Host preferences of Rhus radicans (Anacardiaceae) in a southern deciduous hardwood forest.
Although woody vines, or lianas, are important components of both temperate and tropical forests (Darwin 1867, Putz 1984b, Gentry 1991, Hegarty and Caballe 1991, Collins and Wein 1993), and can present serious silvicultural problems (e.g., Lutz 1943, Nicholson 1958, Fox 1968, Appanah and Putz 1984, Putz et al. 1984, Putz 1991), there are basic aspects of the biology of vines that remain poorly understood. In particular, the simple question of whether vine species grow more successfully upon some kinds of trees than others remains largely unanswered.
Lianas, like epiphytes and hemiepiphytes, are mechanical parasites, dependent upon trees to provide support as lianas grow from the forest floor to better lit areas in the forest canopy. This allows lianas to grow, as Darwin (1867) put it, ". . . with wonderfully little expenditure of organized matter, in comparison with trees, which have to support a load of heavy branches by a massive trunk." As a result of this growth efficiency lianas may contribute much more to primary productivity, litterfall, and nutrient cycling than their basal area or biomass would immediately suggest (Kato et al. 1978, Putz 1983). But trees are paying the support costs for themselves and the lianas they support, and the mechanical burden imposed by lianas in combination with their competitive presence, must select for trees that are less suitable hosts (Putz 1980, 1984a, b, 1991, Stevens 1987, Hegarty 1991, Teramura et al. 1991). A number of potentially defensive measures against liana invasion have been suggested (see, e.g., Putz 1984b, Stevens 1987, Hegarty 1991), but, as Hegarty (1991) points out, the efficacy of these in natural settings is difficult to determine. If, however, defenses against lianas differ in their relative effectiveness, lianas should not be randomly distributed among trees in a forest.
This paper addresses the following basic questions: (1) Does the liana Rhus radicans L. (poison ivy) climb tree species independently of their abundance? That is, does poison ivy exhibit host preferences, or host selectivity? (2) Do R. radicans vines occupy trees independently of the host's size? (3) Is growth from the time of host colonization to reproductive maturity independent of host species and host size? Or is growth to maturity more likely on some hosts than others? Finally, (4) Are some tree barks allelopathic towards R. radicans? In particular, is seed germination and early seedling growth of R. radicans influenced by host bark allelochemicals?
Rhus radicans L. (poison ivy, Anacardiaceae), an adventitious root-climbing woody vine, is found in a variety of eastern North American habitats ranging from cypress swamps to dry rock outcrops, but is especially abundant in open woods and second-growth thickets (Radford et al. 1968). It is a perennial plant with alternate, compound leaves with three leaflets. All parts of the plant contain the contact allergen urushiol (Merck Index 1989). Although R. radicans can exist as a free-standing shrub in open areas, it more commonly uses adventitious roots to climb rocks and trees. In mature forests of the southeastern U.S. Rhus radicans vines grow into the lower crowns of canopy trees, flowering and fruiting on horizontal branches extending from the upper boles and large limbs of the host trees. R. radicans are, like root climbers in general (Darwin 1867), unable to grow on small branches and twigs. Gartner (1991a, b) has suggested in an examination of the congeneric Toxicodendron diversilobum Greene (= Rhus diversiloba T.&G.) that there are are benefits to the viny growth form relative to that of the free-standing shrub. Vines allocate less biomass to mechanical support, and thus can allocate more to resource acquisition by root growth and leaf production. Furthermore, the vine growth form allows for more rapid vertical extension into well-lit regions.
The primary study area was a 200 x 60 m plot, gridded into 30 contiguous 20 x 20 m quadrats, in an old-growth mixed mesophytic hardwood stand in the City of Huntsville's Cold Springs Natural Area on Monte Sano Mountain, a mesa dissected off the southwestern edge of the Cumberland Plateau (34 [degrees] 45[minutes] N, 86 [degrees] 32[minutes] W). The plot is at 480 m elevation in a north-facing hollow with a permanent spring. The soil is derived from deep colluvial debris accumulated at the foot of a bluff of interbedded limestones and shales of the Pennington Formation. The colluvium rests upon on a terrace of more resistant limestones of the Bangor Formation.
