| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ecology |
2Forest Ecology and Forest Management Group, Wageningen University, P.O. Box 47, 6700 AA Wageningen, the Netherlands; 3Instituto Boliviano de Investigación Forestal, P.O. Box 6204, Santa Cruz de la Sierra, Bolivia
Received for publication July 26, 2006. Accepted for publication February 2, 2007.
ABSTRACT
Our understanding of leaf acclimation in relation to irradiance of fully grown or juvenile trees is mainly based on research involving tropical wet forest species. We studied sun–shade plasticity of 24 leaf traits of 43 tree species in a Bolivian dry deciduous forest. Sampling was confined to small trees. For each species, leaves were taken from five of the most and five of the least illuminated crowns. Trees were selected based on the percentage of the hemisphere uncovered by other crowns. We examined leaf trait variation and the relation between trait plasticity and light demand, maximum adult stature, and ontogenetic changes in crown exposure of the species. Leaf trait variation was mainly related to differences among species and to a minor extent to differences in light availability. Traits related to the palisade layer, thickness of the outer cell wall, and Narea and Parea had the greatest plasticity, suggesting their importance for leaf function in different light environments. Short-lived pioneers had the highest trait plasticity. Overall plasticity was modest and rarely associated with juvenile light requirements, adult stature, or ontogenetic changes in crown exposure. Dry forest tree species had a lower light-related plasticity than wet forest species, probably because wet forests cast deeper shade. In dry forests light availability may be less limiting, and low water availability may constrain leaf trait plasticity in response to irradiance.
Key Words: Bolivia • crown exposure • leaf traits • light acclimation • plasticity • tropical dry deciduous forest
Trees are long-lived and sessile organisms that subsist in a spatially and temporally highly heterogeneous environment. Trees should therefore possess the capacity to adjust their leaves to their environment. In tropical wet forests light is considered to be the most limiting resource for tree growth and survival (Whitmore, 1996
) and a major axis of differentiation for tropical tree species. In contrast, studies on leaf trait variability from (semi-)arid woody vegetations, such as the Mediterranean maquis (e.g., Gratani and Varone, 2004
) and the Californian chaparral (e.g., Ackerly, 2004
), have generally focused on leaf traits of shrub species in relation to water availability. Recently, several studies have focused on leaf trait acclimation in response to both light and water availability, thus more closely approximating the reality in the field (Sack et al., 2003
; Sánchez-Gómez et al., 2006
; Quero et al., 2006
). Although authors found that shaded conditions enhance seedling tolerance to drought (Quero et al., 2006
) and that the impact of drought on seedling survival and growth rates was stronger in high light then in low light conditions (Sánchez-Gómez et al., 2006
), clear functional types of species able tolerate a combination of shade and drought are not yet clearly defined (Sack et al., 2003
).
Our current knowledge of leaf responses to irradiance in the tropics is mainly based on research conducted in wet forests (e.g., Bongers and Popma, 1988
; Poorter et al., 2000
) or on small seedlings. Sun leaves grow in a high resource environment and are often relatively thick and small with a low surface to volume ratio. Thicker leaves have a reduced light absorption per unit biomass (Agusti et al., 1994
) and an increased photosynthesis per unit leaf area (Björkman, 1981
; Klich, 2000
). Photosynthetic capacity is enhanced through investment in thicker and/or extra layers of palisade parenchyma tissue and increased nitrogen concentration per unit leaf area (Björkman, 1981
; Poorter et al., 1995; Poorter, 1999
). Overheating might be a severe problem in high light when leaf temperatures exceed the photosynthetic optimum, especially if water availability is low (Smith, 1978
). Smaller, and/or slender leaves have a reduced boundary layer resistance (Givnish, 1984
) and are thus capable of regulating their temperature through better convective cooling of the leaf area (Parkhurst and Loucks, 1972
). Leaf cuticles and epidermides may reduce water loss through evaporation (Gamage et al., 2003
; Mendes et al., 2001) and protect the photosynthetic tissue from excessive irradiance through increased reflectance (Roth, 1984
). High radiation loads and high vapor pressure deficits result in greater transpiration rates of sun leaves, and thus a large water flow to the leaves is needed. This can be facilitated by large xylem conduits (Zimmermann, 1983
), relatively thick internodes in relation to the leaf area (cf. Westoby and Wright, 2003), and a high leaf hydraulic conductance (Sack et al., 2005
).
Light is a limiting resource for growth in the shaded understory. Trees growing in the shaded understory enhance their light interception through the formation of relatively large, thin leaves with a low leaf mass per unit leaf area (LMA) (Evans and Poorter, 2001
). They may forage for light and minimize self shading through the formation of cheap petioles with a high petiole length per unit petiole mass. Respiratory carbon losses are reduced through low protein and nitrogen concentrations (Sims and Pearcy, 1989
).
The magnitude of acclimation that a species can realize in response to differences in irradiance can be referred to as plasticity. Plasticity enhances plant performance and is thought to differ predictably among functional groups. Pioneer species that regenerate in open areas and forest gaps were hypothesized to have a higher plasticity than shade tolerant species, because they grow in a more variable environment. (Bazzaz, 1979
; Bazzaz and Wayne, 1994
). It is a certainty that formed gaps will close, allowing for selection on high acclimation potential. The high resource availability in early successional habitats allows pioneer species to support the carbon investment costs that come along with a higher acclimation potential. Still there is little consensus about this hypothesis. Some studies found plasticity to be similar for pioneer and shade tolerant species (Sims and Pearcy, 1989
; Kitajima, 1994
), while others actually found less plasticity in pioneers than in shade tolerant species (Popma et al., 1992
). Grubb (1998)
suggested that these apparently contradictory results might have been found because pioneers in the seedling stage are more plastic, whereas in the adult stage shade tolerant species are more plastic. Contradictory results may also partly be explained by the fact that in many studies only a few species and a few leaf traits have been evaluated and differing methods were used to calculate and define plasticity. In addition, researchers have often assigned species arbitrarily to functional groups, without an objective, quantitative measure of the light demand of the species.
Popma et al. (1992)
argued that pioneers do not need to have a high plasticity because they always grow in high light and do not survive deep shade. Instead, plasticity should be highest for tall species that establish in the shaded understory and are subject to large ontogenetic changes light availability during their life cycle (cf. Thomas and Bazzaz, 1999; Poorter et al., 2005
). Close relations of leaf trait plasticity with maximum adult stature and ontogenetic crown exposure may thus be expected.
