Biomechanics of climbing palms and how they climb

Abstract

Climbing plants have fascinated botanists since the pioneering works of Darwin and his contemporaries in the 19^th century. Diverse plants have evolved different ways of climbing and a wide range of attachment devices and stem biomechanics to cope with the particular physical demands of life as a climber. We investigated the biomechanics of attachment in a range of climbing palms, including true rattans from Southeast Asia and the genus Desmoncus from South America. We found that hook strength and orientation is coordinated with rachis geometry and rigidity. These findings support the notion of a ratchet-type attachment mechanism and partly explain why these spiny plants are so catchy and efficient at attaching to supports.

Hooks and Climbing Plants

An aspect of climbing plants that has interested our research group over recent years is how the type of attachment—hooks, twining, tendrils or roots—are coadapted with the structure, organization and growth of the stem.^1–5 Comparisons of a wide range of climbers suggest that the type of mechanical attachment is linked with a particular type of mechanical development of the stem.⁴

Slender climbing stems that are securely attached to the crowns and branches of host plants must be flexible to avoid critical stresses transmitted to the climbing stem when host plants sway in the wind or mechanically fail during tree-falls. Climbing stems attached via hooks are probably less prone than twiners to extreme mechanical stresses—at least early on during their climbing development. Hooks can be disengaged by movement against the host or they can fail mechanically and thus release strains on the stem system before they become critical. In order to ensure the latter, hooks must be adapted to break before failure of the stem bearing them. The hook strategy comes with the risk that the climber could actually fall from its support unless the stem retains sufficient rigidity to remain upright. Climbers that attach by open hooks might be expected to retain relatively stiff climbing stems, at least early in development and towards apical parts of climbing axes, before sufficient hooks have been deployed to ensure a secure attachment.

These functional constraints are possibly some of the factors underlying the wide diversity of climbing growth forms we see today. Phylogenetic constraints have also played a role in the evolution of climbing habits in palms.⁶ Climbing palms of the subfamily Calamoideae (rattans) and the genus Desmoncus, a group of predominantly climbing palms of the subfamily Arecoideae, do not have the secondary vascular cambial activity and wood production of many specialized dicot lianas. This rules out the ability to secure attachment from localized woody growth—a feature common to dicotyledonous twiners and tendril climbers—and drastically limits active modification of mechanical properties during growth of the stem. In rattans and other climbing palms, mechanical properties of the main axis can, however, change during the climbing life of the plant. Unlike dicotyledonous lianas that produce flexibility by developing flexible wood, climbing palm stems become less rigid following senescence and shedding of the outer leaf sheath, which leaves behind the familiar cane-like stems that in some species can be highly flexible.^3,5

Biomechanics of Climbing Palms

Climbing rattans (Calamoideae) of Southeast Asia and Africa and the distantly related but morphologically similar genus Desmoncus (Arecoideae) of the neotropics, have a highly characteristic mode of climbing.^5,7 Both groups are renowned in the tropics for their catchiness and efficacy in attaching to supports and anything else that tries to pass them by.⁸ Attachment is secured either by modified leaf apices known as cirri or modified inflorescences known as flagella (Fig. 1).^7,9 These attachment structures bear a variety of reflexed spines or hooks along their length (Fig. 2). We carried out mechanical tests on species from Southeast Asia and South America that measured (A) via pull-tests—the maximum force that hooks and spines withstand up to failure and (B) via bending tests and image analysis/anatomy—the stiffness, rigidity and cross-sectional geometry of the hook-bearing axes. We wanted to test how the mechanical strength of hooks varied along these attachment organs, how hook strength and arrangement varied with bending properties of the axes bearing them, and how these features might contribute to their efficiency and mode of climbing.²

Figure 1.

Attachment organs of (A) the flagellate rattan Calamus acanthospathus (Calamoideae, S. China) and (B) the cirrate palm Desmoncus orthacanthos (Arecoideae, S. America).

Figure 2.

Hook organization in (A) one surface of the cirrus rachis (bilaterally symmetrical cross-section) of Desmoncus polyacanthos (Arecoideae, S. America) and (B) whorls of spines on the rachis (circular in cross-section) of the cirrus of Plectocomia himalayana (Calamoideae, S. China).

Pull tests showed that spine and hook strength increased from the apex towards the base of the cirrus or flagellum and that hooks always failed before the axes bearing them. Hook size and strength was also linked to body size. Small, understory climbers bear small sharp hooks suited for small diameter supports and embedding into branches or trunks. Larger canopy climbers bear bigger hooks capable of engaging twigs and branches of a size that could sustain the weight of the climber.

Axes of the cirrus and flagellum are relatively stiff and mechanically reinforced by an outer layer of fibres. Some cirrate species show abrupt changes in rigidity along the rachis, linked to modifications in cross-sectional geometry—from V-shaped (vertically resistant) to more flattened (vertically compliant) cross-sections (Fig. 3). Flagellar species show drops in stiffness and rigidity at the intersections of the overlapping bracts that comprise these modified fertile axes. We also found that spines and hooks of cirri and flagella are orientated and engage in the direction of least resistance to flexure of the axis bearing them. Species with bilaterally symmetrical “V”-shaped or flattened shaped axes bear spines along one surface of the structure, whereas spines borne on circular to centrisymmetric rachises or flagella were borne all around the structure, sometimes in compact whorls.

Figure 3.

Hook-bearing rachis of (A) Desmoncus polyacanthos, (Arecoideae, S. America) showing transition from basal “v” shaped cross-section geometry (bottom) to flattened outline of the cirrus (top), with hooks confined to abaxial surface, and (B) Calamus tetradactylus, (Calamoideae, S. China) flagellum with flattened cross-sectional geometry at base and rounded cross-section in hook-bearing distal part. Note the adaxial reinforcement in distal part, with packed lignified fibres.

How Climbing Palms Climb

The results are consistent with the idea that climbers attaching by open hooks should bear them on structures that are relatively rigid in bending so they can be deployed and re-engaged when the climber becomes disengaged from its host. The results also demonstrated a diversity of structures based on this general principle. First, hooks increase in strength towards the base of the rachis suggesting that the stiff rachis and series of increasingly strong hooks are specialized as a ratchet mechanism suggested by Francis Putz.¹⁰ A series of increasingly strong hooks, aligned in the same direction along a rachis or axis will effectively ensure that movement of the host plant and engaging-disengaging of the structure will eventually hold the plant tauter against the surrounding branches under stronger loads, on stronger hooks and thus provide greater stability against further disengaging and risk of falling. Second, bilateral or circular cross-sectional geometries of rachises are apparently coupled with an arrangement of hooks on one surface or all around the axis. Third, abrupt changes in rachis geometry produce differences in bending rigidity of the hook-bearing axis that are possibly linked to the direction of hook deployment. Any disturbance or movement of the host vegetation transmitted to the attached palm stem will tend to produce swaying/bending movements of the cirri or flagella bearing hooks in a direction induced by the geometry of the rachis.

What we initially viewed as a relatively simple and “passive” type of attaching structure (compared with relatively “active” and “behavioral” twining/tendril climbing) turned out to be more complex than we first thought. The apparently straight forward business of attaching a climbing plant to a host via hooks appears to be linked to a range of physical constraints and dependant factors such as overall size, weight, axis stiffness, presence of secondary growth and the stiffness and geometry of the axes bearing the hooks.

Addendum to: Isnard S, Rowe NP. The climbing habit in palms: Biomechanics of the cirrus and flagellum. Am J Bot. 2008;95:1538–1547. doi: 10.3732/ajb.0700005.