Within the study plot each of the 301 trees [greater than]10 cm diameter at breast height (dbh) was identified to species (nomenclature follows Radford et al. 1968), measured (dbh and height), and examined for R. radicans. The diameter of each R. radicans vine ascending a tree was measured 10 cm above the ground using vernier calipers. In this study, fruiting of R. radicans was not observed on vines that lacked horizontal branching. R. radicans is thus a vine of the type described by Lee and Richards (1991) in which heteroblastic changes occur as branches without adventitious roots extend away from the juvenile form's support. Consequently, horizontal branching was used to distinguish between juvenile (without horizontal branching) and adult life history stages. Since erect juvenile sprouts of R. radicans are common on the forest floor, each such sprout was tallied in 30 1 x 1 m plots established in a stratified random design (1 plot per 20 x 20 m quadrat). Because vines were not physically removed from the soil it was not possible to determine whether individual vines were sexually produced individuals (genets) or genetically identical (ramets).
Two tree species with well-known allelopathic properties, Juglans nigra (De Scisciolo et al. 1990) and Sassafras albidum (Gant and Clebsch 1975), were rare on the study plot, so adjacent forest of similar composition and structure was searched to increase the sample of these species. Each of these additional trees (25 S. albidum and 13 J. nigra) was measured and examined in the same manner as trees within the plot.
Bark samples were collected from the Cold Springs Natural Area, and were stored at -40 [degrees] C until extracted. Strips [approximately equal to]2 cm thick were carved from the deeply ridged barks of Juglans nigra, Sassafras albidum, and Quercus rubra. Large peeling strips were taken from the bark of Carya ovata, and large flakes were taken from the bark of mature Acer saccharum. The bark samples were each pulverized and extracted using a Soxhlet extractor (Fisher Scientific, Pittsburgh, Pennsylvania, USA) and refluxing 95% ethanol for 8 h. The ethanol solutions were concentrated by rotary evaporation to give the crude extracts. Solutions (1% mass/volume in ethanol) of each of the crude extracts were prepared for biological testing.
Rhus radicans seed germination
Fresh R. radicans fruit were collected from vines located on the University of Alabama in Huntsville campus. The fruits were allowed to dry for 4 d and the seeds were then isolated mechanically (USDA 1974): the dried fruits were placed in a canvas bag and pounded for 1 h; the hulls and branches were rinsed off with water. The seeds were then soaked in 12 mol/L sulfuric acid for 1 h, and then rinsed with water, aqueous sodium bicarbonate, and again with water. The R. radicans seeds were then stratified: the seeds were packed in wet paper towels and stored at 5 [degrees] C for 7 d, then stored at -5 [degrees] C for 14 d, and then another 7 d at 5 [degrees] C.
For each bark extract, three replicates of 100 seeds were placed in a petri dish on a moistened filter paper (11 cm diameter), which had been pretreated with 1% ethanol extract (filter paper soaked in the extract and then thoroughly air-dried to remove the ethanol). Controls consisted of three such replicates on moistened filter paper, which had been pretreated with ethanol, but then thoroughly air-dried. The seeds and the filter paper were kept moist with deionized water, and maintained under plant grow lamps (Sylvania Gro-Lux 15-W Horticultural Fluorescent Bulbs) (illuminated 12 h on, 12 h off) for 15 d. The success of germination and the radicle and hypocotyl lengths (in millimetres) were monitored during that period.
Population structure of Rhus radicans
Rhus radicans is very abundant on the study plot; 401 vines were observed ascending 115 trees (38% of the 301 trees [greater than]10 cm dbh on the plot). Of these vines, 105 (26%) had horizontal branches and were judged to be reproductively mature; 296 (74%) were juvenile vines without horizontal branching (Table 1). These correspond to densities of 87.5 vines/ha and 246.6 vines/ha for mature and juvenile vines, respectively. The largest vine in the study plot was 8.7 cm in diameter and ascended 22 m into the lower crown of a Liriodendron tulipifera tree (76 cm dbh and 29.5 m tall). In the 30 1 x 1 m quadrats 129 erect stems were observed (a density of 43 000 stems/ha), all [less than]25 cm tall, and none with fruits or inflorescences. These small erect stems were not randomly dispersed on the forest floor ([[[Chi].sup.2].sub.6] = 151.5; P [less than] 0.005; for fit to Poisson distribution with mean = 4.3 stems/m), but rather were clumped. In 14 quadrats there were no Rhus radicans, but in 7 quadrats there were eight vines or more.