The magnitude of plasticity is likely smaller in tropical dry forest than in tropical wet forest tree species. Tropical dry forests are characterized by a prolonged dry season in which the vegetation is subject to low soil water availability and high vapor pressure deficit of the air (Bullock et al., 1995
). Light penetration in dry forests is relatively high compared to wet forests (5–10% in the wet season in dry forests vs. 0.4–2.0% yr-round in wet forests; Coomes and Grubb, 2000), due to a low and open canopy and low stem densities. These levels are even higher during the dry season, when many canopy and subcanopy species shed their leaves (Parker et al., 2005
). Leaf trait acclimation in response to light availability is therefore likely to be less pronounced in tropical dry forests than in wet forests.
In this study we evaluate leaf characteristics and evaluate sun–shade plasticity of 43 tropical dry forest tree species. Twenty-four morphological, anatomical and chemical leaf traits that are important for the heat, water, and carbon balance of the plant are assessed. Leaf trait plasticity is related to quantitative measures of maximum adult stature, juvenile crown exposure, and ontogenetic changes in crown exposure of the species. We made the following three predictions: (1) Functional groups of species related to shade tolerance differ in leaf trait plasticity in response to irradiance. Pioneer species grow in a more variable habitat with higher resource availability and are thus expected to have high leaf trait plasticity, whereas shade tolerant species tend to spend their entire life cycle in the low resource environment of the forest understory, with a marginally positive carbon balance, and therefore should have less potential to acclimate and less plasticity. (2) There is a positive relationship between leaf trait plasticity and maximum adult stature, juvenile crown exposure, and ontogenetic changes in crown exposure of the species. Tall species establish in the shaded forest understory, but, during their life cycle, endure more changes in light availability than small species. This requires a greater ability to adjust to their changing growth environment. (3) Dry tropical forests tree species will have smaller leaf trait plasticity in response to irradiance then tree species form wetter forest types. As light is less of a limiting factor in dry forests, the necessity of a high light-related plasticity is smaller for dry forest tree species.
MATERIALS AND METHODS
Study area
This study was conducted in the INPA forest (16°07' S, 61°43' W) in the lowlands of eastern Bolivia. The forest can be classified as a tropical lowland dry deciduous forest, situated at the transition zone between the Amazonian wet forests in the north and the thorn-shrub formations of the Gran Chaco in the south (Killeen et al., 1998
; Jardim et al., 2003
).
The study area has a mean altitude of 458 m and is located on the Precambrian Brazilian shield. Soils are oxisols and are low in nutrients. Mean annual temperature at Concepcíon, ca. 40 km from the study site, is 24.3°C, and the mean annual precipitation is 1160 mm with a dry season (<100 mm rainfall per mo) from April until October. From June through September, the potential evapotranspiration exceeds the mean monthly rainfall, which can result in a water deficit. The forest canopy has an average height of 22 m with emergent trees growing up to 30 m. The forest has a density of 437 stems·ha–1, a basal area of 19.7 m2·ha–1, and a species richness of 34 ha–1 (trees 
10 cm dbh; Instituto Boliviano de Investigación Forestal [IBIF], Santa Cruz de la Sierra, Bolivia, unpublished data). The forests in the region have previously been classified as semi-deciduous forests (Killeen et al., 1998
), because most of the subcanopy trees, shrubs and lianas are evergreen or semi-evergreen (Killeen et al., 1998
). Because the canopy of the forest at INPA is fully deciduous in the dry season, we chose to classify this forest as a dry deciduous forest in stead of a semi-deciduous forest.
The most dominant species at INPA (listed in decreasing order of basal area [m2·ha–1] are Acosmium cardenasii H.S. Irwin & Arroyo (Fabaceae), Casearia gossypiosperma Briquet (Flacourtiaceae), Caesalpinia pluviosa DC (Fabaceae), Neea cf. steinbachii (Nyctaginaceae), Machaerium acutifolium Vogel (Fabaceae), Anadenanthera macrocarpa Benth (Fabaceae), Piptadenia viridifolia (Kunth.) Benth (Fabaceae), and Centrolobium microchaete (Benth.) H.C. Lima (Fabaceae) (IBIF, unpublished data). Nomenclature follows that of the nomenclature database of the Missouri Botanical Garden (W3TROPICOS; http://www.mobot.org).
Study species
We studied leaf traits of 43 tree species belonging to 40 genera, 24 families, and 19 orders. Among the selected species are some of the most abundant species in this type of forest, as well as several commercially valuable timber species (Table 1). With six species, Fabaceae is the biggest family in this study, which is in line with its dominance in this dry deciduous forest. All species together represent more than 77% of the stems larger than 10 cm dbh in the permanent sample plots (IBIF, unpublished data).
|
) using species-specific height–diameter relationships (Verweij, 2004
) and the diameter of the third thickest tree in the permanent sample plots (thus avoiding outliers). We classified the species into four guilds of shade tolerance (cf. Finegan, 1992
) based on their light requirements and longevity (Jardim et al., 2003
; Mostacedo et al., 2003
; Justiniano et al., 2004
) and additional field observations on the habitat preference of the species (L. Poorter, personal observation). Short-lived pioneers (SLP) are species that need high light to establish and grow to their adult stature and have a lifespan up to 30 yr, These SLP are the "typical" pioneers sensu stricto, that can form dense stands in disturbed areas or large treefall gaps (cf. Kammesheidt, 2000
). Long-lived pioneers (LLP) need intermediate light to establish and grow to the high light environment of the forest canopy. They live longer than 30 yr. LLP are pioneers sensu lato that can establish under a wider range of light conditions. Partial shade tolerant species (PST) can establish in the shaded understory, but need more light in later stages of their lifecycle to reach their maximum stature in the high light environment of the canopy. Shade tolerant species (ST) are species that can complete their entire life-cycle in the shade. We included 20 ST species, 10 PST species, nine LLP species, and four SLP species.
