2088 American Journal of Botany 101 ( 12 ): 2088 – 2096 , 2014 ; http://www.amjbot.org/ © 2014 Botanical Society of America American Journal of Botany 101 ( 12 ): 2088 – 2096 , 2014 . Lianas are ubiquitous elements of tropical and temperate for- ests and are well distributed across plant families ( Schnitzer and Bongers, 2002 ). Similar to the case for trees, liana taxo- nomic diversity should refl ect a wide range of functional types and regeneration strategies ( Gallagher et al., 2011 ), which changes as succession progresses ( Letcher and Chazdon, 2012 ). However, most lianas are considered typical pioneers based on their photosynthetic performance, fast growth, and preference for colonizing disturbed habitats, forest edges, canopy gaps, and the top of the canopy ( Avalos and Mulkey, 1999a ; Sanches and V álio, 2002 ; Avalos et al., 2007 ; Toledo-Aceves and Swaine, 2008 ; Ledo and Schnitzer, 2014 ). Despite their pioneer character as adults, many lianas germinate in the shaded forest understory, and as seedlings, they start as self-supporting plants and can tolerate variable periods of deep shade while storing resources to climb to the forest canopy ( Nabe-Nielsen, 2002 ; Feild and Balun, 2008 ; Sanches and V álio, 2002 ; Letcher and Chazdon, 2012 ; Celis and Avalos, 2013 ). It is not clear how consistently lianas maintain their pattern of resource use, habitat choice, and regeneration strategy throughout their ontogeny. Adult lianas respond rapidly to light changes and heterogeneous light conditions ( Avalos and Mulkey, 1999a ; Ledo and Schnitzer, 2014 ) by investing more biomass in leaves relative to supporting tissues as compared with trees ( Putz, 1984 ; Castellanos, 1991 ), which maximizes light interception, enhances mobility, and facilitates the coloni- zation of disturbed habitats ( Paul and Yavitt, 2011 ). In contrast, seedlings may show very restricted physiological responses not necessarily comparable with those of adults, especially when resources are limited. Therefore, it is reasonable to expect the expression of ontogenetic niche shifts, as lianas move from shaded conditions in the understory to well-lit conditions along forest edges, canopy gaps, and the top of the canopy. However, very few studies have discussed differences in physiological performance among different life stages in lianas, from the un- derstory to the canopy. The evidence for ontogenetic niche shifts in tropical plants is still inconclusive ( Poorter et al., 2005 ; Gilbert et al., 2006 ), and it is likely that most species have inter- mediate light requirements and benefi t from moderate light early in life ( Wright et al., 2003 ). In this study, we measured the light acclimation capacity of young seedlings of three twining lianas in response to contrasting 1 Manuscript received 22 March 2014; revision accepted 4 November 2014. This study was fi nancially supported by a Mellon Predoctoral Fellowship to G.A., the International Center for Tropical Ecology, and the Graduate School of the University of Missouri-St. Louis. The authors express their gratitude to S. J. Wright, the Smithsonian Tropical Research Institute, and the managers of Parque Natural Metropolitano for logistic support. Brandon Pratt and two anonymous reviewers edited and signifi cantly improved the initial manuscript. 5 Author for correspondence (e-mail: avalos@fi eldstudies.org) doi:10.3732/ajb.1400127 PHOTOSYNTHETIC AND MORPHOLOGICAL ACCLIMATION OF SEEDLINGS OF TROPICAL LIANAS TO CHANGES IN THE LIGHT ENVIRONMENT 1 GERARDO AVALOS 2,4,5 AND STEPHEN S. MULKEY 3 2 Escuela de Biología, Universidad de Costa Rica 11501-2060 San Pedro, San José, Costa Rica; 3 Unity College, 90 Quaker Hill Road, Unity, Maine 04988 USA; and 4 The School for Field Studies, Center for Sustainable Development Studies, 100 Cummings Center, Suite 534-G Beverly, Massachusetts 01915-6239 USA  Premise of the study: Few studies have analyzed the physiological performance of different life stages and the expression of ontogenetic niche shifts in lianas. Here, we analyzed the photosynthetic and morphological acclimation of seedlings of Stigma- phyllon lindenianum , Combretum fruticosum , and Bonamia trichantha to distinctive light conditions in a tropical dry forest and compared their response with the acclimation response of adult canopy lianas of the same species. We expected acclimation to occur faster through changes in leaf photochemistry relative to adaptation in morphology, consistent with the life history strate- gies of these lianas.  Methods: Seedlings were assigned to the following light treatments: high light (HH), low light (LL), sun to shade (HL), and shade to sun (LH) in a common garden. After 40 d, HL and LH seedlings were exposed to opposite light treatments. Light re- sponse curves, the maximum photosynthetic rate in the fi eld ( A max ), and biomass allocation were monitored for another 40 d on leaves expanded before transfer.  Key results: Photosynthetic responses, A max , and biomass of Stigmaphyllon and Combretum varied with light availability. Physiological characters were affected by current light environment. The previous light environment (carryover effects) only infl uenced A max . Morphological characters showed signifi cant carryover effects. Stigmaphyllon showed high morphological and physiological plasticity. Sun-exposed seedlings of this liana increased stem biomass and switched from self-supporting to climbing forms.  Conclusions: Acclimation in seedlings of these lianas is consistent with the response of adult lianas in the canopy in direction, but not in magnitude. There was no evidence for ontogenetic niche shifts in the acclimation response. Key words: canopy ecology; lianas; ontogenetic niche shift; photosynthetic plasticity; Panama; seedling ecology; shade avoidance; tropical dry forests. AVALOS AND MULKEY—LIGHT ACCLIMATION IN LIANA SEEDLINGS 2089December 2014] Sequence of sun and shade treatments — After developing the third leaf node (8–10 cm in stem length), 1-mo-old seedlings were randomly assigned to one of four light sequences (HH, HL, LH, and LL) in a common garden set up at Parque Metropolitano. Seedlings of all species were self-supporting when experiments started. Lianas under the high to low sequence (HL) began under sun and after 40 d were exposed to shade by relocating the shade cloth during the second part of the experiment. Responses to new light conditions were monitored for another 40 d. Seedlings under the low to high sequence (LH) were subjected to the opposite transfer. Seedlings in the high light (HH) and low light groups (LL) were maintained only under sun or shade during the entire experimental period. The experiment lasted 10 and a half weeks and included 76 lianas in Combretum , 72 in Bonamia , and 109 in Stigmaphyllon after remov- ing damaged seedlings by herbivory or manipulation. Plants were watered when necessary, and sprayed with an insecticide when signs of herbivory