We have said that the circumnutation of climbing plants differs from that of ordinary plants chiefly by its greater amplitude. But most leaves circumnutate in an almost vertical plane, and therefore describe very narrow ellipses, whereas the many kinds of tendrils which consist of metamorphosed leaves, make much broader ellipses or nearly circular figures; and thus they have a far better chance of catching hold of a support on any side. The movements of climbing plants have also been modified in some few other special ways. Thus the circumnutating stems of Solnanum dulcamara can twine round a support only when this is as thin and flexible as a string or thread. The twining stems of several British plants cannot twine round a support when it is more than a few inches in thickness; whilst in tropical forests some can embrace thick trunks;[1] and this great difference in power depends on some unknown difference in their manner of circumnutation. The most remarkable special modification of this movement which we have observed is in the tendrils of Echinocystis lobata; these are usually inclined at about 45° above the horizon, but they stiffen and straighten themselves so as to stand upright in a part of their circular course, namely, when they approach and have to pass over the summit or the shoot from which they arise. If they had not possessed and exercised this curious power, they would infallibly have struck against the summit of the shoot and been arrested in their course. As soon as one of these tendrils with its three branches begins to stiffen itself and rise up vertically, the revolving motion becomes more rapid; and as soon as it has passed over the point of difficulty, its motion coinciding with that from its own weight, causes it to fall into its previously inclined position so quickly, that the apex can be seen travelling like the hand of a gigantic clock. [1] ‘The Movements and Habits of Climbing Plants,’ p. 36. A large number of ordinary leaves and leaflets and a few flower-peduncles are provided with pulvini; but this is not the case with a single tendril at present known. The cause of this difference probably lies in the fact, that the chief service of a pulvinus is to prolong the movement of the part thus provided after growth has ceased; and as tendrils or other climbing-organs are of use only whilst the plant is increasing in height or growing, a pulvinus which served to prolong their movements would be useless.