GROWTH AND MOVEMENT 469 
portion, which steadily wraps itself around the support. In the nature 
of the case it is not possible for twiners to wrap about large supports, 
nor those that are too nearly horizontal. Plants differ much in their 
capacity in these two points, a difference which depends chiefly upon the 
relative length of the growing portion of their stems, and consequently 
upon the precise distance of the most actively growing region from the 
apex. Few twiners encircle supports more than 15 cm. in diameter, or 
those that lie nearer the level than 45. 
Straightening. The coils that a twiner forms at first are loose and of 
low inclination. Later they become steeper and hug the support tightly. 
This seems to be due to a return, in the last stages of growth, to the 
apogeotropism that they possessed in the seedling period, so that the 
stem starts to erect itself, with the effect stated. Very commonly the 
surface of the stem is rough, being ridged or angled or furnished with 
stiff hairs, which prevents slipping from a support too easily or sliding 
along it. Inspection of the stem in the regions no longer growing shows 
that it is twisted, the longitudinal ridges coursing spirally around the 
axis in a direction the reverse of the twining. This torsion is mainly 
the result of the final erection of the stem, though other causes cooperate 
to increase or diminish it. 
This also is a mechanical necessity of the behavior. It can be imitated by coiling 
a long piece of rubber tubing on a table, marking a crayon line along the upper 
surface, and then lifting the inner end of the coil while the other end is held on the 
table, both ends being prevented from twisting in the fingers. Then it will be seen 
that the line apparently passes spirally around the tube, because the latter is twisted 
by the steepening of its coils. 
The tardy development of the leaves and branches is very evidently 
an advantage in twining, for they would greatly impede the revolving 
movement and the subsequent tightening of the coils. When the 
branches do develop, they show the same behavior as the main axis. 
This explanation of twining is not wholly satisfactory, because there are details 
of the process, and some features that appear only under experiment, that are not 
clearly accounted for ; but it is far the best of the many theories that have been pro- 
posed, and in the major outlines that have been presented here it is certainly correct. 
(2) THIGMOTROPISM 
Tendrils. Many plants are sensitive to mechanical stimuli such as 
contact or friction, as shown by the alterations of the rate of growth 
that lead to curvature. This phenomenon is thigmotropism* The 