were detected. Common garden experiment — Transplant to the common garden took place in May of 1997 (start of wet season). Tree trunks and other plant debris were cleared from an area of 30 × 30 m to expose the soil surface. The fi rst 40 cm of soil of an area of 20 × 4.5 m were removed, homogenized, and cleared of roots to provide homogeneous conditions for seedling planting. In addition, branches from neighboring trees and lianas overtopping the experimental area were cut to secure homogeneous illumination. Shade treatments were created with 63% neutral greenhouse shade cloth put over shade houses made of bam- boo poles 1.8 m tall. Seedlings were planted in four contiguous subplots of 5 × 4.5 m, keeping homogeneous soil and light conditions. To quantify relative differences in photosynthetic photon fl ux density (PFD) among plots, we used four quantum light sensors (Li-190SA, Li-COR, Lincoln, Nebraska, USA) positioned on the ground in the middle of each plot. The light sensors were hooked to a Li-1000 datalogger (Li-COR) programmed to scan PFD levels at 1-min intervals, starting at 8:00 and ending at 16:00 solar time. Since the goal was to measure relative differences in PFD to categorize plots rather than to compile a record of light variation at the site, the measurements were done for 2 days 2 weeks before, and for 2 days 2 weeks after the switch in light conditions. Before the light transfer, HH and HL plots received an average of 720 and 1045 µmol PFD m -2 s −1 , whereas LL and LH plots received 136 and 270 µmol PFD m -2 s −1 , respectively. This was suffi cient to separate high light from low light plots. The distribution of PFD during the day differed little for plots exposed to similar light conditions. After the transfer of light treatments, average PFD levels of sun environments (HH and LH) were close to 1700 µmol PFD·m −2 s −1 , whereas in shade treatments irradiance values remained under 100 µmol PFD·m −2 s −1 . Response variables — We monitored changes in leaf structure, physiology, and patterns of biomass allocation in seedlings before and after the switch in light treatments. Physiological measurements were restricted to those leaves fl ushed and fully expanded under sun or shade at the start of the experiment. Although new leaves were produced after the transfer of light treatments, they were not present in suffi cient numbers across light sequences and, thus, were not included in the analyses of photosynthetic variation. Morphological traits included the total biomass distributed to roots, stems, and petioles, leaves, root to shoot ratio, leaf to stem ratio, and leaf to total mass ratio as determined from the individual dry mass for half of the replicates in each treatment, harvested after 40 d of development before the light switch, and at the end of the experi- ment, 40 d after the light transfer. We took special care in collecting all root material (including fi ne roots) by washing them in situ. All plant material used in biomass measurements was stored and dried in an oven at 60 ° C for 2 d until constant mass prior to measurement. Stem length was measured at the beginning and at the end of the second part of the experiment during the light transfer. Photosynthetic light response curves — To characterize the photosynthetic capacity of seedlings, we collected leaf discs right after dawn from fully ex- panded leaves, then stored them in a darkened, humidifi ed container. This mate- rial was brought immediately to the laboratory for measurement of dark respiration rate (µmol CO 2 ·m −2 s −1 ), light compensation point (µmol PFD·m −2 s −1 ), apparent quantum yield ( Q ), and photosynthetic rate at light saturation (µmol O 2 ·m −2 s −1 ) from light response curves measured with an oxygen electrode (Hansatech In- struments, Norfolk, UK) using 10% CO 2 at 28 ° C. Leaf specifi c mass (LSM) was determined from the leaf discs used in the oxygen electrode. Two sets of light curves were obtained, one before and one after the transfer of light treat- ments. Oxygen electrodes measure the rate of oxygen evolution, which is light conditions and compared their response with the acclima- tion response of adults of the same species described previously ( Avalos and Mulkey, 1999a ; Avalos et al., 2007 ). Most studies on seedling ecology have focused on the responses of already established seedlings, or juveniles (sensu Garwood, 1996 ; Whitmore, 1996 ; Baraloto et al., 2005 ), whereas very few have examined the implications of initial morphology on seedling establishment and its role determining the ecological amplitude of tropical species ( Kitajima, 1994 , 1996a , b ), especially when compared with adults. Our aim here is to evaluate the consis- tency of the regeneration span of these lianas in which adults behave like light-demanding species ( Avalos and Mulkey, 1999a ; Avalos et al., 2007 ). We expect initial light conditions to determine the extent of adjustment to subsequent light changes. Thus, the magnitude of phenotypic differences in response to light changes should be higher in seedlings that started growth under high light. Light- demanding species will perform poorly in the shade, while spe- cies able to withstand the shade will express more conservative strategies, showing fewer phenotypic differences across habi- tats. Expanding our knowledge on the range of responses ex- pressed by lianas to the continuum of light environments found in tropical forests, especially in early regeneration stages, is crucial to understand the increased abundance of lianas in neo- tropical forests ( Wright et al., 2004 ) and their role in carbon cycling and to predict the future behavior of rainforest ecosys- tems under global warming scenarios ( Schnitzer and Bongers, 2002 ; Malhi and Phillips, 2004 ; Malhi, 2012 ; Van der Heijden et al., 2013 ). MATERIALS AND METHODS Study site — This research was conducted in central Panama at Parque Natu- ral Metropolitano, a 100–150-yr-old tropical dry forest. The site has an average annual rainfall of 1740 mm and a well-defi ned dry season (December through late April). Here, the Smithsonian Tropical Research Institute maintains a can- tilevered construction crane that allows repeated, nondestructive access to the upper canopy from a suspended cage ( Parker et al., 1992 ). Study species — The neotropical liana Stigmaphyllon lindenianum A. Juss. (Malpighiaceae) is abundant throughout the study area as well as in trees under the crane. Combretum fruticosum (Loefl .) Stunz (Combretaceae) and Bonamia trichantha Hallier f. (Convolvulaceae) were the most abundant twining lianas on trees under the crane during the wet season of 1994 and the dry season of 1995 ( Avalos and Mulkey, 1999b ; Avalos et al., 2007 ). Canopy lianas of C. fruticosum and B. trichantha show alternative leaf phenotypes. Leaves pro- duced during the wet season have lower photosynthetic capacity relative to those produced during the dry season ( Avalos et al., 2007 ). Although the infor- mation on the ecology of seedlings is limited, we have found that S. lindenia- num and C. fruticosum germinate both in large gaps and in shaded environments, whereas B. trichantha is found only in conditions of direct sunlight. As adults, the three species are associated with high light habitats. All are twining lianas with very narrow stems (~1 cm in diameter). Hereafter, the species are referred to by their generic designation. Growth conditions — Seeds were collected from at least 10 reproductive in- dividuals throughout Parque Metropolitano and from lianas growing on trees under the crane during the dry season. Seeds were taken to the greenhouse of the Smithsonian Tropical Research Institute in Panama City and allowed to germinate under moderate sun conditions in fl ats fi lled with alluvial soil taken from the fi eld site. Seeds were watered daily via a sprinkler system. Combretum and Bonamia started to germinate after 10–12 d, whereas Stigmaphyllon required 28 d to emerge. Immediately after germination, each seedling was transplanted to a fl exible plastic bag with a mixture of vermiculite and peat moss soil. These containers were large enough to minimize the adverse effects of root binding and provided appropriate conditions before transplant into the common garden. 2090 AMERICAN JOURNAL OF BOTANY [Vol. 101 treatment, β j is the effect the previous light treatment, ( t β ) ij is the interaction between previous and present light treatments (i.e., the effect of light sequence), and e ijk is the error term. The fi rst crossover model tested the signifi cance of present and previous light treatments on the principal components emerging from physiological variables, whereas the second model tested for differences in the components coming from morphological characters. We considered ac- climation to low and high light to occur if the response of HL and LH seedlings matched that of LL and HH seedlings after the transfer. Principal components were normalized using the Box–Cox method when necessary ( Quinn and Keough, 2002 ) to conform to the normal distribution and equality of variances. All sta- tistical analyses were done using JMP 10.0 Statistical Software (SAS Institute, Cary, North Carolina, USA). RESULTS Photosynthetic responses before transfers — Photosyn- thetic rates at light saturation in the three lianas varied little among light treatments before the transfer ( Table 1 ) . The low- est rates were observed in Bonamia . Q , LSM, compensation points, and respiration rates showed little variation among treatments. Before the light transfer, we found two principal components, which explained 64% of the variation in photo- synthetic parameters. The fi rst component (40.56% of the variation) was dominated by the photosynthetic rate at light saturation and Q , whereas the second component (23.65%) was dominated by respiration rate. Both components were normally distributed. The fi rst component showed no differ- ences between species and light environments (two-way ANOVA, r 2 = 0.15, F 5,33 = 1.17, P = 0.34), whereas the second component showed signifi cant differences only among spe- cies ( r 2 = 0.32, F 2,33 = 4.88, P < 0.01), with shade seedlings of Bonamia diverging form seedlings of Combretum and Stigma- phyllon (Tukey HSD, P < 0.05). The magnitude of A max was higher under high light treat- ments for all species, with Stigmaphyllon showing the highest rates, followed by Combretum ( Fig. 1A ) . Differences were stoichiometrically equivalent to the rate of carbon fi xation. We also measured A max (the maximum photosynthetic rate under fi eld conditions in µmol CO 2 ·m −2 s −1 ) using fully expanded leaves and a portable photosynthesis system (Li-6400, Li-COR) set to an ambient CO 2 concentration of 386 µmol, a water vapor concentration of 24–28 mol, and a light intensity of 1500 µmol PFD·m −2 s −1 keeping the temperature in the cuvette close to ambient tempera- ture (30 ° C). Statistical analyses and test of hypotheses — Before transfers— Suites of morphological and photosynthetic characters were highly correlated. Thus, we consolidated these variables using a principal component analysis (PCA). This procedure removed the correlation among variables and improved the statistical power of subsequent analyses. Differences in morphological and photosyn- thetic characters between sun and shade treatments were tested using a two-way ANOVA applied to the scores of the principal components. We used two ANOVA models. The fi rst one measured differences in the principal compo- nents resulting from photosynthetic characters (dark respiration rate, light com- pensation point, Q , photosynthetic rate at light saturation and LSM), whereas the second model tested differences in the components derived from morpho- logical characters between sun and shade environments. Root to shoot and leaf to mass ratios were analyzed separately from the rest of the biomass allocation variables since they resulted from the combination of variables already in- cluded in the PCA. Tukey’s honestly signifi cant difference (HSD) was used as post hoc test after fi nding signifi cant main effects. After transfers— During the second part of the experiment, a crossover de- sign was used to test the effects of present light environment (light treatment after the transfer), the effects of exposure to previous light conditions (carryover effects), and the interaction between previous and present light treatments (which is analogous to a test of the effect of light sequence) on physiological and mor- phological characters. Carryover effects are the effects of the fi rst light environ- ment that persist and infl uence phenotypic responses in subsequent periods of measurement ( Ratkowsky et al., 1993 ). We used the following crossover design: N C C Spijk i j j ijkijY t t e , where Y ijk is the response of the seedling of the species i within the current light environment j after being exposed to the previous light conditions k , μ is the population parameter, Sp i is the species effect, t j is the effect of present light TABLE 1. Photosynthetic responses in seedlings of the lianas Bonamia (BM), Combretum (CF), and Stigmaphyllon (SL) before and after the light transfer in a common garden set in Parque Natural Metropolitano, Panama. Values are averages ( ± 1 SE); N = sample size. Treatment and species Q (apparent quantum yield) LSM (cm 2 ·g −1 ) Saturation (µmol O 2 ·m 2 s −1 ) Compensation (µmol PFD·m 2 s −1 ) Respiration (µmol CO 2 ·m 2 s −1 ) N High light BM 0.01 (0.004) 0.003 (0.0004) 5.4 (0.7) 10.88 (1.7) −2.1 (0.2) 6 CF 0.03 (0.0006) 0.004 (0.0004) 9.71 (1.2) 14.22 (1.98) −2.16 (0.3) 7 SL 0.03 (0.004) 0.005 (0.004) 9.28 (1.6) 13.47 (2.89) −1.2 (0.2) 7 Low light BM 0.02 (0.008) 0.004 (0.0004) 5.6 (1.11) 15.71 (3.84) −3.08 (0.74) 6 CF 0.02 (0.002) 0.006 (0.003) 7.43 (0.8) 13.02 (1.81) −1.70 (0.21) 7 SL 0.02 (0.006) 0.001 (0.0002) 7.37 (1.71) 18.18 (2.66) −1.57 (0.4) 7 Permanent high light BM 0.03 (0.007) 0.004 (0.0006) 10.5 (2.47) 2.6 (1.8) −0.84 (0.42) 3 CF 0.04 (0.004) 0.005 (0.0005) 14.44 (1.7) 13.2 (0.93) −2.14 (0.28) 4 SL 0.02 (0.001) 0.003 (0.0004) 13.7 (2.01) 22.44 (2.82) −1.5 (0.11) 5 High to low light BM 0.03 (0.002) 0.004 (0.0003) 7.08 (0.2) 5.4 (3.5) −2 (0.008) 2 CF 0.04 (0.001) 0.004 (0.0002) 10.5 (0.93) 8 (0.66) −1.28 (0.12) 4 SL 0.03 (0.004) 0.002 (0.0004) 7.41 (0.41) 15.9 (7.1) −1.18 (0.4) 2 Low to high light BM 0.01 (0.006) 0.004 (0.001) 7.15 (1.94) 2.9 (1.53) −1.3 (0.41) 4 CF 0.04 (0.004) 0.004 (0.0002) 10.79 (1.86) 12.56 (1.85) −2 (0.2) 4 SL 0.04 (0.005) 0.003 (0.0001) 12.56 (1.6) 18.45 (2.02) −2.24 (0.34) 4 Permanent low light BM 0.01 (0.004) 0.004 (0.0001) 4.8 (0.26) 1.73 (1.63) −0.8 (0.33) 3 CF 0.04 (0.006) 0.003 (0.0001) 7.11 (0.98) 8.83 (1.68) −1.52 (0.14) 3 SL 0.03 (0.002) 0.001 (0.0002) 7.67 (3.18) 11.6 (1.15) −2.16 (0.9) 3 AVALOS AND MULKEY—LIGHT ACCLIMATION IN LIANA SEEDLINGS 2091December 2014] F 2,29 = 16.38, P < 0.0001) with Stigmaphyllon diverging from Combretum and Bonamia . The A max responses after the light transfer maintained the overall differences among species and light treatments ob- served before the light transfer, with Stimaphyllon showing the highest rates under HH ( Fig. 1B ). Stimaphyllon varied A max as a function of light availability, decreasing A max under shade and increasing it under high light. Combretum maintained low A max under shade, whereas LH seedlings did not surpass A max relative to HL seedlings. In Bonamia A max was higher under HH, but no differences were observed among the rest of the light treat- ments ( Fig. 2 ). The lianas responded to previous as well as to present light treatments ( Table 2 ) , but current light conditions had a stronger effect relative to the previous light environment. Biomass distribution before transfers — Access to high light increased seedling biomass across species ( Table 3 ) . Under high light, Stigmaphyllon and Combretum more than doubled the biomass of Bonamia , whereas under shade Stigmaphyllon had the lowest biomass. Biomass allocated to leaves under high light was the dominant allocation compartment in Stigmaphyl- lon and Combretum . Bonamia maintained consistently small signifi cant among species (two-way ANOVA, r 2 = 0.67, F 2,105 = 27.82, P < 0.0001) and light treatments ( F 1,105 = 7.06, P = 0.009). Photosynthetic responses after transfers — We extracted two principal components from the second set of photosyn- thetic parameters, which explained 71% of the variation. The fi rst component (47%) was dominated by photosynthetic rate at light saturation, whereas the second one (24%) was dominated by LSM. Both components were normally distributed. In the crossover design for the fi rst component ( r 2 = 0.61) there was a signifi cant species effect ( F 2,29 = 9.48, P = 0.0007), with Bon- amia separating from Combretum and Stigmaphyllon (Tukey HSD, P < 0.05; Fig. 2 ) . The current light treatment had a sig- nifi cant effect on this component ( F 1,29 = 7.4, P = 0.01), whereas carryover effects were not signifi cant. Stigmaphyllon and Com- bretum varied their photosynthetic responses as a function of increasing or decreasing light ( Fig. 2 ). The second component presented signifi cant differences only among species ( r 2 = 0.57, Fig. 1. Differences among species and light treatments in A max (µmol CO 2 ·m −2 ·s −1 ) (A) before and (B) after the light transfer treatments. Abbre- viations refer to species names (SL: Stigmaphyllon , CF: Combretum , BM: Bonamia ). Letters above bars indicate statistical signifi cance at P < 0.05 following Tukey’s HSD test. Error bars refer to ± 1 SE. Fig. 2. Differences among species and light treatments in photosyn- thetic variables combined in a principal component dominated by the pho- tosynthetic rate at light saturation (µmol O 2 ·m −2 s −1 ) before the light transfers. Abbreviations refer to species names as in Fig. 1 . Letters above bars indicate statistical signifi cance at P < 0.05 following Tukey’s HSD test. Error bars refer to ± 1 SE. TABLE 2. Crossover design measuring the effects of exposure to present and previous light treatments (carryover effects) on A max (µmol CO 2 ·m −2 s −1 ) in seedlings of Stigmaphyllon , Combretum , and Bonamia in a common garden experiment set in Parque Natural Metropolitano, Panama ( r 2 = 0.83). Source F ratio P Species F 2,373 = 186.98 <0.0001 Previous light treatment (carryover effects) F 1,373 = 162.60 <0.0001 Current light treatment F 1,373 = 226.37 <0.0001 Previous light treatment × species F 2,373 = 22.28 <0.0001 Current light treatment × species F 2,373 = 57.84 <0.0001 Light sequence F 2,373 = 11.82 0.0007 Light sequence × species F 2,373 = 4.17 0.01 2092 AMERICAN JOURNAL OF BOTANY [Vol. 101 TABLE 3. Biomass distribution in liana seedlings before and after the light transfer under high and low light environments. Abbreviations refer to species names (SL: Stigmaphyllon , CF: Combretum , BM: Bonamia ). Values are averages ( ± 1 SE) of dry mass in grams; N = sample size. Species Leaves Stems Roots Biomass Root/shoot Leaf/mass N High light BM 0.024 (0.004) 0.046 (0.004) 0.06 (0.003) 0.17 (0.02) 0.7 (0.08) 0.15 (0.02) 20 CF 0.17 (0.01) 0.06 (0.006) 0.07 (0.006) 0.30 (0.02) 0.31 (0.02) 0.54 (0.01) 27 SL 0.22 (0.03) 0.12 (0.01) 0.08 (0.01) 0.42 (0.06) 0.25 (0.02) 0.48 (0.01) 46 Low light BM 0.028 (0.002) 0.04 (0.002) 0.047 (0.003) 0.14 (0.006) 0.5 (0.02) 0.20 (0.01) 40 CF 0.093 (0.01) 0.041 (0.005) 0.054 (0.01) 0.18 (0.02) 0.43 (0.06) 0.48 (0.01) 29 SL 0.033 (0.003) 0.02 (0.001) 0.022 (0.002) 0.08 (0.007) 0.42 (0.02) 0.38 (0.02) 35 Permanent high light BM 0.04 (0.01) 0.06 (0.01) 0.06 (0.005) 0.18 (0.02) 0.6 (0.08) 0.19 (0.05) 3 CF 0.25 (0.06) 0.12 (0.02) 0.10 (0.02) 0.48 (0.11) 0.3 (0.03) 0.50 (0.02) 10 SL 0.76 (0.12) 0.55 (0.11) 0.41 (0.06) 1.73 (0.28) 0.3 (0.06) 0.45 (0.02) 21 High to low light BM 0.02 (0.004) 0.05 (0.008) 0.04 (0.01) 0.13 (0.02) 0.59 (0.1) 0.12 (0.01) 3 CF 0.13 (0.02) 0.07 (0.01) 0.08 (0.01) 0.28 (0.06) 0.36 (0.04) 0.51 (0.01) 9 SL 0.2 (0.02) 0.14 (0.04) 0.06 (0.01) 0.4 (0.1) 0.21 (0.02) 0.50 (0.01) 17 Low to high light BM 0.03 (0.003) 0.04 (0.003) 0.04 (0.001) 0.14 (0.008) 0.48 (0.02) 0.21 (0.02) 17 CF 0.04 (0.01) 0.02 (0.01) 0.03 (0.01) 0.10 (0.01) 0.41 (0.04) 0.44 (0.01) 4 SL 0.04 (0.004) 0.04 (0.01) 0.02 (0.001) 0.10 (0.01) 0.29 (0.04) 0.34 (0.04) 13 Permanent low light BM 0.02 (0.005) 0.04 (0.008) 0.06 (0.006) 0.16 (0.01) 0.66 (0.06) 0.14 (0.01) 9 CF 0.08 (0.01) 0.04 (0.006) 0.06 (0.01) 0.18 (0.03) 0.52 (0.04) 0.45 (0.02) 11 SL 0.06 (0.01) 0.05 (0.009) 0.04 (0.008) 0.14 (0.02) 0.4 (0.09) 0.42 (0.01) 11 seedlings across treatments. We consolidated biomass distribu- tion (mass of leaves, stems, roots, and overall biomass) in one principal component, which explained 86% of the variation across species. The coeffi cients of the eigenvectors for this component were similar among variables and ranged 0.45 