492 TRANSFORMATION OF ENERGY 



discuss them at length, and we may content ourselves with referring to HABER- 

 LANDT'S (1901) exhaustive memoir on the subject. HABERLANDT, in addition to 

 anatomical data, records certain physiological arguments as to the nature of 

 contact sensitivity. He points out that a vigorous radial pressure, a com- 

 pression of the protoplasm, does not lead to a stimulation, and that the 

 stimulus-movement is induced rather by tangential tensions in the protoplasm. 

 Whether these observations help to explain the nature of the phenomenon we 

 are not prepared to say. 



It is very remarkable that tendrils which possess so great a capacity for 

 distinguishing different mechanical influences should also react to stimuli very 

 different in character from those of contact. We are indebted to CORRENS (1896 a) 

 for proof of the fact that tendrils exhibit curvatures, when subjected to sudden 

 changes of temperature (cooling or heating), quite similar to those induced by 

 contact. As far as we know (FITTING, 1903 a, 614) the mechanics of growth are 

 the same as those seen in contact curvatures. Further CORRENS has shown that 

 chemical stimuli also induce curvature in tendrils, as for instance when they are 

 treated with such diverse substances as iodine solution, acetic acid, chloro- 

 form, or ammonia. Finally PFEFFER has succeeded in obtaining responses by 

 employing weak induction currents as stimuli. 



We shall return to these phenomena at the close of this lecture, but we must 

 not omit to draw attention to them here by way of showing that the sensitivity 

 of tendrils is not so restricted as one might think from the publications on the 

 subject. On the other hand curvatures due to heat or chemical stimuli may be 

 the more readily disregarded since these are entirely unnatural and do not affect 

 tendrils in the wild state. 



Let us now turn to the curving itself that generally follows on the application 

 of a stimulus. The extraordinary rapidity of the process has not unnaturally 

 led to the assumption (DARWIN, 1876 a; MACDOUGAL, 1896) that the movement 

 was due to a diminution of turgidity on the concave side, afterwards rendered 

 permanent by growth. From FITTING'S (1903 a) researches it would appear, 

 however, that turgor has no special part to play in the process, which is rather 

 to be attributed to special growth phenomena. FITTING showed, by making 

 microscopic measurements of the movements of ink-lines made at appro- 

 priate distances apart on the upper and under sides, that immediately after the 

 stimulation the marks on the convex side separated much more rapidly than 

 before. The elongation may be so great in the course of a few minutes as to 

 amount to an increase of 50, 100, or even 160 per cent, in an hour, values not 

 attained by the non-stimulated tendril in 24 hours, and this, too, independently 

 of the age of the tendril. At the same time the indices on the under or concave 

 side approximate somewhat, resulting in an absolute contraction of about I per 

 cent, per hour. From these measurements it is obvious that not only do all the 

 tissues of the stimulated zone lying between the axis (middle line) and the convex 

 outer surface suffer extension, but that most of those towards the concave side 

 also participate in the growth increment, in other words, the neutral line, which 

 is neither elongated nor contracted, lies on the concave side of the medulla ; 

 the middle line itself exhibits a marked increase in growth. That can only be 

 definitely established by direct measurements on the flanks which suffer 

 elongation to the same degree as the middle line. 



We cannot, however, draw any conclusions from these calculations as to 

 whether the growth acceleration, which reaches a maximum at a point on the 

 other side opposite to the point of contact, is the first and the only factor 

 leading to curvature, or whether an inhibition of growth on the concave side 

 takes place at the same moment or j ust previously. The observed approximation 

 of the indices on the concave side may be active, or they may be caused passively 

 by compression induced by the curvature. It is more probable that the growth 

 acceleration is not active in all the parts which it affects but that the deeper 