Data from a separate study (L. Poorter, unpublished data) provided an objective and quantitative measure of light demand, based on the analysis of height–light trajectories of each species. For a median of 133 (range 16–9064) individuals per species ranging from seedlings to adult trees, the height and crown exposure (CE) were estimated. The CE varies from 1 if the tree crown does not receive any direct light, to 2 if it receives lateral light, 3 if it receives overhead light on part of the crown, 4 when it receives overhead light on the whole crown, to 5 if it has an emergent crown that receives light from all directions (Dawkins and Field, 1978
). CE can be measured repeatedly, and there is a close relation between CE and both canopy openness (Davies et al., 1998
) and incident radiation (Clark et al., 1993
). For each species, we related CE to tree height, using a multinomial regression analysis (L. Poorter et al., unpublished manuscript; cf. Poorter et al., 2005
) and calculated the average crown exposure at 2 m height (CEjuv) and maximum adult stature (CEadult). The difference between CEjuv and CEadult was used to estimate the ontogenetic change in CE for each species (CEchange) (Table 1). It should be emphasized that these values indicate the average CE values at the population level. At a given height, individuals of the same species may be found under a wide range of crown exposures, and we used these extreme individuals as sun and shade trees for the sampling of leaves used in the present study. We used the CE and guilds as two complementary approaches to evaluate hypotheses about plasticity. The CE is a quantitative measure of light demand, whereas the guilds capture both juvenile crown exposure and ontogenetic changes in crown exposure (PST and LLP have larger CEchange than the two other guilds).
Leaf collection
Sun and shade leaves of each species were collected in the wet season of 2003–2004, by sampling five trees per species in full sunlight and five trees growing in shaded conditions. Sun and shade trees were selected based on the canopy openness (CO) above their crowns. CO was estimated for every sample tree as the percentage of the hemisphere above the tree crown that is not covered by crowns of other trees. Although the maximum and minimum CO at which individuals could still be found is rather species-specific, we applied the rule of thumb that the CO of sun individuals was at least 50% and that of shade individuals at most 15%. For species that could not be found within these ranges, the most highly and least illuminated individuals found were sampled. Average CO of the sampled sun individuals was 56.5 ± 0.3 % (mean ± SE) and of the shade individuals 9.8 ± 0.1 (mean ± SE) (Table 1). Leaves were sampled from trees between 10–20 cm dbh and 10–20 m height, to reduce the confounding effect of tree size on leaf traits (Koch et al., 2004
). Jacaratia sp., Erythroxylum daphnites, Esenbeckia almawillia, Actinostemon concepcionis, Myrciaria cauliflora, Manihot guaranitica, Phyllanthus sp. nov., and Pogonopus tubulosus, reached only a maximal height of 2–9 m (Table 1). For these species, the leaves were collected from the tallest individuals. For every sample tree, we estimated the diameter at breast height (dbh), total height, and the percentage canopy openness. To minimize trait variation related to the position of sampled leaves within the crown of the individual trees, we selected trees of the same height (especially within species) and sampled five leaves per tree from the outer leaf shell midway between the top and bottom of the crown. All sun leaves were collected from the most exposed side and all shade leaves from the least exposed side of the crown. Leaves were cut with a pair of extendable pruning shears and transported to the field station in plastic bags. To minimize leaf trait variation caused by the age of the sampled leaves, we selected leaves that were young and fully expanded. We also selected leaves with minimal signs of herbivore or pathogen damage.
In the field station a cross-section of one leaf per tree was stored in 70% ethanol (EtOH) for anatomical analysis. For compound and lobed leaves, a cross-section was made of an average-sized foliole or lobe.
Leaf morphology
For each tree, four leaves were measured for the length (LL) and width (LW) of the leaf blade. Folioles were considered to be functional equivalents of simple leaves and were treated accordingly in this study. Average-sized folioles were used in the case of compound leaves. The length of the petiole and the length and diameter of the internode section, below the corresponding node, were also measured. Leaf thickness (LT; µm) was measured with a micrometer. Leaves were rehydrated overnight in wet tissue in a refrigerator, dried with a tissue, and weighed to determine the saturated fresh mass. Leaves and folioles were scanned with a desktop-scanner, and their surface area (LA; cm2) was determined using pixel-counting software (Van Berloo, 1998
). Leaf toughness (LTO) was determined with a penetrometer (punch-head; ø 3 mm; 7 mm2). With this device, the leaves were punctured between the veins with the head of a nail. The nail was connected to a reservoir, which was gradually filled with water to increase the mass on the nail until the lamina finally ruptured. The mass at the moment of penetration was converted to a measure for leaf toughness (N·cm–2), a proxy for leaf toughness. Leaf toughness was not determined for Caesalpinia pluviosa because its folioles were too small. Thereafter, the leaves, petioles and internodes were oven dried for 48 h at 65°C and measured again for their dry mass.
From these measurements, we calculated the following morphological traits: leaf density [LD; leaf dry mass/(leaf area x leaf thickness); g·cm–3], leaf slenderness (LS; leaf length/leaf width; cm·cm–1), leaf mass per area (LMA; leaf dry mass/leaf area; g·m–2), leaf dry matter content (LDMC; leaf dry mass/fully saturated leaf fresh mass; g·g–1), specific petiole length (SPL; petiole length/dry petiole mass; cm·g–1), and the internode to leaf area ratio (ILAR; internode cross-sectional area/leaf area; cm2·cm–2) (Table 2). LTO is an indicator of resistance to herbivory, the LS of the capacity to reduce overheating by a reduction of the boundary layer. LDMC indicates the amount of dry mass invested per unit leaf fresh mass and LMA, the amount of biomass a plant invests to produce a unit leaf surface for light capture. Both are proxies for leaf construction costs. SPL indicates the light foraging capacity given a fixed amount of biomass for petiole construction, and the ILAR is an estimate of the water supply capacity to the leaf.
|
From the digital images, we measured the thickness of the outer cell wall (TOC; µm), upper epidermis (UET; µm), palisade parenchyma (PPT; µm), spongy parenchyma (SPT; µm), and lower epidermis (LET; µm) and counted the number of palisade parenchyma cell layers (PPL). We measured the diameter of the five largest xylem conduits (XCD; µm) in the midrib of the leaf as the average of two perpendicular diameters and determined the xylem conduit density (XCDE; mm–2) by counting the number of conduits in a given cross-sectional area of the xylem tissue. Additional observations were made whether additional tissue layers, such as a hypodermis, were present. From these data, we calculated the palisade to spongy parenchyma ratio (PSPR; palisade parenchyma thickness/spongy parenchyma thickness; µm·µm–1) (Table 2).