to 0.53 indicating high correlation among morphological vari- ables. Accordingly, we termed this component “biomass”. This component was normalized with the Box–Cox transformation and was entered as a response variable into a two-way ANOVA using species, light treatment, and their interaction, as main ef- fects. Differences were found among species and light treat- ments ( r 2 = 0.56, Table 4 , Fig. 3A ) . Stigmaphyllon showed the steepest differences between sun and shade groups, Combretum had moderate differences, and Bonamia did not show differ- ences among light environments ( Fig. 3A ). Biomass distribution after transfers — After the light trans- fer, biomass distribution was consolidated into one principal component, which explained 91% of the variation. The load- ings of this component ranged from 0.47 to 0.52, showing high correlation among roots, leaves, stems, and overall biomass. The scores of this component were normalized using the Box– Cox procedure. A two-way ANOVA ( Table 5 , Fig. 3B ) applied to the transformed scores showed signifi cant differences among species, light sequences, and their interaction. The highest biomass was found in Stigmaphyllon under continuous high light. Both Stigmaphyllon and Combretum varied the magnitude of bio- mass as a function of light conditions, although LL seedlings showed higher biomass than LH seedlings in both lianas. Bonamia showed a weaker response to differences in light availability after the transfer, and no differences were found among treatments ( Fig. 3B ). Current light treatment did not af- fect biomass allocation. None of the effects for which the cur- rent light treatment was involved infl uenced seedling biomass ( Table 5 ). Similar to Bonamia and Combretum , shade seedlings of Stigmaphyllon showed little growth over the course of the experiment. Root to shoot ratio — Root to shoot ratios were higher in sun seedlings of Bonamia relative to Combretum and Stigmaphyl- lon before the transfer ( Table 3 , r 2 = 0.33, F 2,123 = 26.89, P < 0.0001). Differences were found among species under sun con- ditions, whereas shade seedlings of Stigmaphyllon and Com- bretum had similar root to shoot ratios (Tukey HSD, P < 0.05). The tendency was maintained after the transfer in light treat- ments. Root to shoot ratios were transformed using the Box– Cox procedure to correct for lack of normality. There was only a signifi cant effect of species ( Table 3 , r 2 = 0.45, F 2,176 = 32.20, P < 0.0001), with Bonamia showing higher ratios relative to Stigmaphyllon and Combretum . Differences among sequences and carryover effects were not signifi cant. Leaf to mass ratio — This variable showed a strong effect of species ( F 2,123 = 202.51, P < 0.0001), light treatment ( F 1,123 = 7.96, P < 0.006) and the interaction species × light treatment ( F 2,123 = 9.43, P < 0.0002) before the light transfer ( Table 3 ; Fig. 4 ) . The highest leaf to mass ratios were found in Combre- tum and Stigmaphyllon under sun and shade conditions, whereas Bonamia showed the lowest ratio ( Fig. 4A ). After the transfer, differences were signifi cant among species ( F 2,175 = 135.86, P < 0.0001; Fig. 4B ) with Combretum and Stigmaphyllon showing TABLE 4. Two-way ANOVA measuring the effects of species and light treatment on biomass allocation before the light transfer in seedlings of Stigmaphyllon , Combretum , and Bonamia in a common garden set in Parque Natural Metropolitano, Panama. Source F ratio P Species F 2,123 = 7.51 0.0008 Light treatment F 1,123 = 82.60 <0.0001 Species × light treatment F 2,123 = 26.11 <0.0001 AVALOS AND MULKEY—LIGHT ACCLIMATION IN LIANA SEEDLINGS 2093December 2014] Fig. 3. (A) Differences among species and light treatments of high light (H) and low light (L) in biomass allocation variables consolidated using PCA before the light transfer in Stigmaphyllon (SL), Combretum (CF), and Bonamia (BM). (B) Differences in biomass allocation variables after the light transfer combined into one principal component according to sequence of light treatments. Different letters above bars show statistical signifi cance at P < 0.05 (Tukey’s HSD). Error bars refer to ± 1 SE. TABLE 5. Crossover model measuring the effects of species exposure to present and previous light conditions (carryover effects) on biomass allocation of seedlings of Stigmaphyllon , Combretum , and Bonamia in a common garden experiment set in Parque Natural Metropolitano, Panama ( r 2 = 0.60). Source F ratio P Species F 2,115 = 4.20 0.01 Previous light treatment (carryover effects) F 1,115 = 46.50 <0.0001 Current light treatment F 1,115 = 0.36 0.54 Previous light treatment × species F 2,115 = 9.05 0.0002 Current light treatment × species F 2,115 = 0.81 0.44 Light sequence F 1,115 = 14.76 0.0002 Light sequence × species F 2,115 = 1.02 0.36 Fig. 4. Leaf to mass ratio (A) before and (B) after the light transfer in Stigmaphyllon (SL), Combretum (CF), and Bonamia (BM). Letters above box plots indicate statistical signifi cance at P < 0.05 following Tukey’s HSD test. increase in stem length and the high variation observed in the HH and HL groups where some plants increased stem length by 88% in less than 2 wk. DISCUSSION Acclimation is a complex response whose expression scales with the extent of light changes, spanning several orders of magnitude, from sunfl ecks to canopy gaps, and from days to higher ratios in HH and HL relative to Bonamia . Exposure to previous light environment had a signifi cant effect ( F 1,175 = 8.60, P < 0.003), whereas the current light conditions did not affect leaf to mass ratios. Bonamia showed a weak response to light treatments, and only seedlings under HH had a slightly higher ratio relative to the rest of the light treatments. Changes in stem length — Differences among species in stem length were signifi cant ( r 2 = 0.34, F 2,213 = 27.63, P < 0.0001) and were driven by the increase in stem length of Stig- maphyllon under the HH and HL sequences ( Fig. 5 ) . Stem length was not affected by present light conditions, but was infl uenced by previous light environments ( F 1,213 = 4.97, P < 0.02). HH and HL seedlings were taller than LL and LH seed- lings in Stigmaphyllon ( F 2,213 = 6.37, P < 0.002). Bonamia did not show signifi cant increases in stem length, whereas Comb- reum had shorter seedlings only under the LL treatment. In contrast, in Stigmaphyllon initial exposure to sun conditions fa- cilitated the accumulation of suffi cient biomass to express the climbing response. After the light transfer, 17 Stigmaphyllon seedlings, taller than 15 cm, shifted from self-supporting to climbing forms. These seedlings were responsible for the steep 2094 AMERICAN JOURNAL OF BOTANY [Vol. 101 months ( Pearcy, 2007 ). Changes that take place over days are initially dealt with through the adjustment of the photosynthetic apparatus of mature, expanded leaves, present before the light change. These changes involve the rearrangement of