The young internodes of the Lophospermum are sensitive as well as the petioles, and by their combined movement seize any object. The flower-peduncles of the Maurandia semperflorens revolve spontaneously, and are sensitive to a touch, yet are certainly useless for climbing. The leaves of at least two and probably of most of the species of Clematis, and of Fumaria and Adlumia, spontaneously curve from side to side, like the internodes, and are thus better adapted to seize any distant object. The petioles of the perfect leaves, as well as the rudimentary or tendril-like leaves of Tropæolum tricolorum move spontaneously and slowly towards their own stem or the supporting stick, which they then clasp; these petioles also show some tendency to contract spirally. The tips of the uncaught leaves of the Gloriosa, as they grow old, contract into a flat spire. These several facts are interesting, as we shall see, in relation to true tendrils.
It was observed in some cases that, as with twining plants, so with leaf-climbers, the first internodes which rise from the ground do not spontaneously revolve; nor are the petioles or tips of the first-formed leaves sensitive. In certain species of Clematis the high development and spontaneous movements of the leaves, with their highly sensitive petioles, apparently have rendered almost superfluous the spontaneous movements of the internodes, which have consequently become enfeebled. In certain species of Tropæolum it would appear as if both the spontaneous movements of the internodes and the sensitiveness of the petioles have become enfeebled; and in one species they have been completely lost.
By tendrils I mean filamentary organs, sensitive to contact and used exclusively for climbing. By this definition, spines or hooks and rootlets, all of which are used for climbing, are excluded. True tendrils are formed by the modification of leaves with their petioles, of flower-peduncles, perhaps also of branches and stipules. Mohl, who includes with true tendrils various organs having a similar external appearance, classes them according to their homological nature, as being modified leaves, flower-peduncles, &c. This would be an excellent scheme; but I observe that botanists, who are capable of judging, are by no means unanimous on the nature of certain tendrils. Consequently I will describe tendril-bearing plants by natural families, following Lindley, and this will in most, or in all, cases keep those of the same homo-
