MOVEMENTS OF IRRITATION. 499 



stream of water against an expanded leaf, the well-known closing 

 movements very rapidly ensue. Tendrils are not irritable to all 

 kinds of mechanical shock ; they only react when in their sensi- 

 tive zone discrete points of limited extent are subjected simul- 

 taneously or in sufficiently rapid succession to push or pull. 

 We have therefore to distinguish between impact stimulus 

 [Stossreiz], to which, e.g., the leaves of Mimosa react, and contact 

 stimulus \_Gontactreiz], which cause the movements of tendrils. 



If a support, e.g. a wire or a thin stick, is placed in the neigh- 

 bourhood of tendrils of Sicyos or Cucurbita, the irritable organs 

 soon come into contact with it owing to the nutation movements 

 already referred to. The contact acts as a stimulus, and the ten- 

 drils curve. Owing to this curvature, fresh points of the tendrils 

 come into contact with the support, causing further stimulation, 

 and thus, rapidly or slowly, the tendril comes to wind round the 

 support (see Fig. 166, tendril branch, d). When a tendril has 

 grasped a support, notable changes very rapidly appear in the 

 portion of the tendril between the support and the plant. This 

 part of the tendril, viz., as represented in Fig. 166, forms cork- 

 screw-like coils, and for purely mechanical reasons so-called 

 reversal points originate, which separate spirals wound in opposite 

 directions (see Fig. 166, at TF). I found, e.g., that a Sicyos tendril, 

 which had grasped a support at about 4 p.m. on July 1st, was 

 already coiled corkscrew fashion in the part stretched between 

 the supports and the plant on the morning of the next day ; 

 reversal points were also already present. I found further that 

 a Cyclanthera tendril, which had grasped a support, at a high 

 summer temperature already exhibited the first spiral turns be- 

 tween the support and the plant at the end of eight hours. These 

 first turns formed in the immediate neighbourhood of the point of 

 attachment of the tendril to the support. The development of 

 the coils proceeds in the free part of the tendril from the tip 

 towards the base. 



Tendrils of our Cucurbitacese which have not grasped a support 

 exhibit coiling phenomena like organs which have already attached 

 themselves, but we can readily satisfy ourselves that while the coils 

 form rapidly in attached tendrils, they develop only very slowly 

 in free ones. These facts leave no room for doubt that the ac- 

 celerated coiling in attached tendrils is due to the contact stimulus 

 to which they have been subjected, and it is moreover clear that 

 propagation of stimulus must play an important part in the 



