494 TRANSFORMATION OF ENERGY 



movements in geotropic curvatures. We have regarded the compression of the 

 concave side there as a cause of this autotropism (Baranetzky, 1901) and there- 

 fore the second growth acceleration in the case of tendrils could only take place 

 if actual curvature had previously been effected. Fitting (1903) showed, how- 

 ever, that in tendrils also which are mechanically prevented from carrying 

 out their curvature, there are likewise two temporary growth accelerations 

 separated by a period of rest. It is very desirable that measurements 

 should be taken of the distribution of growth during the period when 

 curving is ceasing, since in those structures which are affected by a geotropic 

 stimulus, a growth acceleration manifests itself in the median zone (compare 

 p. 435) ; it is possible that the same relations may obtain in the present case. 

 Meanwhile it is impossible to say with certainty whether the autotropism of 

 tendrils is of a different nature or not from that already observed in cases of geo- 

 tropism. If both phenomena be identical then the cause of autotropism must 

 lie in an effort to curve, a tension of the parts concerned, and not in the first 

 instance in the completed curvature (compare Fitting, p. 612). 



Having now studied the movements of tendrils which result from a temporary 

 contact, let us turn next to problems connected with the actual encircling of the 

 support by the tendril in nature. The circumnutation of the tendril, exhibiting 

 as it does movements very similar to those seen in twining stems, must aid it 

 considerably in its efforts to find a suitable support ; we might almost say that 

 the tendril feels round about for a support to attach itself to. If the hapto- 

 tropically reacting region comes in contact with some solid body the further 

 nutatory movements enable it to place itself under conditions in which the 

 haptotropic stimulus may operate. As in the experiments described, the tendril 

 rtibs itself against the support, and forthwith at once curves, whereby new 

 regions of the tendril come in contact with the support. Since the stimulus, 

 as we have seen, is transmitted for a few millimetres on either side of the point 

 of contact, the tendril in a very short time describes a complete circle, provided 

 that the support be not too thick or too thin. If the support be of suitable 

 diameter the tendril is still unable to carry out completely the curvature 

 striven after ; a tension arises which exerts a pressure on the support, a tension 

 which may be easily demonstrated by using compressible material, such as a roll 

 of paper, as the supporting structure ; such a substance will be found to have 

 been squeezed by the tendril. The first coil is followed by a second and a third, 

 if the tip of the tendril be still free. How comes it that these coils, it may be 

 asked, remain permanently encircling the support if each incurving be followed 

 by a recurving of the tendril ? The reverse curvature may indeed be observed 

 in tendrils which have managed to grasp a support, manifesting itself in a loosen- 

 ing of the existing coil. When this reverse curvature appears, however, the 

 movement of the tendril or of the support may bring about a fresh contact 

 stimulus, which again induces incurving, so finally effecting a permanent en- 

 circling of the support. So long as no loosening of the coil takes place, no new 

 contact stimulus is possible, for the pressure exerted on the support does not 

 act as a stimulus. As we have already seen, the stimulus is transmitted from 

 the point of contact both proximately and distally. For purely mechanical 

 reasons the base of the tendril cannot exhibit any curvature, since both ends 

 are firmly fixed, to the plant on the one hand and the support on the other. 

 But if the spiral on the support becomes slack, then curvature ensues in the 

 neighbouring portions of the tendril basaUy, so that these come in contact with 

 and surround the support, pushing in front of them the previously formed but 

 now loose coils. A glance at Fig. 155 will make this clear. When the coils 

 have succeeded in obtaining a permanent hold on the support further basipetal 

 curvature ceases. 



After the completion of the permanent twining, growth in length completely 

 ceases, and there appears not only in these coils but also in the remainder of the 

 tendril a number of important changes, which are apparently induced not by 