TABLE 1. Size class structure of the portion of the Rhus radicans population ascending trees as vines in 1.2 ha of old-growth mixed mesophytic forest at Cold Springs, Huntsville, Alabama.
Stem diameter (cm)(*) No. juveniles No. adults
[less than]0.25 142 0 [0.25-0.5] 135 11 (0.5-1.0] 16 23 (1.0-2.0] 0 16 (2.0-3.0] 0 14 (3.0-4.0] 0 10 (4.0-5.0] 2(**) 14 (5.0-6.0] 0 7 (6.0-7.0] 0 6 [greater than]7.0 1(**) 4
* Parentheses in class limits indicate exclusion of end points from the class; brackets indicate inclusion.
(**) Broken stems with sprouts, but lacking horizontal branches.
The question of whether R. radicans is more successful at ascending some trees than others is complicated by the fact that host abundance can be assessed in several ways (e.g., Daniels and Lawton 1991). If one considers host abundance as the number of stems available for R. radicans to climb, then relative abundance of different host tree species can be determined from their proportional contribution to the number of stems in the forest. This approach, however, ignores the fact that bigger trees are bigger targets for colonization. When colonizing hosts, R. radicans exhibits ascending primary stems that are [approximately equal to]2 mm diameter, which climb out of the leaf litter of the forest floor, and are appressed to the base of the host trunk, attached by adventitious roots. It was therefore reasoned that tree basal circumference was probably better than simple presence as a measure of the abundance of potential hosts. Both measures of tree species abundance are presented in Table 2.
For statistical purposes the less abundant tree species, indicated in Table 2, have been combined as "other trees," and Quercus prinus, Q. rubra, and Q. velutina have been grouped as "hardbarked oaks" (as distinct from Q. alba and Q. muehlenbergii, which have bark of a quite different, scaly or flaky texture).
Rhus radicans vines, considering both juveniles and adults together, were not distributed among tree species in proportion to their relative stem abundances; ([[[Chi].sup.2].sub.9] = 192.5; P [less than] 0.001). Carya ovata and the hardbarked oaks supported twice as many R. radicans as expected from their relative stem densities, while Acer saccharum, common as small trees in the understory, support half the R. radicans expected (Table 2). A somewhat different picture arose from a consideration of stem circumference as a measure of abundance. R. radicans, considering adults and juveniles together, were not distributed over tree species in proportion to their relative summed circumferences ([[[Chi].sup.2].sub.9] = 100.03; P [less than] 0.001). Subsequent one-degree-of-freedom [[Chi].sup.2] tests revealed that R. radicans vines were roughly twice as abundant as expected on Carya ovata and on the hard hardbarked oaks, 1.3 times the expected abundance on Aesculus octandra, two-thirds the expected abundance on Acer saccharum, half the expected abundance on Tilia americana, and a third the expected abundance on Quercus muehlenbergii. R. radicans vines were less abundant than expected upon the less common tree species grouped together as "other trees," but given the array of tree species involved this is impossible to interpret. Otherwise R. radicans vines occupied tree species in proportion to the opportunities for colonization that those species present.
The fate of juvenile R. radicans vines was uninfluenced by the host species. Adult vines were generally distributed among tree species in the same manner as juveniles ([[[Chi].sup.2].sub.9] = 14.3; P = 0.11). This suggests that the patterns of host preferences by R. radicans are established at the time young vines first start to ascend potential hosts.
The influence of host size
Some of the above patterns of tree occupation by R. radicans may be attributable to tree size (Table 3). When the trees of the study area were grouped into diameter classes independently of species, and both juvenile and adult R. radicans vines were considered together, the vines were not found to be distributed independently of tree size ([[[Chi].sup.2].sub.8] = 70.08; P [less than] 0.005). R. radicans were about three-fifths of the expected abundance on trees [less than]60 cm dbh, and 1.5 times the expected abundance on trees [greater than]60 cm dbh. Adults seemed to be distributed across host sizes as juveniles were ([[[Chi].sup.2].sub.8] = 11.10; P = 0.2). This reinforced the notion that patterns of host occupation were determined by factors influencing host colonization by juvenile sprouts from the forest floor.