Finally, we determined for each species the mass-based nitrogen concentration (Nmass; mg·g–1), with the Kjeldahl microassay protocol (Archibald, 1958
), and the phosphorus concentration (Pmass; mg·g–1) for pooled leaf samples per light environment. From these data, we calculated the area-based nitrogen (Narea; mmol·m–2) and phosphorus concentration (Parea; mmol·m–2).
Statistics
For each leaf trait, an arithmetic average was calculated per tree from four sampled leaves. PPL was untransformed, LDMC was arcsine-transformed and the other leaf characters were log10-transformed prior to analysis to improve normality and homoscedasticity.
Leaf traits of sun and shade leaves were compared using a full factorial two-way ANOVA with species and light level as fixed factors. The amount of variation explained (
2 x 100%) by the species, light, and the interaction effect was calculated as the sum of squares of the effect divided by the total sum of squares of the model. 2 is an equivalent of R2. If species have a smaller among-species variance of a leaf trait in one light environment compared to another, then this would be an indication that species show a convergent evolution of the trait in that light environment (cf. Bongers and Popma, 1988
). We therefore analyzed the differences in variance between sun and shade leaves with a two-tailed F test, using the mean trait values per species in each light environment. Leaf trait plasticity was calculated for each species based on the mean leaf trait values in the sun and in the shade (cf. Valladares et al., 2000b
). This plasticity index was calculated as the absolute difference between the maximum trait value in one of the light environments and the minimum trait value in the other light environment, divided by the maximum value, and multiplied by 100% (cf. Valladares et al., 2000b
). Total plasticity per species was expressed as the average plasticity of all 24 leaf traits. Using Pearson correlation analyses, we investigated the relations between sun and shade values of the leaf traits and the relations between leaf trait plasticity and CEjuv, CEchange, and Hmax using the species average. Statistical analyses were performed with SPSS 12.0.1 (SPSS Inc. Chicago, Illinois, USA).
RESULTS
Species and light environment
Species and light environment had a large effect of leaf morphology, anatomy, and chemistry (Table 3). All leaf traits differed strongly among species (P < 0.001), which explained most of the leaf trait variation (mean 77%, range 34–95). Light had a significant effect on 16 of 24 leaf traits and explained considerably less of the variation in leaf traits (mean 3.4%, range 0.2–10). There was a significant species and light interaction effect for 14 of 24 leaf traits. For these traits, species responses to light explained an additional 3.9% of the variation (range 1.1–7.6). Light did not have a significant direct or interaction effect on leaf slenderness and xylem conduit diameter. Light did not have a significant effect on the chemical traits Nmass and Pmass either, and no interaction effect could be calculated here because N and P were determined for each species based on pooled leaves. An absence of a significant species-light interaction effect indicates that all species have a similar response to light. Species thus had a similar response to the light for leaf slenderness, LDMC, internode to leaf area ratio, upper and lower epidermis thickness, and xylem conduit density.
|
Sun leaves were shorter, narrower, smaller, and thicker than shade leaves. Sun leaves had a higher LMA, LDMC, and a higher N and P concentration per unit leaf area (Narea, Parea), but sun and shade leaves did not differ in slenderness, leaf toughness and density, or specific petiole length.
Anatomically, the greater thickness of sun leaves could be attributed to their thicker outer cell wall, upper epidermis, palisade parenchyma, and lower epidermis. Sun leaves had a higher palisade to spongy parenchyma ratio. Part of this increase can be ascribed to the increased number of palisade parenchyma cell layers in sun leaves compared to shade leaves. Sun and shade leaves did not differ in the mean thickness of the spongy parenchyma nor the diameter and density of xylem conduits of the midrib.
For all leaf traits sun and shade values were highly correlated (average r = 0.86, range 0.52–0.97, p < 0.001 in all cases) (Table 3). Few traits showed differences in species variance between sun and shade leaves. The species variance in leaf area and internode to leaf area ratio was lower in sun leaves than in shade leaves, suggesting convergent evolution in these leaf traits in the sun. The variance in thickness of the outer cell wall and palisade to spongy parenchyma ratio was lower in shade leaves than in sun leaves.
Leaf trait plasticity in response to irradiance
For the morphological leaf traits, internode to leaf area ratio (15.7%), LMA (15.0%), leaf area (14.0%), and leaf thickness (10.5%) were among those with the greatest plasticity. LDMC (6.7%), leaf length (6.9%), and width (7.4%) were among the traits with the lowest morphological plasticity, while still being significantly different between sun and shade leaves. Pmass (0.3%) had the lowest plasticity (Table 3). For the anatomical traits, three of the four that had the highest plasticity were related to the palisade parenchyma tissue, namely palisade parenchyma thickness (24.6%), palisade to spongy parenchyma ratio (22.2%), and the number of palisade parenchyma cell layers (18.8%). Also the thickness of the outer cell wall (20.2%) had a high plasticity. The thickness of the lower and upper epidermis (4.6% and 7.2%) had the lowest plasticity, while still significantly different between sun and shade leaves. Xylem conduit diameter (2.9%) had the lowest plasticity among all anatomical traits (Table 3). Morphological, anatomical, and chemical traits did not differ significantly in their mean plasticity (ANOVA: F3, 20 = 0.7; P > 0.05).
Plasticity among functional groups
Plasticity in leaf length, width, and area differed significantly among functional groups related to shade tolerance, while all other plasticity in leaf traits did not (Table 4). Short-lived pioneers had greater plasticity in these traits than the three other functional groups.
|
|
) and a tropical wet forest (Los Tuxtlas, Mexico, Bongers and Popma, 1988
). The three forests are distinctly different in mean annual precipitation and length of the dry period (Table 5). The plasticity in all but one of the leaf traits differed among the three forest types (one-way ANOVA, all P < 0.001, Table 5). Leaf trait plasticity was greatest in the wet forest, but hardly any differences in leaf plasticity were found between the dry and the moist forest (Fig. 2), despite considerable differences in forest structure and canopy openness.
|
|
Leaf trait plasticity in response to irradiance
Most of the variation in leaf traits could be explained by differences among species. Light had only a minor effect: for one-third of the traits no effect at all and for two-thirds the effect on leaf trait variation was low, despite large differences in canopy openness (CO) above the sampled sun (CO = 56.5 ± 0.3%) and shade trees (CO = 9.8 ± 0.1%). Species respond differently to an increase in light, as indicated by the high number of significant species–light interaction terms (Table 3). Different tree species thus have different ways of coping with resource capture and conservation.
