THE TWINING OF STEMS 35 



a mechanical obstacle determining the production and nature of the per- 

 manent coils formed by the growing circumnutating apex. 



In the absence of distinct circumnutation no twining is possible and 

 hence the seedling stem of the scarlet-runner, the shoots developed from 

 the rhizome of the hop, and in general all the branches of twiners which 

 are unable to circumnutate sufficiently are also unable to twine 1 . Hence 

 also etiolated stems of Tropaeolum majus t and of Polygonum fagopyrum 

 gain the power of circumnutating and of twining at the same time. 



Circumnutation, and with it twining, cease when a twiner is rotated 

 upon a klinostat so that the action of gravity is eliminated, while as the 

 result of the plant's orthotropism the coiled younger portions may untwine 

 and straighten 1 . It still remains possible, however, that gravity may act 

 directly upon the process of twining, as well as indirectly by influencing 

 circumnutation. At the same time it is evident that the contact with the 

 support exerts no stimulus capable of preventing the uncoiling of the young 

 shoot on a klinostat. 



Under such circumstances, however, the stem of Cuscuta loses not only 

 its power of circumnutation, but also the contact irritability which it 

 exhibits under normal conditions. Rubbing the stems of Cuscuta and 

 Lophospermum scandens with a solid body suffices to produce an irritable 

 curvature, but not in the case of the stem of Phaseolus or of other twining 

 plants. The same negative result is also obtained when one side of the 

 shoot is repeatedly rubbed, or when permanent contact against an edge 

 of wood is assured (Darwin), or when the circumnutating shoot presses 

 against a rod attached to an appropriate turn-table (de Vries 2 ). 



The normal symmetric circumnutation is not sufficient to produce 

 twining, for if it were, the horizontal or sloping free end when attached 

 to a support at its base would continue to circumnutate and coil around 

 a horizontal or inclined ideal axis (Baranetzsky's asymmetric circumnuta- 

 tion 3 ). By attaching a piece of india-rubber tube to a retort-stand it is 

 easy to show that in this way no twining about an erect support could 

 be produced. It is evident, therefore, that sure and regular coiling involves 

 a regulation of the growth and circumnutation of the growing apex. 



According to Schwendener 4 this is produced by grasping movements, 



1 Bowiea vohMlis twines when horizontally rotated on a klinostat, according to W. Voss (Bot. 

 Ztg., 1902, Originale, p. 231), if illuminated from the apical side. This plant is more strongly 

 heliotropic than other climbers, and its circumnutation is dependent upon illumination. 



2 H. Mohl (1. c., p. 112) considered twining to be the result of contact irritability, but Palm 

 (Ueber d. Winden d. Pflanzen, 1827, pp. 20, 97) of rotary nutation. Darwin (1. c., 1875, pp. 16, 39) 

 and later de Vries (1. c., p. 321) showed the absence of any contact irritability, which has, however, 

 again been brought forward by F. G. Kohl (Jahrb. f. wiss. Bot., 1884, Bd. xv, p. 327). Ambronn 

 (Zur Mechanik des Windens, 1884, 1, p. 32, Repr. from Sitzungsb. d. sachs. Ges. d. Wiss.) has shown 

 that Kohl's experiments are inconclusive. 



3 Baranetzsky, 1. c., pp. n, 16, distinguishes between symmetric and asymmetric nutation. 



4 Schwendener, Gesammelte bot. Mitth., Bd. i (1881), p. 405; (1886), p. 441. 



D 2 



