portion of the main stem is only slightly sensitive. The terminal branches lightly rubbed with a twig did not curve until from 30 m. to 42 m. had elapsed: they slowly became straight again in between 10 h. and 20 h. A loop of thread weighing one-eighth of a grain plainly caused the thinner branches to curve, as did occasionally a loop weighing one-sixteenth of a grain; but this latter slight weight, though left suspended, was not sufficient to cause a permanent flexure. The whole leaf with its tendril and the young upper internode together revolve vigorously and quickly, though irregularly, and sweep a wide space. The figure traced on a bell-glass was either an irregular spire or a zigzag line. The nearest approach to an ellipse was an elongated figure of 8, with one end a little open; this was completed in 1 h. 53 m. During a period of 6 h. 17 m. another shoot made a complex figure, apparently representing three and a half ellipses. When the lower part of the petiole bearing the leaflets was securely fastened, the tendril itself described similar but much smaller figures.
This species climbs well. The tendrils after clasping a stick become thicker and more rigid; but the blunt hooks do not turn and adapt themselves to the supporting surface, as is the case in so perfect a manner with some of the Bignoniaceæ and the Cobæa. In young plants 2 or 3 feet in height, the tendrils, which are only half the length of those borne by the same plants when grown taller, do not contract spirally after clasping a support, but only become slightly flexuous. Pull-sized tendrils, on the other hand, contract spirally, excepting the thick basal portion. Tendrils which have caught nothing simply bend downwards and inwards, like the extremities of the leaves of the Corydalis claviculata. But in all cases the petiole after a time becomes angularly and abruptly bent like that of the Eccremocarpus.
Cucurbitaceæ.—The tendrils in this family have been ranked by several competent judges as modified leaves, stipules, and branches; or the same tendril as part leaf and part branch. De Candolle considers the tendrils in two of the tribes as different in their homological nature[1]. From facts recently adduced, Mr. Berkeley thinks that Payer's view is the most probable, namely, that the tendril is "a separate portion of the leaf itself"[2].
- ↑ I am indebted to Prof. Oliver for information on this head. In the Bulletin de la Société Botanique de France, 1857, there are numerous discussions on the nature of the tendrils in this family.
- ↑ Gardeners' Chronicle, 1864, p. 721. From the affinity of the Cucurbitaceæ to the Passifloraceæ, it might be argued that the tendrils of the former are