TABLE 2. Forest composition and Rhus radicans occupation of hosts in 1.2 ha of old-growth mixed mesophytic forest at Cold Springs, Huntsville, Alabama.
Summed circum- No. ju- No. No. ference venile adult Tree species trees (cm) vines vines
Acer saccharum 89 5816.0 41 10 Carya ovata 35 3682.0 63 28 Liriodendron tulipifera 33 5790.3 57 16 Aesculus octandra 29 3540.0 40 20 Tilia americana 21 2115.2 12 0 Quercus muehlenbergii 16 1657.8 4 3 Carya glabra 14 1158.6 10 2 Quercus alba 12 1617.0 18 9 Quercus rubra(*) 12 1906.3 39 9 Carya tomentosa 11 1144.5 6 4 Magnolia acuminata 6 563.6 3 2 Quercus prinus(*) 5 593.4 2 0 Liquidambar styraciflua(**) 4 302.5 1 0 Fraxinus americana(**) 3 177.2 0 0 Juglans nigra 3 369.1 0 0 Sassafras albidum 2 272.1 0 0 Quercus velutina(*) 1 100.5 12 0 Cornus florida(**) 1 41.5 0 0 Nyssa sylvatica(**) 1 49.3 0 0 Prunus serotina(**) 1 23.2 0 0 Ulmus americana(**) 1 22.5 0 0
* Grouped in analyses as "hardbarked oaks."
(**) Grouped in analyses as "other trees." TABLE 3. Size class structure of 1.2-ha old-growth mixed mesophytic forest at Cold Springs, Huntsville, Alabama, and size-specific occupation of host trees by Rhus radicans.
Summed No. No. Tree dbh No. circum- juvenile adult (cm)(*) trees ference (cm) vines vines
[10-20] 92 4300.2 26 5 (20-30] 69 5408.9 58 16 (30-40] 47 5099.1 40 21 (40-50] 35 4923.8 32 12 (50-60] 21 3572.3 21 6 (60-70] 16 3240.2 52 19 (70-80] 9 2087.0 31 16 (80-90] 8 2140.4 21 4 (90-100] 3 885.9 14 7
* Parentheses in class limits indicate exculsion of end points from the class; brackets indicate inclusion.
The marked preference of R. radicans for Carya ovata is apparently not, however, just a matter of host size. The size class distribution of C. ovata differs from that of other trees in the forest (all species except C. ovata lumped together) ([[[Chi].sup.2].sub.8] = 13.14; P = 0.044). There were roughly twice as many C. ovata 40-60 cm dbh as expected, and only one-fourth the expected number [greater than]60 cm dbh. Furthermore, if Carya ovata, and the R. radicans upon it, were removed from consideration, R. radicans vines were still not distributed among tree size classes in proportion to the summed basal circumference of those size classes ([[[Chi].sup.2].sub.8] = 100.9; P [less than] 0.005). Instead R. radicans were found in about two-thirds the expected abundance on trees [less than]60 cm dbh, and were [approximately equal to]1.6 times as abundant as expected on trees [greater than]60 cm dbh. Clearly then, the preference of R. radicans for Carya ovata occurs in spite of its size.
Similarly, the preference of R. radicans for large trees was not just a reflection of the fact it grew less commonly than expected upon Acer saccharum, the most abundant small tree in the stand. When Acer saccharum and the R. radicans vines upon them were removed from consideration, R. radicans were still not distributed among tree size classes in proportion to their basal circumference ([[[Chi].sup.2].sub.8] = 64.48; P [less than] 0.005).
Since Aesculus octandra, Quercus muehlenbergii, and Tilia americana had size class distributions like that of the other trees of the stand ([[[Chi].sup.2].sub.6] = 8.34, P = 0.2; [[[Chi].sup.2].sub.5] = 3.72, P = 0.3; and [[[Chi].sup.2].sub.6] = 5.19, P = 0.5, respectively), the preference of adult R. radicans for A. octandra, and the avoidance of Q. muehlenbergii and T. americana were apparently not influenced by host size per se.
Sassafras albidum and Juglans nigra
Neither S. albidum nor J. nigra support the number of R. radicans expected from tree basal circumference. On J. nigra, juvenile R. radicans vines were one-third the expected abundance ([[[Chi].sup.2].sub.1] = 7.17; P = 0.007), and adults were simply absent ([[[Chi].sup.2].sub.1] = 5.57; P = 0.02). On S. albidum there were 1/20 the expected number of juvenile vines ([[[Chi].sup.2].sub.1] = 22.95; P [less than] 0.005), and one-third the expected number of adults ([[[Chi].sup.2].sub.1] = 3.77; P = 0.055).
Bark, seed germination, and seedling growth
Under the control conditions mean ([+ or -]1 SD) germination of Rhus radicans seeds was 24.3 [+ or -] 3.5% but germination varied among experimental treatments (one-way ANOVA; [F.sub.6,14] = 23.56; P [less than] 0.001). Ethanol extracts of barks of Carya ovata, Quercus rubra, Acer saccharum, Juglans nigra and Sassafras albidum all inhibited seed germination of Rhus radicans (Tukey hsd multiple comparisons; Table 4). Furthermore, extract of the bark of Juglans nigra more effectively inhibited germination than those of Carya ovata or Quercus rubra.
Seedling growth was less influenced by bark extracts. Hypocotyl and radicle lengths at day 15 were neither normally nor lognormally distributed, so pairwise Kolmogorov-Smirnov tests were used for comparisons between controls and treatments, and among treatments. The distribution of Rhus radicans seedling hypocotyl lengths did not differ between the controls and any of the bark extract treatments (Kolmogorov-Smirnov [[[Chi].sup.2].sub.2] = 4.4, P = 0.12, for control vs. Carya ovata; [[[Chi].sup.2].sub.2] = 3.78, P = 0.17, for control vs. Quercus rubra; [[[Chi].sup.2].sub.2] = 3.61, P = 0.18, for control vs. Juglans nigra, [[[Chi].sup.2].sub.2] = 2.14, P [greater than] 0.25, control vs. Sassafras albidum; [[[Chi].sup.2].sub.2] = 2.70, P = 0.25, for control vs. Acer saccharum). Hypocotyl lengths differed only between the treatments involving bark extracts of S. albidum in comparison to those of C. ovata ([[[Chi].sup.2].sub.2] = 9.15, P = 0.011) and Q. rubra ([[[Chi].sup.2].sub.2] = 8.68, P = 0.017). In each case there was a larger proportion (62% vs. 27% and 28%, respectively) of very small seedlings ([less than]2.5 mm hypocotyls) in the presence of S. albidum bark extract. Similarly, the distribution of radicle lengths did not differ between the controls and any of the bark extract treatments ([[[Chi].sup.2].sub.2] = 4.40, P = 0.12, for control vs. C. ovata; [[[Chi].sup.2].sub.2] = 0.64, P [greater than] 0.5, for control vs. Q. rubra; [[[Chi].sup.2].sub.2] = 0.70; P [greater than] 0.5, for control vs. J. nigra; [[[Chi].sup.2].sub.2] = 3.05, P = 0.22, for control vs. S. albidum; [[[Chi].sup.2].sub.2] = 3.90, P = 0.17, for control vs. A. saccharum). The treatment with C. ovata bark extract did have a smaller proportion of seedlings with very short ([less than]2.5 mm) radicles than did those with S. albidum and A. saccharum (27% vs. 65% and 68%, [[[Chi].sup.2].sub.2] = 11.27 and 12.46, respectively, P [less than] 0.005 for both).
TABLE 4. Germination of Rhus radicans seeds in the presence of bark extracts. Three replicates of 100 seeds each in each treatment.(*)
Germination Extract [Mathematical Expression Omitted] SD
Control 24.3 a 3.5 Carya ovata 15.7 b 1.5 Quercus rubra 14.3 bc 1.5 Juglans nigra 7.3 c 0.6 Sassafras albidum 11.3 bc 2.5 Acer saccharum 10.3 bc 1.5
* Treatments followed by the same letter do not differ significantly at P [less than] 0.05 (Tukey hsd multiple comparisons following one-way ANOVA).
