Answers are grounded in this paper and cite their sources.
Liana (woody vine) loads on 20 trees (dbh >20 cm) were determined for each of 24 species on Barro Colorado Island, Panama. Three species were consistently free of lianas. Trees of predominantly liana-free species were more flexible and had longer leaves than trees of species that more often carried lianas. It is suggested that because lianas are likely to slow host-tree growth rates and may increase tree mortality rates, characteristics that help trees to shed or avoid lianas are advantageous in liana-rich forests. LIANAS ARE WOODY CLIMBING PLANTS that rely on other plants for support. Resources which self-supporting plants allocate for support (wood) may be used by lianas for extension growth and reproduction (Darwin 1867). Lianas are generally detrimental to trees that support them. For example, stem diameter growth rates have been shown to be slower for liana-laden than for liana-free trees by Featherly (1941) and by Putz (1982). Furthermore, lianas can damage their hosts directly by mechanical abrasion and passive strangulation (Lutz 1943, pers. obs.) or indirectly by making their hosts more susceptible to ice (Siccama et al. 1976) and wind damage. Lianas also increase the probability of their host tree falling (Webb 1958, Putz 1982) and increase accessibility of tree crowns to folivorous and other potentially harmful canopy animals (Montgomery and Sunquist 1978, Charles-Dominique et al. 1981). To the extent that lianas decrease host tree fitness, trees that remain free of lianas derive selective advantage from this ability. In this paper, I examine some characteristics that may contribute to a tree's ability to avoid or shed lianas. Two physical characteristics suggested as reducing a tree's susceptibility to lianas are buttresses and smooth bark. Black and Harper (1979) suggest that by casting shade on liana seedlings, buttresses reduce the accessibility of tree trunks to lianas, however Putz (1980) points out that because small trees seldom have large buttresses, this would only be true for lianas that climb directly up tree trunks with roots or adhesive tendrils. Boom and Mori (1982) censused liana-loads on trees in a lowland forest in Brazil and found no support for the buttress-liana hypothesis. They also found smooth bark had no significant effect on the proportion of trees supporting lianas. The most common types of lianas on Barro Coloardo Island (BCI), Panama, climb with twining-stems, tendrils, or modified branches that grow out and surround supports. The need to encircle supports constrains these lianas to climbing on other lianas and on small diameter trees and branches (Penalosa 1982, Putz 1982). Root climbers and those that climb with the aid of adhesive tendrils can climb on supports of all sizes but these types of lianas are relatively rare on BCI. (Because they are not woody and because they sometimes lose connections with the ground, I do not consider plants in the Araceae and Cyclanthaceae to be lianas.) Because most lianas must surround their hosts, the larger the tree the less susceptible it is to lianas climbing up from the ground. This suggests that rapid stem thickening may enable trees to avoid lianas (Putz 1980). Large liana-free trees can be colonized by lianas growing from neighboring tree crowns. By reducing the surface area exposed to adjacent trees, trees with long, branchfree boles relative to overall height (i.e., shallow-crowned trees) may be less susceptible to such lateral invasion. Another way trees may avoid lianas extending over from neighbors is by being mechanically flexible. When trees recently connected by lianas sway in opposite directions, the lianas may be broken or dislodged. I therefore predicted that the probability of a tree carrying lianas should be negatively correlated with its flexibility. Once they reach the top of trees, lianas characteristically grow out onto limbs or large leaves. If the leaves are large enough, when shed they may dislodge attached lianas. This mechanism is most easily envisioned operating with extremely large-leaved plants such as palms; when a large palm leaf is shed, adhering lianas could be pulled off the tree unless firmly attached to other leaves. If this process occurs, trees with large leaves would be expected to host lianas less often than trees with small leaves. In the course of preliminary studies of the effects of lianas on tree saplings (unpublished data), I marked 30 lianas growing on spiny-trunked saplings and 75 lianas growing on spine-free saplings. During the one year observation period, a significantly higher proportion (x2 = 4.4, P < .05) of the lianas on spiny-stemmed saplings died. It appeared that when the spiny-trunked saplings I Received 15 December 1982, revised 5 June 1983, accepted 26 June 1983. BIOTROPICA 16(1): 19-23 1984 19 This content downloaded from 207.46.13.76 on Fri, 09 Sep 2016 04:38:39 UTC All use subject to http://about.jstor.org/terms swayed in the breeze, the spines severed the lianas. This suggested the hypothesis that spiny-trunked trees carry lianas less frequently than spine-free trees. To determine what factors influence the probability that a tree carries lianas, I tested the hypotheses that rapid growth, large leaves, flexible stems, and trunk spines characterize liana-free trees. In addition, the hypothesis that lianas twining around spiny-trunked trees are sawn off when the trees sway was tested experimentally.
No takes yet. Share an insight, caveat, or question.
Francis E. Putz (1984) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: