traction into a firm bundle. In the Cobæa the tendrils alone revolve; these are divided into many fine branches, terminating in sharp little hooks, which crawl into crevices, and are turned by an excellently adapted movement to any object that is seized. In the Ampelopsis, on the other hand, there is little or no power of revolving in any part: the branched tendrils are but little sensitive to contact; their hooked extremities cannot seize any thin object; they will not even clasp a stick, unless in extreme need of a support; but they turn from the light to the dark, and, spreading out their branches in contact with any nearly flat surface, the disks are developed. These can adhere, by the secretion of some cement, to a wall, or even to a polished surface; and this is more than the disks of the Bignonia capreolata can effect.
The formation and rapid growth of these adherent disks is one of the most remarkable peculiarities in the structure and functions of tendrils. We have seen that such disks are formed by two species of Bignonia, by the Ampelopsis, and, according to Naudin[1], by the Cucurbitaceous genus Peponopsis adhærens. Their development, apparently in all cases, depends on the stimulus from contact. It is not a little singular that three families so widely distinct as the Bignoniaceæ, Vitaceæ, and Cucurbitaceæ should all have species bearing tendrils with this same remarkable peculiarity. Most tendrils, after they have clasped any object, rapidly increase in strength and thickness throughout their whole length; but some tendrils, when wound round a support either by the middle or the extremity, become swollen at these points in a remarkable manner; thus I have seen the clasped portion of a tendril of the Bignonia Chamberlaynii grown twice as thick as the free basal portion, and become wonderfully rigid. In the Anguria the lower surface of the tendril, after it has wound round a stick, forms a coarsely cellular layer, which closely fits the wood, but is not adherent; in the Hanburya similar layer is developed, which is adherent; lastly, in the Peponopsis adherent disks are formed at the tips of the tendrils. These three last-named genera belong to the Cucurbitaceæ, so that, in this one family, we have a nearly perfect gradation from a common tendril to one that forms an adherent disk at its tip; the one small step which is wanted is a tendril in a state between that of the Anguria and Hanburya—that is, adherent only in a slight degree or occasionally.
Finally, it may be added that America, which so abounds with arboreal animals, as has lately been insisted on by Mr. Bates,
- ↑ Annales des Sc. Nat. Bot. 4th series, tom. xii. p. 89.