Lianas, being mechanical parasites like epiphytes and hemiepiphytes, must deal with host trees in addition to the environmental conditions that typically constrain plant growth. While all plants have to deal with neighbors, the interaction between mechanical parasites and their hosts is obviously more intimate in terms of the physical contacts, and contains costs to the host not incumbent in ordinary competitive interactions. The patterns of host occupation exhibited by Rhus radicans in the old-growth mesic hardwood forest of our study area have interesting implications for evolutionary diversification among climbing plants and for the study of the mechanisms of liana-host interactions. First, the fact that R. radicans grew more abundantly than expected on some host tree species, and less abundantly than expected on others, tells something of the nature of the niche space exploited by climbing plants. Although twining and tendril-climbing lianas are dependent upon the availability of trellis-like structure in the vegetation for ascent to the canopy, root climbers are free of this constraint. R. radicans shows, however, that the availability of particular host species also influences climbing opportunities. R. radicans joins a growing set of mechanically parasitic plants known to exhibit host preferences, which includes epiphytes (e.g., Schlesinger and Marks 1977), hemiepiphytes (e.g., Todzia 1986), and stranglers (e.g., Putz and Holbrook 1989, Daniels and Lawton 1991, Laman 1994).
Why R. radicans prefers some host species more than others is still not entirely clear. Our results show, however, that it is not just a matter of host size; R. radicans was more abundant than expected on large trees, but also on Carya ovata, although it very seldom grew [greater than]60 cm dbh in our area. Some hosts might provide better physical environments for R. radicans growth. Crown morphology, leaf area, and leaf presentation of some tree species might, for example, result in better lit boles and lower canopies. Examination of forest canopy light environments (e.g., Parker 1995) should determine whether such heterogeneity exists. Carter and Teramura (1988) provide a description of the photosynthetic capacities of R. radicans and other temperate vines, but the impact of heterogeneity of the physical environment on growth of vines in the wild remains largely unexplored (Teramura et al. 1991). In any case, since it appears that the host preferences of R. radicans are established at the colonization stage, the critical determinants of its success must be acting at the bases of potential host trunks.
Some host trees, of course, might offer better root attachment opportunities, and therefore be easier to climb. As adventitious root-climbers, R. radicans vines need stable surfaces for root attachment. It seems that root adherence to rapidly exfoliating or crumbling bark would be difficult (Stevens 1987). Indeed, gentle tugs on small R. radicans on Quercus alba often pull roots free with attached pieces of host bark. Although Carya ovata (shagbark hickory), the most favored host, has exfoliating bark, the large, thick plates are strongly attached to the trunk, persist for many years, and have a hard surface. Indeed, the crevices and tunnels of its shaggy surface may offer more secure attachment than afforded by other trees. In the absence of data, however we are left with the basic question: Do R. radicans roots attach more firmly to some barks than others?
Host colonization may also be influenced by allelopathy. Frei and Dodson (1972), for instance, have shown that bark extracts of some Quercus spp. from Mexico adversely affect germination and early growth of epiphytic orchids, but information regarding allelochemical interactions between liana and their hosts is lacking (Hegarty 1991, Teramura et al. 1991, but see Talley et al., in press). Our extracts of the barks of Juglans nigra and Sassafras albidum, however, inhibited germination and seedling growth, respectively, relative to bark extract of Carya ovata. S. albidum bark contains safrole, [Alpha]-pinene, [Alpha]-phellandrene, camphor, and eugenol, which are phytotoxic (Gant and Clebsch 1975). J. nigra, which is well known to inhibit the growth of neighboring plants (Gabriel 1975, Rice 1984), produces juglone, a broadly phytotoxic phenolic compound (Neave and Dawson 1989, Heijl et al. 1993), which promotes abnormal tissue development of cucumber radicles and cotyledons (Tekintas et al. 1988). Since the host preferences of Rhus radicans seem to be established at the point of colonizing host trunks, allelochemical inhibition of sprouts on the forest floor by soluble toxins leached from foliage (Hooks and Stubbs 1967, McPherson and Thompson 1972, Lodhi 1976), decaying litter (Lodhi 1978), or roots (Whittaker and Feeney 1971) may also be an important influence on host preferences. This leaves us with yet another question: Does the distribution of R. radicans sprouts on the forest floor reflect the observed patterns of host preference?