within- leaf resource allocation, including leaf anatomy and nitrogen redistribution ( Hikosaka and Terashima, 1996 ; Evans and Poorter, 2001 ). Morphological adjustment takes place once physiological responses have stabilized after longer time scales (days to weeks) and involve the reorganization of bio- mass allocation, plant architecture and leaf display, as well as the production of leaves adapted to the new light conditions ( Chazdon and Kaufmann, 1993 ; Turnbull et al., 1993 ; Oguchi et al., 2003 , 2005 ). However, in seedlings growing in the shaded understory, small size results in less leaf and root area, limiting the capacity to absorb light and nutrients and modify- ing biomass allocation rapidly. Here, the fi nal acclimation re- sponse will depend on the intensity and direction of the light change, as well as on the seedling capacity to adjust existent foliage to new light conditions, given competing trade-offs among morphological and physiological characters ( Evans and Poorter, 2001 ; Oguchi et al., 2005 ). For these reasons, we expected initial light conditions to determine morphological adjustment and to infl uence the extent of physiological re- sponses to subsequent light changes (signifi cant carryover effects). The overall physiological and morphological responses of the lianas analyzed in this study conformed to these expecta- tions. Photosynthetic variables showed small differences be- tween sun and shade treatments before the light transfer, but after the switch, variables like saturation and respiration rates (integrated into one principal component), moved in the direc- tion of light availability, especially in the lianas Stigmaphyllon and Combretum . Carryover effects on physiological variables were not signifi cant or were weaker than the effect of current light environment, as in the case of A max . Physiological re- sponses were mostly determined by the current light conditions. As discussed later, the physiological and morphological re- sponse of Bonamia was very limited and was more consistent with a stress response, especially within the LL and LH treat- ments, which showed the lowest Q values ( Table 1 ). The direction of the photosynthetic responses in seedlings of Stigmaphyllon was similar to that of adult lianas subjected to the same sequence of light treatments in the canopy ( Avalos and Mulkey, 1999a ). However, seedlings exposed to sun condi- tions had photosynthetic rates at light saturation equivalent in magnitude to that of adult lianas exploiting shaded canopy mi- crohabitats ( Avalos and Mulkey, 1999a ; Avalos et al., 2007 ). Photosynthetic capacities in seedlings were thus signifi cantly limited compared with that of adults. In these seedlings, A max better refl ected the extent of environmental differences, espe- cially in Stigmaphyllon , and was affected by previous light con- ditions, but the last light treatment had a stronger effect. The magnitude of A max was similar to the photosynthetic saturation rates of adult canopy lianas in Stigmapyllon and Combretum . Bonamia, in contrast, showed a very restricted A max and overall photosynthetic responses, even for seedlings maintained under continuous sun or exposed to an initial period of high light. Stigmaphyllon and Combretum acclimated to a light increase or decrease using leaves produced and expanded in the previous light environment, as it has been observed in adults of these species ( Avalos and Mulkey, 1999a ; Avalos et al., 2007 ). In these lianas, adults have shown postexpansion acclimation mostly to a light increase ( Avalos and Mulkey, 1999b ; Avalos et al., 2007 ). Of the three lianas, Bonamia showed the most restricted response. Adults of this species are typical shade- avoiders and aggressively pre-empt the canopy by reaching peaks in leaf production quickly, which creates very dark con- ditions within the canopy of their host trees. In addition, canopy lianas of Bonamia decrease leaf life span and fi nally drop their leaves in shaded microhabitats while increasing leaf production in well-lit sites ( Avalos et al., 2007 ). This behavior corresponds to an opportunistic, shade-avoiding character and implies lack of capacity to adapt already expanded leaves to shade. The lim- ited adjustment to shade in seedlings and canopy lianas is con- gruent with the habitat choice of Bonamia , which prefers forest edges, large gaps, and the top of the canopy. Patterns of biomass allocation were mostly infl uenced by ex- posure to previous light conditions, and thus, carryover effects were signifi cant. Stigmaphyllon was the most sensitive to light differences, increasing biomass in sites of greater irradiance. Combretum followed a similar pattern, but to a lesser degree, since it did not show the high biomass accumulation observed in Stigmaphyllon . Exposure to a previous period of high light facilitated biomass increase in Stigmaphyllon , which extended stem length, became unstable, and shifted from self-supporting to climbing forms. This behavior is similar to that of the eco- typic forms of western poison oak Toxicodendron diversilobum (i.e., Gartner, 1991a , b ), in which vines showed higher leaf bio- mass and longer stems than self-supported shrubs. However, in Ipomoea purpurea , the climbing habit was expressed with more intensity in the shade, where plants showed reduced branching and increased stem length more than plants in the sun ( Gianoli, 2003 ). In this latter study, sun plants were reproductive and thus showed a strikingly different strategy of biomass alloca- tion relative to shade plants. Given suffi cient resources to ac- cumulate biomass, shade plants can develop longer and more slender stems than plants in the sun ( Gartner, 1991b ; Den Dubbelden and Oosterbeek, 1995 ; Leicht-Young et al., 2011 ). In our case, seedlings showed higher respiration rates relative to canopy lianas ( Avalos and Mulkey, 1999a ; Avalos et al., 2007 ) and were clearly limited by reduced biomass accumulation in the shade. Fig. 5. Box plots representing the extent of variation in stem length increase between the beginning and end of the second part of the experi- ment during the light transfer. Abbreviations refer to species names (BM: Bonamia , CF: Combretum , SL: Stigmaphyllon ). Letters above box plots indicate statistical signifi cance at P < 0.05 following Tukey’s HSD test. AVALOS AND MULKEY—LIGHT ACCLIMATION IN LIANA SEEDLINGS 2095December 2014] in abundance and biomass across the tropics ( Schnitzer and Bongers, 2011 ; Laurance et al., 2014 ) and have signifi cant im- pacts on carbon sequestration in these ecosystems ( Van der Heijden et al., 2013 ), the analysis of their adaptation to succes- sional gradients, especially during the initial stages of regenera- tion, is crucial to improve the management of disturbed habitats. This knowledge is becoming urgent as tropical environments are increasingly dominated by secondary forests, a favorite habitat for lianas ( Laurance et al., 2014 ). LITERATURE CITED ANDRADE , J. L. , F. C. MEINZER , G. GOLDSTEIN , AND S. A. SCHNITZER . 