that in this respect the power of spiral contraction gives a tendril-climber an advantage over leaf-climbers which have no contracting power, and therefore no means of hauling themselves up to supporting objects.
But the spiral contraction of tendrils has another use, and this is probably the most important one. This use depends on the fact that a contracted tendril acts like a spiral spring, and is thus converted into a yielding, instead of an unyielding, body. The spirally-wound tendril yields like an elastic thread to a pull which would break the tendril in its original condition. The meaning of this arrangement is to enable the plant to weather a gale which would tear it from its support by snapping the tendrils, if they were not converted into spiral springs.
My father describes how he went in a gale of wind to watch the bryony on an exposed hedge, and how, in spite of the violent wind which tossed the branches of the plant about, the bryony safely rode out the gale, "like a ship with two anchors down, and with a long range of cable ahead, to serve as a spring as she surges to the storm." It may also serve to divide the weight which has to be supported equally among a number of tendrils; and this is the meaning of the spiral contraction seen in the tendrils of the Virginia creeper.
It can be seen in Fig. 4 that all the coils of the spiral are not in the same direction. First, there are two in one direction, then six in the other, and then three again in the first direction, making six turns in one way and five in the other. And this is universally the case; the turns in one direction are always approximately equal in number to those in the opposite direction. It can be shown to be a mechanical necessity that a tendril which has its two ends fixed, and which then coils into a spiral, should behave in this way.
A simple model made to show this mechanical necessity is described by Sachs in his "Text-book." It is made by stretching a strip of India-rubber and cementing it to an unstretched strip. The strips being united in a state of longitudinal strain, form a spiral when released. If the model is held by one end only, the turns of the spiral are all in one direction. And this represents the behavior of a tendril which has not managed to seize a support; for some unknown reason such tendrils contract into spirals, and the turns of such spirals are all in one direction. But, if the India-rubber is held at both ends, half the turns are in one direction, half in the other, just as with a tendril the same thing happens.
Now, let us consider the general relations that exist between twining plants, leaf-climbing plants, and tendril-climbing plants. To an evolutionist the question how these various classes of climbing plants have been developed is perhaps of most interest. What is the relationship between them? Have all classes been developed separately from ordinary non-climbing plants, or has one class been developed out of one of the others; and, if so, which is the oldest form of climbing plant?
