dril thus standing vertically up sweeps a circle right above the stem, and is well adapted to catch some object above, and to favour the ascent of the plant. The whole leaf, with its tendril, after a short time, bends downwards to one side, allowing the next succeeding leaf to become vertical, and ultimately it assumes a horizontal position; but, before this has occurred, the tendril, supposing it to have caught nothing, has lost its powers of movement and has spirally contracted into an entangled mass. In accordance with the rapidity of all the movements, their duration is short: in a plant growing vigorously from being placed in a hot-house, a tendril only revolved for about 36 hours, counting from the period when it became sensitive; but during this period it probably made at least 27 revolutions.
When the branches of a revolving tendril strike against a stick, they quickly bend round and clasp it; but the little hooks play an important part, especially if only the extremity of the tendril be caught, in preventing its being dragged by the rapid revolving movement away too quickly for its irritability to act. As soon as a tendril has bent round a smooth stick or a thick rugged post, or has come into contact with planed wood (for it can at least temporarily adhere even to so smooth a surface as this), the same peculiar movements begin in the branchlets as have been described in those of the Bignonia capreolata and the Eccremocarpus, namely, the branchlets lift themselves up and down; those, however, which have their hooks already directed downwards remain in this position and secure the tendril, whilst the others twist about till they arrange themselves in conformity with every irregularity of the surface, and bring their hooks, originally facing in various directions, into contact with the wood. The use of the hooks was shown by giving the tendrils tubes and slips of glass to catch; for these, though temporarily seized, were afterwards invariably lost, either during the arrangement of the branches or when the spiral contraction ensued.
The perfect manner in which the branches arrange themselves, creeping like rootlets over all the inequalities and into any deep crevice, is quite a pretty sight; for it is perhaps more effectually done than by the tendrils of the former species, and is certainly more conspicuous, as the upper surfaces of the main stem and of every branch to the extreme hooks are angular and coloured green, whilst the lower surfaces are rounded and purple. I was led to infer, as in the former cases, that light guided these conforming movements of the branches of the tendrils. I made