Although the light effect explained minor proportions of the leaf trait variation, sun and shade leaves did differ in most traits. Results are overall in line with past studies reviewing the matter of leaf acclimation to irradiance (see, e.g., Bongers and Popma, 1988
; Sims and Pearcy, 1989
; Cao, 2000
; Evans and Poorter, 2001
; Rozendaal et al., 2006
). While species differed substantially in their response to irradiance, there is a high correlation between species trait values in the sun and the shade (Table 3). This means that the species ranking for trait values is largely maintained in both light environments and that there is no crossover in trait performance between low and high irradiance. Shade tolerant and pioneer species may therefore specialize for different environments because of differences in their inherent traits (cf. Kitajima and Poorter, in press
), rather than through phenotypic differences in trait values at low and high irradiance (cf. Sack and Grubb, 2003
).
The among-species variance of four of the 24 traits differed significantly between sun and shade leaves. Bongers and Popma (1989) argued that, if the among-species variance of a trait in a given light environment is smaller, the state of this trait has a greater importance for the performance of the leaf in that environment. A smaller among-species variance therefore indicates convergent evolution in that light environment. We found a smaller variance in leaf area in the sun. Sun leaves are smaller to allow more effective cooling of the leaf area as they have a thinner boundary layer (Parkhurst and Louks, 1972; Givnish, 1984
), thus avoiding supraoptimal leaf temperatures for photosynthesis. We also found a smaller variance in the internode to leaf area ratio. Sun leaves have a higher internode to leaf area ratio to compensate for the higher transpiration with an increased water supply (cf. Westoby and Wright, 2003). In this way, plants increase the vessel number, rather than the vessel density or diameter. The variances in thickness of the outer cell wall and in the palisade to spongy parenchyma ratio were smaller in the shade (Table 3). Shade leaves have a thinner upper epidermis with a thinner outer cell wall, apparently because they need less protection than sun leaves. The outer cell wall and upper epidermis protect the leaf as they minimize the damaging effect of high irradiance through reflection of the excessive light (Roth, 1984
; Bondada et al., 1996
) and reduce cuticular evaporation (Hall and Jones, 1961). Shade leaves have a smaller palisade to spongy parenchyma ratio. The relatively thick spongy parenchyma layer is especially useful to enhance backscattering within the leaf of diffuse understory light. Bongers and Popma (1988)
also found smaller variances in palisade to spongy parenchyma ratios in shade leaves in Los Tuxlas. Sun plants may regulate leaf temperatures through smaller leaves or an increased transpiration. Such a water-spending strategy is counterintuitive, however, given the limited water availability in the dry season. Sun trees may avoid dry-season water stress by exploring deeper soil layers, a larger soil volume for water (Poorter and Hayashida, 2002), or by having a deciduous leaf habit. Of the species included in this study, we observed that at least 22 species showed a deciduous leaf habit in the dry season. Despite the fact that light acclimation and dynamic responses to light have been reported vary substantially with leaf longevity (Kursar and Coley, 1993
, 1999
), leaf trait plasticity did not differ between evergreen (N = 14) and deciduous species (N = 22) in our study (t test: –1.45 
t 1.32; P > 0.05; df = 34).
Leaf trait plasticity varies between 0.3 and 24.6%. Leaf traits that have the greatest plastic response could be more important for leaf functioning in different light environments (Bongers and Popma, 1988
). If this were the case in our forest, palisade parenchyma thickness, palisade to spongy parenchyma ratio, thickness of the outer cell wall, the number of palisade parenchyma cell layers, Narea and Parea are most critical for light acclimation in our species. Three of these traits are anatomical and directly related to the palisade parenchyma tissue. The palisade parenchyma, Narea and Parea all play a distinct role in enhancing the photosynthetic capacity of the leaf (Evans, 1999
). The internode to leaf area ratio, the most plastic morphological trait in this study, is related to water supply and the thickness of the outer cell wall is related to water conservation. A high light-related plasticity was also found in Narea and palisade parenchyma thickness of 61 Mexican wet forest species (Bongers and Popma, 1988
). Rozendaal et al. (2006)
analyzed the plasticity of 39 Bolivian moist forest species and found a high plasticity in internode to leaf area ratio, SLA, Narea, and Parea.
Plasticity among functional groups
Plasticity of four leaf traits was greatest for short-lived pioneers, in line with our first hypothesis. Yet, we did not expect that functional groups would differ in the plasticity of so few leaf traits (cf. Sack et al., 2003
). Many other studies showed considerable differences among functional groups (Popma et al., 1992
; Kitajima, 1994
; Strauss-Debenedetti and Bazzaz, 1996; Valladares et al., 2002). We may argue that, for relatively open forests such as tropical dry forests, plant traits and plasticity might be more closely related to drought tolerance than to shade tolerance.
In our study only the plasticity of leaf size parameters (LL, LW, and LA) and upper epidermis thickness differed among functional groups. For these traits, short-lived pioneer species had the greatest plasticity. This is in line with the hypothesis as formulated by Strauss-Debenedetti and Bazzaz (1996), which assumes that pioneer species have greater plasticity because they grow in more heterogeneous habitats, but in contrast with results from other studies (Sims and Pearcy, 1989
; Popma et al., 1992
; Kitajima, 1994
; Rozendaal et al., 2006
).
Plasticity in relation to adult size and crown exposure
Most leaf trait plasticity parameters were not correlated to maximum adult stature or ontogenetic crown exposure of the species. The lack of a relation with Hmax is surprising because tall trees generally have to cope with all the changes in light environment that occur from the understorey to the forest canopy (cf. Poorter et al., 2005
). Therefore we expected Hmax and CEchange to be positively related to plasticity. That this is not the case may be because ontogenetic changes in crown exposure are related to ontogenetic plasticity, rather than the sun–shade plasticity derived from sun and shade trees of similar age. It might also be related to the fact that differences in light availability, especially during the dry period, are less pronounced in the deciduous dry forest, which therefore may not act as the only selective force. Low water availability in the dry season might constrain sun–shade plasticity and partly explain our counterintuitive results.