2005 . Water uptake and transport in lianas and co-occurring trees of a sea- sonally dry tropical forest. Trees 19 : 282 – 289 . AVALOS , G. , AND S. S. MULKEY . 1999a . Photosynthetic acclimation of the liana Stigmaphyllon lindenianum (Malpighiaceae) to light changes in a tropical dry forest canopy. Oecologia 120 : 475 – 484 . AVALOS , G. , AND S. S. MULKEY . 1999b . Seasonal changes in liana cover in the canopy of a neotropical dry forest. Biotropica 31 : 186 – 192 . AVALOS , G. , S. S. MULKEY , K. KITAJIMA , AND S. J. WRIGHT . 2007 . Colonization Strategies of two liana species in a tropical dry forest canopy. Biotropica 39 : 393 – 399 . BARALOTO , C. , D. E. GOLDBERG , AND D. BONAL . 2005 . Performance trade-offs among tropical tree seedlings in contrasting microhabitats. Ecology 86 : 2461 – 2472 . BAZZAZ , F. A. 1996 . Plants in changing environments . Cambridge University Press, New York, New York, USA. BAZZAZ , F. A. , AND R. W. CARLSON . 1982 . Photosynthetic acclimation to variability in the light environment of early and late successional plants. Oecologia 54 : 313 – 316 . CASTELLANOS , A. E. 1991 . Photosynthesis and gas exchange of vines. In F. E. Putz and H. A. Mooney [eds.], The biology of vines, 181–204. Cambridge University Press, Cambridge, New York, New York, USA. CELIS , G. , AND G. AVALOS . 2013 . Acclimation of seedlings of Gnetum leyboldii Tul. (Gnetaceae) to light changes in a tropical rain forest. Revista de Biología Tropical 61 : 1859 – 1868 . CHAZDON , R. L. , AND S. KAUFMANN . 1993 . Plasticity of leaf anatomy of two rain forest shrubs in relation to photosynthetic light acclimation. Functional Ecology 7 : 385 – 394 . CONDON , M. A. , T. W. SASEK , AND B. R. STRAIN . 1992 . Allocation patterns in two tropical vines in response to increased atmospheric carbon di- oxide. Functional Ecology 6 : 680 – 685 . DALLING , J. W. , K. WINTER , J. D. NASON , S. P. HUBBELL , D. A. MURAWSKI , AND J. L. HAMRICK . 2001 . The unusual life history of Alseis blacki- ana : A shade-persistent pioneer tree? Ecology 82 : 933 – 945 . DEN DUBBELDEN , K. C. , AND B. OOSTERBEEK . 1995 . The availability of ex- ternal support affects allocation patterns and morphology of herba- ceous climbing plants. Functional Ecology 9 : 628 – 634 . EVANS , J. R. , AND H. POORTER . 2001 . Photosynthetic acclimation of plants to growth irradiance: The relative importance of specifi c leaf area and nitrogen partitioning in maximizing carbon gain. Plant, Cell & Environment 24 : 755 – 767 . FEILD , T. S. , AND L. BALUN . 2008 . Xylem hydraulic and photosynthetic function of Gnetum (Gnetales) species from Papua New Guinea. New Phytologist 177 : 665 – 675 . GALLAGHER , R. V. , M. R. LEISHMAN , AND A. T. MOLES . 2011 . Traits and ecological strategies of Australian tropical and temperate climbing plants. Journal of Biogeography 38 : 828 – 839 . GARTNER , B. L. 1991a . Is the climbing habit of poison oak ecotypic? Functional Ecology 5 : 696 – 704 . GARTNER , B. L. 1991b . Structural stability and architecture of vines vs. shrubs of poison oak, Toxicodendron diversilobum. Ecology 72 : 2005 – 2015 . GARWOOD , N. C. 1996 . Functional morphology of tropical tree seedlings. In M. D. Swaine [ed.], The ecology of tropical forest tree seedlings, 59–129. Man and the biosphere series, UNESCO, New York, New York, USA. High respiration rates, combined with a limited leaf and root area, most likely limited biomass increase across light treat- ments, especially in Bonamia . The higher root to shoot ratios of Bonamia relative to Com- bretum and Stigmaphyllon and the relatively low leaf to mass ratio, limits this liana to well-lit environments at the beginning of the regeneration cycle. High root to shoot ratios are indica- tive of habitat specialization, being predominant in species adapted to dry habitats, such as drier slopes or drier forest types ( Garwood, 1996 ). It is likely that plants in the understory of the dry forest of Parque Metropolitano experience prolonged peri- ods of drought during the dry season. Under these conditions, small seedlings that did not produce a large shoot during the wet season benefi t from higher root allocation, which favors access to deeper, more humid soil. For lianas of tropical dry forests (which often die back during the dry season), higher root allocation enhances root storage, which is important to support new growth during the wet period (see Condon et al., 1992 ). Lianas are among the most deeply rooted species in tropical forests ( Holbrook and Putz, 1996 ; Andrade et al., 2005 ), a con- dition that gives them access to the water necessary to maintain an extensive leaf area using very narrow stems. The seedling and adult lianas studied here exhibit charac- teristics typical of pioneer species ( Bazzaz and Carlson, 1982 ; Bazzaz, 1996 ; Wright et al., 2004 ; Paul and Yavitt, 2011 ). In the understory, seedling survival and growth depend on morpho- logical traits that enhance defense against herbivores or patho- gens, such as tougher leaves and a well-developed root system. Similar to our case, high respiration costs for root construction and maintenance result in lower carbon gain in shade-intolerant species (see Kitajima, 1994 ). Acclimation at the physiological level increases the chances for biomass accumulation before the resource is depleted by canopy closures or overtaken by other plants, and buffers seedling mortality in low-resource en- vironments. In contrast, the response to long-term light changes depends on the production of new organs and loss of tissues formed in the previous light environment or on the post expan- sion acclimation of the same tissues ( Kamaluddin and Grace, 1992 ). The fi nal response is determined by the interaction be- tween morphology and function, since acclimation depends on physiological, as well as morphological plasticity. Although there are a few instances in which seedlings of pioneer species can survive prolonged periods in the shade (i.e., Dalling et al., 2001 ), for the most part, light-demanding species perform poorly under prolonged shade. Understanding the expression of niche differentiation throughout ontogeny (ontogenetic niche shifts), as an adapta- tion mechanism to the continuum of light environments in trop- ical forests, faces enormous challenges given the diversity of life forms and functional roles of tropical plants. In this study, we did not fi nd strong evidence supporting ontogenetic niche shifts. Seedlings expressed a range of physiological and mor- phological responses to changing light conditions consistent with their light-demanding character as adults. The amount and direction of phenotypic changes across a gradient of light conditions must be considered when judging the plasticity of morphological and physiological characters in relation to re- generation strategy and ontogenetic stages ( Strauss-Debenedetti and Bazzaz, 1996 ). For instance, the response of Bonamia was very restricted in relation to the response of Stigmaphyllon , which reacted faster to light variation. Both responses were consistent with the strategy of resource use and habitat selec- tion of these species in the canopy. Since lianas are increasing 2096 AMERICAN JOURNAL OF BOTANY OGUCHI , R. , K. HIKOSAKA , AND T. HIROSE . 