We thank the Huntsville Land Trust and Monte Sano State Park for allowing us to conduct this research project. We are indebted to Thiokol Corp. (Huntsville, Alabama) for a generous donation of laboratory glassware used for carrying out this research. Partial support of this research provided by the Petroleum Research Fund administered by the American Chemical Society is gratefully acknowledged. We thank J. K. Baird, S. C. Thomas, J. D. Daniels, and P. Cothran for helpful discussions.
Appanah, S., and F. E. Putz. 1984. Climber abundance in virgin dipterocarp forest and the effect of pre-felling climber cutting on logging damage. Malaysian Forester 47:335-342.
Carter, G. A., and A. H. Teramura. 1988. Vine photosynthesis and relationships to climbing mechanics in a forest understory. American Journal of Botany 75:1011-1018.
Collins, B. S., and G. R. Wein. 1993. Understory vines: distribution and relation to environment on a southern mixed hardwood site. Bulletin of the Torrey Botanical Club 120:38-44.
Daniels, J. D., and R. O. Lawton. 1991. Habitat and host preference of Ficus crassiuscula, a neotropical strangling fig of the lower-montane rain forest. Journal of Ecology 79:129-141.
Darwin, C. 1867. On the movements and habits of climbing plants. Journal of the Linnean Society (Botany) 9:1-118.
De Scisciolo, B., D. J. Leopold, and D. C. Walton. 1990. Seasonal patterns of juglone in soil beneath Juglans nigra (black walnut) and influence of J. nigra on understory vegetation. Journal of Chemical Ecology 16:1111-1130.
Fox, J. E. D. 1968. Logging damage and the influence of climber cutting prior to logging in the lowland dipterocarp forest of Sabah. Malayan Forester 31:326-347.
Frei, J. K., and C. H. Dodson. 1972. The chemical effect of certain bark substrates on the germination and early growth of epiphytic orchids. Bulletin of the Torrey Botanical Club 99:301-307.
Gabriel, W. J. 1975. Allelopathic effects of black walnut on white birches. Journal of Forestry 73:234-237.
Gant, R. E., and E. E. C. Clebsch. 1975. The allelopathic influences of Sassafras albidum in old-field succession in Tennessee. Ecology 56:604-615.
Gartner, B. L. 1991a. Structural stability and architecture of vines vs. shrubs of poison oak, Toxicodendron diversilobum. Ecology 72:2005-2015.
-----. 1991b. Relative growth rates of vines and shrubs of western poison oak, Toxicodendron diversilobum (Anacardiaceae). American Journal of Botany 78:1345-1353.
Gentry, A. H. 1991. The distribution and evolution of climbing plants. Pages 3-52 in F. E. Putz and H. A. Mooney, editors. The biology of vines. Cambridge University Press, Cambridge, England.
Hegarty, E. E. 1991. Vine-host interactions. Pages 357-376 in F. E. Putz and H. A. Mooney, editors. The biology of vines. Cambridge University Press, Cambridge, England.
Hegarty, E. E., and G. Caballe. 1991. Distribution and abundance of vines in forest communities. Pages 313-336 in F. E. Putz and H. A. Mooney, editors. The biology of vines. Cambridge University Press, Cambridge, England.
Heijl, A. M., F. A. Einhellig, and J. A. Rasmussen. 1993. Effects of juglone on growth, photosynthesis, and respiration. Journal of Chemical Ecology 19:559-568.
Hooks, D. D., and J. Stubbs. 1967. An observation of understory growth retardation under three species of oaks. United States Department of Agriculture Forest Service Research Note SE-70.
Kato, R., Y. Tadaki, and H. Ogawa. 1978. Plant biomass and growth increment studies in Pasoh forest. Malayan Nature Journal 30:211-224.
Laman, T. G. 1994. The ecology of strangler figs (hemiepiphytic Ficus spp.) in the rain forest canopy of Borneo. Dissertation. Harvard University, Cambridge, Massachusetts, USA.
Lee, D. W., and J. H. Richards. 1991. Heteroblastic development in vines. Pages 205-243 in F. E. Putz and H. A. Mooney, editors. The biology of vines. Cambridge University Press, Cambridge, England.
Lodhi, M. A. K. 1976. Role of allelopathy as expressed by dominating trees in a lowland forest in controlling the productivity and pattern of herbaceous growth. American Journal of Botany 63:1-8.