Leaf trait plasticity in dry and wet forests
Although sun and shade leaves differed in most leaf traits, the percentages of explained variation were generally low (Table 3). Only 3% of the total variation in leaf traits could be explained by differences between sun and shade leaves, while the same factor explained 8% of the leaf trait variation of 39 Bolivian moist forest tree species (Rozendaal et al., 2006
). The smaller light effect in our study may well be the result of the more open and deciduous character of the dry forest compared to the moist forest.
When we compared the plasticity in leaf characteristics between three forests differentiated by water availability (Table 5), it was clear that the wet forest had greatest plasticity, while the differences between the other two forests were minor (Fig. 2). This is most probably related to variability in light availability, which is much higher in Los Tuxtlas. The vegetation there is much denser with a much deeper shade year-round in the non-gap areas, while the vegetation in the other two forests is rather open. For such forests, leaf trait acclimation to differences in light availability may thus be not welldefined, because light is not as limiting as in hyperwet forests. An alternative explanation is that in the wet forest the shade leaves have been sampled in deeper shade. Given that the largest changes in leaf traits occur at the lowest light levels (Poorter, 1999
), this might have led to a different observed plasticity among forest types. We acknowledge that water and light availability are often interacting factors, especially in drier ecosystems. Lower water availability has been reported to reduce plant response to irradiance (Quero et al., 2006
; Sánchez-Gómez et al., 2006
). We do not believe that in this deciduous dry forest, water is a limiting factor during the wet season when these leaves are formed. Still tree species may follow a "conservative resource-use strategy" involving relatively low leaf-level responses to irradiance (cf. Valladares et al., 2000a
).
Summarizing, we found that in the deciduous dry forest leaf trait variation is mainly explained by differences among species and only to a minor extent by differences in light availability. Sun–shade plasticity is therefore not large, and as a result, this plasticity is only very sporadically related to (ontogenetic changes in) the light requirements of the species. In relatively open dry forests, light-related plasticity seems to be less essential for species growth and survival than in wet forests, and the low water availability in dry forests may constrain the magnitude of leaf trait plasticity in response to irradiance.
FOOTNOTES
1 The authors thank all the people involved in the Instituto Boliviano de Investigación Forestal (IBIF) for their support during and after the fieldwork; J. Alvarez-Mendoza and R. Verweij for their unprecedented assistance in collecting the data in the field; N. de Ruyter and H. Kieft for supervising the laboratory work with respect to the leaf anatomy; D. Rozendaal for providing her leaf trait data from La Chonta for comparative analysis; and INPA Parket Ltda. for letting us conduct our studies in the INPA forest. Comments of P. J. Grubb and three anonymous reviewers substantially improved the quality of the manuscript. L.M. was supported by a fellowship from the Wageningen graduate school, Production Ecology and Resource Conservation (PE&RC), and L.P. by Veni grant 863.02.007 from the Netherlands Organization for Scientific Research (NWO). 
4 Author for correspondence (e-mail: Lars.Markestijn{at}wur.nl
; phone: +31 317 478071; fax: +31 317 478078
LITERATURE CITED
Ackerly D.. 2004. Functional strategies of chaparral shrubs in relation to seasonal water deficit and disturbance. Ecological Monographs 74: 25-44..[CrossRef][ISI]
Agusti S. Enriquez S. Frost-Christensen H. Sand-Jensen K. Duarte C. M.. 1994. Light harvesting among photosynthetic organisms. Functional Ecology 8: 273-279..[CrossRef]
Archibald R. M.. 1958. Nitrogen by the Kjeldahl method. In D. Seligson, [ed.], Standard methods of clinical chemistry, 2–9. Academic Press, New York, New York, USA..
Bazzaz F. A.. 1979. The physiological ecology of plant succession. Annual Review of Ecology and Systematics 10: 351-371..[CrossRef][ISI]
Bazzaz F. A. Wayne P. M.. 1994. Coping with environmental heterogeneity: the physiological ecology of tree seedling regeneration across the gap–understory continuum. In M. M. Caldwell, R. W. Pearcy, [eds.], Exploitation of environmental heterogeneity by plants, 349-390. Academic Press, San Diego, California, USA..
Björkman O.. 1981. Responses to different quantum flux densities. In O. L. Lange, P. S. Nobel, C. B. Osmond, H. Ziegler, [eds.], Encyclopedia of plant physiology, 57-107. Springer-Verlag, Berlin, Germany..
Bondada B. R. Oosterhuis D. M. Murphy J. B. Kyung S. K.. 1996. Effect of water stress on the epicuticular wax composition and ultrastructure of cotton (Gossypium hirsutum L.) leaf, bract, and boll. Environmental and Experimental Botany 36: 61-69..[CrossRef][ISI]
Bongers F. Popma J.. 1988. Is exposure-related variation in leaf characteristics of tropical wet forest species adaptive?. In M. J. A. Werger, P. van der Aart, H. During, J. T. A. Verhoeven, [eds.], Plant form and vegetation structure: adaptation, plasticity, and relation to herbivory, 191-200. SPB Academic Publishing, The Hague, Netherlands..
Buisson D. Lee D. W.. 1993. The developmental responses of papaya leaves to simulated canopy shade. American Journal of Botany 80: 947-952..[CrossRef][ISI]
Bullock S. H. Mooney H. A. Medina E., [eds.],. 1995. Seasonally dry tropical forests. Cambridge University Press, Cambridge, UK..
Cao K.. 2000. Leaf anatomy and chlorophyll content of 12 woody species in contrasting light conditions in a Bornean heath forest. Canadian Journal of Botany 78: 1245-1253..
Chapin F.. 1980. The mineral nutrition of wild plants. Annual Review of Ecology and Systematics 11: 233-260..[CrossRef][ISI]
Chazdon R. L. Fetcher N.. 1984. Light environments of tropical forests. In E. Medina, H. A. Mooney, C. Vasquez-Yanes, [eds.], Physiological ecology of plants in the wet tropics, 27-36. W. Junk, The Hague, Netherlands..
Chazdon R. L. Kaufmann S.. 1993. Plasticity in leaf anatomy of two rainforest scrubs in relation to photosynthetic light acclimation. Functional Ecology 7: 385-394..[CrossRef]
Chiariello N.. 1984. Leaf energy balance in the wet lowland tropics. In E. Medina, H. A. Mooney, C. Vasquez-Yanes, [eds.], Physiological ecology of plants in the wet tropics, 85-98. W. Junk, The Hague, Netherlands..