2005 . Leaf anatomy as a con- straint for photosynthetic acclimation: Differential responses in leaf anatomy to increasing growth irradiance among three deciduous trees. Plant, Cell & Environment 28 : 916 – 927 . PARKER , G. G. , A. P. SMITH , AND K. P. HOGAN . 1992 . Access to the upper forest canopy with a large tower crane. BioScience 42 : 664 – 670 . PAUL , G. S. , AND J. B. YAVITT . 2011 . Tropical vine growth and the effects on forest succession: A review of the ecology and management of tropical climbing plants. Botanical Review 77 : 11 – 30 . PEARCY , R. W. 2007 . Responses of plants to heterogeneous light envi- ronments. In F. I. Pugnaire and F. Valladares [eds.], Functional plant ecology, 234–257. CRC Press, Boca Raton, Florida, USA. POORTER , L. , F. BONGERS , F. J. STERCK , AND H. WÖLL . 2005 . Beyond the regeneration phase: Differentiation of height–light trajectories among tropical tree species. Journal of Ecology 93 : 256 – 267 . PUTZ , F. E. 1984 . The natural history of lianas on Barro Colorado Island, Panama. Ecology 65 : 1713 – 1724 . QUINN , G. P. , AND M. J. KEOUGH . 2002 . Experimental design and data analysis for biologists . Cambridge University Press, Cambridge, UK. RATKOWSKY , D. A. , M. A. EVANS , AND J. R. ALLDREDGE . 1993 . Crossover experiments: Design, analysis, and application . Marcel Dekker, New York, New York, USA. SANCHES , M. C. , AND I. F. M. V ÁLIO . 2002 . Seedling growth of climbing species from a southeast Brazilian tropical forest. Plant Ecology 159 : 51 – 59 . SCHNITZER , S. A. , AND F. BONGERS . 2002 . The ecology of lianas and their role in forests. Trends in Ecology & Evolution 17 : 223 – 230 . SCHNITZER , S. A. , AND F. BONGERS . 2011 . Increasing liana abundance and biomass in tropical forests: Emerging patterns and putative mecha- nisms. Ecology Letters 14 : 397 – 406 . STRAUSS-DEBENEDETTI , S. , AND F. A. BAZZAZ . 1996 . Photosynthetic char- acteristics of tropical trees along successional gradients . In S. S. Mulkey, R. L. Chazdon, and A. P. Smith [eds.], Tropical forest plant ecophysiology, 162–186. Chapman and Hall, New York, New York, USA. TOLEDO-ACEVES , T. , AND M. D. SWAINE . 2008 . Biomass allocation and photosynthetic responses of lianas and pioneer tree seedlings to light. Acta Oecologica 34 : 38 – 49 . TURNBULL , M. H. , D. DOLEY , AND D. J. YATES . 1993 . The dynamics of photosynthetic acclimation to changes in light quantity and quality in three Australina rainforest tree species. Oecologia 94 : 218 – 228 . VAN DER HEIJDEN , G. M. , S. A. SCHNITZER , J. S. POWERS , AND O. L. PHILLIPS . 2013 . Liana impacts on carbon cycling, storage and sequestration in tropical forests. Biotropica 45 : 682 – 692 . WHITMORE , T. C. 1996 . A review of some aspects of tropical rain for- est seedling ecology with suggestions for further enquiry . In M. D. Swaine [ed.], The ecology of tropical forest tree seedlings, 3–39. Man and the biosphere series, UNESCO, New York, New York, USA. WRIGHT , S. J. , O. CALDERÓN , A. HERN ÁNDEZ , AND S. PATON . 2004 . Are lia- nas increasing in importance in tropical forest? A 17-year record from Panama. Ecology 85 : 484 – 489 . WRIGHT , S. J. , H. C. MULLER-LANDAU , R. CONDIT , AND S. P. HUBBELL . 2003 . Gap-dependent recruitment, realized vital rates, and size distributions of tropical trees. Ecology 84 : 3174 – 3185 . GIANOLI , E. 2003 . Phenotypic responses of the twining vine Ipomoea pur- purea (Convolvulaceae) to physical support availability in sun and shade. Plant Ecology 165 : 21 – 26 . GILBERT , B. , S. J. WRIGHT , H. C. MÜLLER-LANDAU , K. KITAJIMA , AND A. HERN ÁNDEZ . 2006 . Life history trade-offs in tropical trees and lianas. Ecology 87 : 1281 – 1288 . HIKOSAKA , K. , AND I. TERASHIMA . 1996 . Nitrogen partitioning among pho- tosynthetic components and its consequence in sun and shade plants. Functional Ecology 10 : 335 – 343 . HOLBROOK , N. M. , AND F. E. PUTZ . 1996 . Physiology of tropical vines and hemiepiphytes: Plants that climb up and plants that climb down . In S. S. Mulkey, R. L. Chazdon, and A. P. Smith [eds.], Tropical forest plant ecophysiology, 363–394. Chapman and Hall, New York, New York, USA. KAMALUDDIN , M. , AND J. GRACE . 1992 . Acclimation in seedlings of a tropi- cal tree Bischofi a javanica following a stepwise reduction in light. Annals of Botany 69 : 557 – 562 . KITAJIMA , K. 1994 . Relative importance of photosynthetic traits and allo- cation patterns as correlates of seedling shade tolerance of 13 tropical trees. Oecologia 98 : 419 – 428 . KITAJIMA , K. 1996a . Cotyledon functional morphology, patterns of seed reserve utilization and regeneration niches of tropical tree seedlings . In M. D. Swaine [ed.], The ecology of tropical forest tree seedlings, 193–210. Man and the biosphere series, UNESCO, New York, New York, USA. KITAJIMA , K. 1996b . Ecophysiology of tropical tree seedlings. In S. S. Mulkey, R. L. Chazdon and A. P. Smith [eds.], Tropical forest plant ecophysiology, 559–596. Chapman and Hall, New York, New York, USA. LAURANCE , W. F. , A. S. ANDRADE , A. MAGRACH , J. L. C. CAMARGO , J. J. VALSKO , M. CAMPBELL , P. M. FEARNSIDE , ET AL. 2014 . Long-term changes in liana abundance and forest dynamics in undisturbed Amazonian forests. Ecology 95 : 1604 – 1611 . LEDO , A. , AND S. A. SCHNITZER . 2014 . Disturbance and clonal reproduc- tion determine liana distribution and maintain liana diversity in a tropical forest. Ecology 95 : 2169 – 2178 . LEICHT-YOUNG , S. A. , A. M. LATIMER , AND J. A. JR. SILANDER . 2011 . Lianas escape self-thinning: Experimental evidence of positive density de- pendence in temperate lianas Celastrus orbiculatus and C. scan- dens. Perspectives in Plant Ecology, Evolution and Systematics 13 : 163 – 172 . LETCHER , S. G. , AND R. L. CHAZDON . 2012 . Life history traits of lianas dur- ing tropical forest succession. Biotropica 44 : 720 – 727 . MALHI , Y. 2012 . The productivity, metabolism and carbon cycle of tropi- cal forest vegetation. Journal of Ecology 100 : 65 – 75 . MALHI , Y. , AND O. L. PHILLIPS . 2004 . Tropical forests and global atmo- spheric change: A synthesis. Philosophical Transactions of the Royal Society, B, Biological Sciences 359 : 549 – 555 . NABE-NIELSEN , J. 2002 . Growth and mortality rates of the liana Machaerium cuspidatum in relation to light and topographic position. Biotropica 34 : 319 – 322 . OGUCHI , R. , K. HIKOSAKA , AND T. HIROSE . 2003 . Does the photosyn- thetic light-acclimation need change in leaf anatomy? Plant, Cell & Environment 26 : 505 – 512 .