Abstract. In tropical forests, lianas (woody vines) are important structural parasites of trees. We assessed the effects of forest fragmentation, treefall disturbance, soils, and stand attributes on liana communities in central Amazonian rain forests. Over 27 500 liana stems (Ն2 cm diameter at breast height [dbh]) were recorded in 27 1-ha plots in continuous forest and 42 plots in 10 forest fragments ranging from 1 to 100 ha in area. For each plot, an index of forest disturbance was determined from a 20-yr study of tree-community dynamics, and 19 soil-texture and chemistry parameters were derived from soil surface samples (top 20 cm).Liana abundance was 187-701 stems/ha, and liana aboveground dry biomass varied from 3.7 to 12.3 Mg/ha. Liana abundance increased significantly near forest edges and was significantly positively associated with forest disturbance and significantly negatively associated with tree biomass. Liana biomass was similarly associated with disturbance and tree biomass but also increased significantly along soil-fertility gradients. Plots near forest edges had a significantly higher proportion of small (2-3 cm dbh) lianas and relatively fewer large (Ն4 cm dbh) lianas than did sites in forest interiors.Liana communities were further assessed by comparing their species richness, composition, climbing guilds, and frequency of tree infestation in three 10-ha fragments. Within each fragment, data were collected in 24 small (400-m 2 ) plots, with half of the plots near edges and half in interiors. Significantly more trees were infested on fragment edges than in interiors. All three major guilds (branch-twiners, mainstem-twiners, tendril-twiners) were significantly more abundant on edges. Species diversity of lianas (as measured by Fisher's diversity index) also was significantly higher on edges, and this was not simply an artifact of increased liana abundance on edges.We conclude that many aspects of liana community structure are affected by habitat fragmentation, and we suggest that lianas can have important impacts on forest dynamics and functioning in fragmented rain forests. By creating physical stresses on trees and competing for light and nutrients, heavy liana infestations appear partly responsible for the dramatically elevated rates of tree mortality and damage observed near fragment edges.

A recent increase in published studies of lianas has been paralleled by a proliferation of protocols for censusing lianas. This article seeks to increase uniformity in liana inventories by providing specific recommendations for the determination of which taxa to include, the location of diameter measurement points on individual stems, the setting of minimum stem diameter cutoffs, the treatment of multiple‐stemmed and rooted clonal groups, and the measurement of noncylindrical stems. Use of more uniform liana censusing protocols may facilitate comparison of independently collected data sets and further our understanding of global patterns in liana abundance, diversity, biomass, and dynamics.

Lianas, or woody climbing plants, are a major constituent of seasonally dry tropical forests, and are thought to impact negatively their host trees. In this study we evaluated whether liana presence was associated with reduced leaf water potentials and growth in adult Senna multijuga trees during the dry season in a lowland Bolivian forest. We used leaf water potentials in trees as a first approach to assess trees' water status, under the assumption that leaf water potentials become more negative when water losses (via transpiration) exceed gains (by uptake). We measured relative growth in girth at 1.5 m height (gbh) to quantify tree growth. At the beginning of the 1996 dry season (early June), we selected 20 S. multijuga trees 10-20 cm dbh, and measured their gbh. We also recorded pre-dawn and mid-day leaf water potentials in these trees. In ten experimental trees all lianas were then cut, while the remaining trees were used as controls. Pre-dawn and mid-day water potentials were re-measured 1 day after liana-cutting, and then every week in all trees for 1 month and then at 3 and 5 months, until the beginning of the next rainy season (November); gbh was measured again in July 1997 to estimate relative growth rate. Liana removal was associated with less negative pre-dawn (-0.3 vs -0.4 MPa) and mid-day (-0.5 vs -0.7 MPa) water potentials in trees during the dry season. This difference appeared as early as 1 day after cutting, and disappeared once the rainy season began. Liana-cut trees grew more (0.4 mm/mm year) than liana-uncut trees (0.2 mm/mm year). These findings suggest that lianas may interfere with water availability to these trees during the dry season, and may also hinder tree growth.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.. Ecological Society of America is collaborating with JSTOR to digitize, preserve and extend access to Ecology.Abstract. Lianas, woody climbing plants, are a conspicuous component of tropical forest canopies that might affect prevailing conditions in the forest floor and thus impact tree seedling regeneration. The effects of lianas on tree seedling survival, growth, and density were studied in a lowland liana forest in Bolivia. Gravimetric soil water content and canopy openness were measured to evaluate whether these factors changed as a result of liana cutting. I established 24 square plots of 900 m2 each, and after an initial set of measurements, all lianas were cut in half of them, while the others plots were used as controls. Tree seedling growth and survival of two tree species were evaluated: Clarisia ilicifolia and Astronium fraxinifolium.Eighteen months after liana cutting, seedlings in liana-cut plots grew significantly taller and produced more leaves than did seedlings in control plots, but survival was not affected by treatment. Seedling growth following liana cutting was significantly higher in A. fraxinifolium than in C. ilicifolia seedlings. Densities of tree and liana seedlings did not change after liana cutting. Gravimetric soil water content was apparently not affected by liana-cut treatment. Canopy openness increased significantly in liana-cut plots, but only by 4%, and only after 26 mo. I conclude that lianas hinder the growth of tree species seedlings differentially, which in turn might shift the balance in competitive interactions between seedlings. Thus, at the study site lianas could affect tree regeneration.

In the tropical rainforest of Los Tuxtlas, Mexico, lianas (woody-climbing plants) had a clumped distribution on trees, and 63.3 percent of trees ≥10 cm in diameter carried at least one liana. Trees with larger diameters supported more lianas and a higher total basal area of lianas than trees with smaller diameters. There was no relationship between liana diameter and the number of trees a liana climbed, but tendril climbers climbed on more trees than stem-twiners. RESUMENEn el bosque tropical de Los Tuxtlas, México, las lianas (plantas trepadoras leñosas) se distribuyeron de forma agregada en losárboles, y 63.3 por ciento de losárboles ≥10 cm de diámetro sostuvieron al menos una liana.Árboles con diámetros mayores soportaron más lianas y mayorárea basal de lianas queárboles con diámetros menores. No hubo relación entre el diámetro de las lianas y el número deárboles que estas trepaban, pero las lianas con zarcillos treparon sobre másárboles que las lianas que enroscan el tallo.