Clark D. B. Clark D. A. Rich P. M.. 1993. Comparative analysis of microhabitat utilization by saplings of nine tree species in neotropical rain forest. Biotropica 25: 397-407..[CrossRef][ISI]
Davies S. J. Palmiotto P. A. Ashton P. S. Lee H. S. LaFrankie J. V.. 1998. Comparative ecology of 11 sympatric species of Macaranga in Borneo: tree distribution in relation to horizontal and vertical resource heterogeneity. Journal of Ecology 86: 662-673..[CrossRef][ISI]
Dawkins H. C. Field D. R. B.. 1978. A long-term surveillance system for British woodland vegetation. Department of Forestry, Oxford University, Oxford, UK..
Evans J. R.. 1999. Leaf anatomy enables more equal access to light and CO2 between chloroplasts. New Phytologist 143: 93-104..[CrossRef][ISI]
Evans J. R. Poorter H.. 2001. Photosynthetic acclimation of plants to growth irradiance: the relative importance of specific leaf area and nitrogen partitioning in maximizing carbon gain. Plant, Cell and Environment 24: 755-767..[CrossRef]
Field T. S. Brodribb T. Jaffré T. Holbrook N. M.. 2001. Acclimation of leaf anatomy, photosynthetic light use, and xylem hydraulics to light in Amborella trichopoda (Amborellacaea). International Journal of Plant Sciences 162: 999-1008..[CrossRef][ISI]
Finegan B.. 1992. Bases ecológicas para la silvicultura. Tema 1. V Curso internacional sobre silvicultura y manejo de bosques tropicales. Centro Agronómico Tropical de Investigación y Enseñanza, Turrialba, Costa Rica..
Gamage H. K. Ashton M. S. Singhakumara B. M. P.. 2003. Leaf structure of Syzygium spp. (Myrtaceae) in relation to site affinity within a tropical rain forest. Botanical Journal of the Linnean Society 141: 365-377..[CrossRef][ISI]
Garnier E. Shipley B. Roumet C. Laurent G.. 2001. A standardized protocol for the determination of specific leaf area and leaf dry matter content. Functional Ecology 15: 288-295..
Givnish T. J.. 1984. Leaf and canopy adaptations in tropical forests. In E. Medina, H. A. Mooney, C. Vasquez-Yanes, [eds.], Physiological ecology of plants in the wet tropics, 51-84. W. Junk, The Hague, Netherlands..
Givnish T. J.. 1988. Adaptation to sun and shade: a whole-plant perspective. Australian Journal of Plant Physiology 15: 63-92..[ISI]
Gratani L. Varone L.. 2004. Leaf key traits of Erica arborea L., Erica multiflora L. and Rosmarinus officinalis L. co-occurring in the Mediterranean maquis. Flora 199: 58-59..
Grubb P. J.. 1998. A reassessment of the strategies of plants which cope with shortage of resources. Perspectives in Plant Ecology, Evolution, and Systematics 1: 3-31..[CrossRef]
Jardim A. Killeen T. J. Fuentes A.. 2003. Guía de los árboles y arbustos del bosque seco Chiquitano, Bolivia. Editorial FAN, Santa Cruz de la Sierra, Bolivia..
Justiniano M. J. Peña-Claros M. Gutíerrez M. Toledo M. Jordán C. Vargas I. Montero J. C.. 2004. Guía dendrológica de especies forestales de Bolivia—volumen II. Bolivia Forestal (BOLFOR) Santa Cruz de la Sierra, Bolivia..
Kammesheidt L.. 2000. Some autecological characteristics of early to late successional tree species in Venezuela. Acta Oecologica 21: 37-48..[CrossRef]
Killeen T. J.. 1997. Southeastern Santa Cruz. In S. D. Davis, V. H. Heywood, O. Herrera McBryde, J. Villa-Lobos, A. C. Hamilton, [eds.], Centers of plant diversity: a guide and strategy for their conservation. 3. The Americas, 416-420. World Wildlife Fund, Washington, D.C., USA and World Conservation Union, Oxford, UK..
Killeen T. J. Jardim A. Mamani F. Rojas M.. 1998. Diversity, composition and structure of a tropical semi-deciduous forest in the Chiquitanía region of Santa Cruz, Bolivia. Journal of Tropical Ecology 14: 803-807..[CrossRef][ISI]
Kitajima K.. 1994. Relative importance of photosynthetic traits and allocation patterns as correlates of seedling shade tolerance of 13 tropical trees. Oecologia 98: 419-428..[CrossRef][ISI]
Kitajima K. Poorter L.. In press. Functional basis for resource niche differentiation by tropical trees. In W. P. Carson, S. A. Schnitzer, [eds.], Tropical forest community ecology. Blackwell Publishing, Oxford, UK..
Klich M. A.. 2000. Leaf variations in Elaeagnus angustifolia related to environmental heterogeneity. Environmental and Experimental Botany 44: 171-183..[CrossRef][ISI][Medline]
Koch G. W. Sillett S. C. Jennings G. M. Davis S. D.. 2004. The limits to tree height. Nature 428: 851-854..[CrossRef][Medline]
Kuroki H. Tang Y. Koizumi H. Washitani I.. 2002. Effects of light and soil water availability on leaf photosynthesis and growth of Arisaema heterophyllum, a riparian forest understorey plant. Journal of Plant Research 115: 419-427..[CrossRef][ISI][Medline]
Kursar T. A. Coley P. D.. 1993. Photosynthetic induction times in shade-tolerant species with long and short-lived leaves. Oecologia 93: 165-170..[CrossRef][ISI]
Kursar T. A. Coley P. D.. 1999. Contrasting modes of light acclimation in two species of the rainforest understory. Oecologia 121: 489-498..[CrossRef][ISI]
McClendon J. H. McMillen G. G.. 1982. The control of leaf morphology and tolerance of shade by woody plants. Botanical Gazette 143: 319-330..
Mostacedo B. Justiniano J. Toledo M. Fredericksen S. T.. 2003. Guía dendrolólogica de especies forestales de Bolivia. Bolivia Forestal (BOLFOR), Santa Cruz de la Sierra, Bolivia..