-----. 1978. Allelopathic effects of decaying litter of dominant trees and their associated soil in a lowland forest community. American Journal of Botany 65:340-344.
Lutz, H. J. 1943. Injuries to trees caused by Celastrus and Vitis. Bulletin of the Torrey Botanical Club 70:436-439.
McPherson, J. K., and G. L. Thompson. 1972. Competitive and allelopathic suppression of understory by Oklahoma oak forests. Bulletin of the Torrey Botanical Club 99:293-300.
Merck Index. Eleventh edition. 1989. Page 1556. Merck, Rahway, New Jersey, USA.
Neave, I. A., and J. O. Dawson. 1989. Juglone reduces growth, nitrogenase activity, and root respiration of actinorhizal black alder seedlings. Journal of Chemical Ecology 15:1823-1836.
Nicholson, D. I. 1958. An analysis of logging damage in tropical rain forest, North Borneo. Malayan Forester 21:235-245.
Parker, G. G. 1995. Structure and microclimate of forest canopies. Pages 73-108 in M. D. Lowman and N. M. Nadkarni, editors. Forest canopies: a review of research on a biological frontier. Academic Press, New York, New York, USA.
Putz, F. E. 1980. Lianas vs. trees. Biotropica 12:224-225.
-----. 1983. Liana biomass and leaf area of a 'tierra firme' forest in the Rio Negro basin, Venezuela. Biotropica 15:185-189.
-----. 1984a. How trees avoid and shed lianas. Biotropica 16:19-23.
-----. 1984b. The natural history of lianas on Barro Colorado Island, Panama. Ecology 65:1713-1724.
-----. 1991. Silvicultural effects of lianas. Pages 493-501 in F. E. Putz and H. A. Mooney, editors. The biology of vines. Cambridge University Press, Cambridge, England.
Putz, F. E., and N. H. Holbrook. 1989. Strangler fig rooting habits and nutrient relations in the llanos of Venezuela. American Journal of Botany 76:781-788.
Putz, F. E., H. S. Lee, and R. Goh. 1984. Effects of post-felling silvicultural treatment on woody vines in Sarawak. Malaysian Forester 47:214-226.
Radford, A. E., H. E. Ahles, and C. R. Bell. 1968. Manual of the vascular flora of the Carolinas. University of North Carolina Press, Chapel Hill, North Carolina, USA.
Rice, E. L. 1984. Allelopathy. Second edition. Academic Press, New York, New York, USA.
Schlesinger, W. H., and P. L. Marks. 1977. Mineral cycling and the niche of Spanish moss, Tillandsia usneoides L. American Journal of Botany 64:1254-1262.
Stevens, G. C. 1987. Lianas as structural parasites: the Bursera simaruba example. Ecology 68:77-81.
Talley, S. M., W. N. Setzer, and B. R. Jackes. In press. Host associations of two adventitious-root-climbing vines in a North Queensland tropical rainforest. Biotropica.
Tekintas, E., A. Tanrisiver, K. Mendilcioglu, and A. Inal. 1988. Effect of juglone on seed germination. Ege Universite Ziraat Fakulte Dergi 25:203-213 (Turkish); Chemical Abstracts (1990) 113:112609.
Teramura, A. H., W. G. Gold, and I. N. Forseth. 1991. Physiological ecology of mesic, temperate woody vines. Pages 245-285 in F. E. Putz and H. A. Mooney, editors. The biology of vines. Cambridge University Press, Cambridge, England.
Todzia, C. 1986. Growth habits, host tree species, and density of hemiepiphytes on Barro Colorado Island, Panama. Biotropica 18:22-27.
USDA. 1974. Seeds of woody plants in the United States. United States Department of Agriculture, Forest Service, Agricultural Handbook 450.
Whittaker, R. H., and P. P. Feeny. 1971. Allelochemics: chemical interactions between species. Science 171:757-770.
|Printer friendly Cite/link Email Feedback|
|Author:||Talley, Sharon M.; Lawton, Robert O.; Setzer, William N.|
|Date:||Jun 1, 1996|
|Next Article:||Luxury uptake and storage of nitrogen in the rhizomatous alpine herb, Bistorta bistortoides.|