 

 the adhesive " pads " at the ter- 

 minations of the slender hooked 

 tips of the branching tendrils, 

 Fig 46 -Climbing peduncle of rncana, Until contact with the surface of 

 (after'Trtubt" '"''^'"^ ^ '"''^''' ^ Wall has occurrcd. On the 



latter species, however, the pads 

 are in course of development before any contact has taken 

 place just as the aerial roots of Ivy begin to appear before 

 contact. It is therefore reasonable to conclude that the 

 effect of contact has become more or less hereditary in 

 the latter Japanese species, though not in the American. 



These tendrils behave exactly like the clasping roots of 

 Orchids, Ivy, etc., as well as the so-called " roots " of Lami- 

 naria, Cutleria, etc. Indeed, the way in which subterra- 

 nean root-hairs fix themselves to particles of the soil is by 

 essentially the same method. The irritation caused by con- 

 tact aided by moisture excites the cell- wall to grow out into 

 protuberant processes, which enables it to adapt itself to the 



* I have examined a considerable number of galls, and can quite 

 corroborate M. Prillieux, who has shown how the normal tissues become 

 hypertrophied {Ann. des Sci. Nat., ser. 6, tom. ii. (1876), p. 113). 



t See Climbing Plants, fig. 1, p. 47, and fig. 4, p. 74. 