Paiva E. A. S. dos Santos Isaisas R. M. Vale F. H. A. de Senna Queiroz C. G.. 2003. The influence of light intensity on anatomical structure and pigment contents of Tradescantia pallida (Rose) Hunt. cv. purpurea Boom (Commelinaceae) leaves. Brazilian Archives of Biology and Technology 46: 617-624..[ISI]
Parker G. Tinoco-Ojanguren C. Martinez-Yrizar A. Maass M.. 2005. Seasonal balance and vertical pattern of photosynthetically active radiation within canopies of a tropical dry deciduous forest ecosystem in Mexico. Journal of Tropical Ecology 21: 283-295..[CrossRef][ISI]
Parkhurst D. F. Loucks O. L.. 1972. Optimal leaf size in relation to environment. Journal of Ecology 60: 505-537..[CrossRef][ISI]
Poorter L.. 1999. Growth responses of fifteen rain forest tree species to a light gradient: the relative importance of morphological and physiological traits. Functional Ecology 13: 396-410..[CrossRef]
Poorter L. Bongers L. Bongers F.. 2006. Architecture of 54 moist forest tree species: traits, trade-offs, and functional groups. Ecology 87: 1289-1301..[ISI][Medline]
Poorter L. Bongers F. Sterck F. J. Wöll H.. 2005. Beyond the regeneration phase: differentiation of height–light trajectories among tropical tree species. Journal of Ecology 93: 256-267..[CrossRef][ISI]
Poorter L. Kwant R. Hernandez R. Medina E. Werger M. J. A.. 2000. Leaf optical properties in Venezuelan cloud forest trees. Tree Physiology 20: 519-526..[ISI][Medline]
Popma J. Bongers F. Werger M. J. A.. 1992. Gap-dependence and leaf characteristics of trees in a tropical lowland rain forest in Mexico. Oikos 63: 207-214..[CrossRef][ISI]
Quero J. L. Villar R. Marañón T. Zamora R.. 2006. Interactions of drought and shade effects on seedlings of four Quercus species: physiological and structural leaf responses. New Phytologist 170: 819-834..[CrossRef][ISI][Medline]
Roth I.. 1984. Stratification of tropical forests as seen in leaf structures, 521. Kluwer, Dordrecht, Netherlands..
Rozendaal D. M. A. Hurtado V. H. Poorter L.. 2006. Plasticity in leaf traits of 38 tropical tree species in response to light: relationships with light demand and adult stature. Functional Ecology 20: 207-216..[CrossRef]
Sack L. Grubb P. J.. 2003. Crossovers in seedling relative growth rates between low and high irradiance: analyses and ecological potential. Functional Ecology 17: 276-287..[CrossRef]
Sack L. Grubb P. J. Marañón T.. 2003. The functional morphology of juvenile plants tolerant of strong summer drought in shaded forest understories in southern Spain. Plant Ecology 168: 139-163..[CrossRef][ISI]
Sack L. Tyree M. T. Holbrook N. M.. 2005. Leaf hydraulic architecture correlates with regeneration irradiance in tropical rainforest trees. New Phytologist 167: 403-417..[CrossRef][ISI][Medline]
Sánchez-Gómez D. Valladares F. Zavala M. A.. 2006. Performance of seedlings of Mediterranean woody species under experimental gradients of irradiance and water availability: trade-offs and evidence for niche differentiation. New Phytologist 170: 795-806..[CrossRef][ISI][Medline]
Sims D. A. Pearcy R. W.. 1989. Photosynthetic characteristics of a tropical forest understorey herb, Alocasia macrorrhiza, and a related crop species, Colocasia esculenta, grown in contrasting light environments. Oecologia 79: 53-39..[CrossRef][ISI]
Sims D. A. Pearcy R. W.. 1992. Response of leaf anatomy and photosynthetic capacity in Alocasia macrorrizha (Araceae) to a transfer from low to high light. American Journal of Botany 79: 449-455..[CrossRef][ISI]
Smith W. K.. 1978. Temperatures of desert plants: another perspective on the adaptability of leaf size. Science 201: 614-616..
Strauss-Debenedetti S. Bazzaz F. A.. 1991. Plasticity and acclimation to light in tropical Moraceae of different successional positions. Oecologia 87: 377-387..[CrossRef][ISI]
Thomas S. C.. 1996. Asymptotic height as a predictor of growth and allometric characteristics in Malaysian rain forest trees. American Journal of Botany 83: 556-566..[CrossRef][ISI]
Valladares F. Martínez-Ferri E. Balaguer L. Pérez-Corona E. Manrique E.. 2000a. Low leaf-level response to light and nutrients in Mediterranean oaks: a conservative resource-use strategy?. New Phytologist 148: 79-91..[CrossRef][ISI]
Valladares F. Wright J. S. Lasso E. Kitajima K. Pearcy R. W.. 2000b. Plastic phenotypic responses to light of 16 congeneric shrubs from a Panamanian rainforest. Ecology 81: 1925-1936..[CrossRef][ISI]
Van Berloo J.. 1998. Pixels! Image analysis 1.3. Wageningen University, Wageningen, Netherlands..
Verweij R.. 2004. Tree architecture: dimensional relationships within 50 tree species from a Bolivian dry forest. Master's thesis, Wageningen University, Wageningen, Netherlands..
Whitmore T. C.. 1996. A review of some aspects of tropical rain forest seedling ecology with suggestions for further enquiry. In M. D. Swaine, [ed.], The ecology of tropical forest tree seedlings, 3-39. Programme on Man and the Biosphere, UNESCO series, vol. 17. Parthenon, Paris, France..
Wright I. J. Reich P. B. Westoby M.. 2001. Strategy shifts in leaf physiology, structure and nutrient content between species of high- and low-rainfall and high- and low-nutrient habitats. Functional Ecology 15: 423-434..[CrossRef]
Wright I. J. Westoby M.. 2002. Leaves at low versus high rainfall: coordination of structure, lifespan and physiology. New Phytologist 155: 403-416..[CrossRef][ISI]
Wright I. J. Westoby M.. 2003. Nutrient concentration, resorption and lifespan: leaf traits of Australian sclerophyll species. Functional Ecology 17: 10-19..[CrossRef]
Zimmermann M. H.. 1983. Xylem structure and the ascent of sap. Springer-Verlag, Berlin, Germany..
| |||||||||||||||||||||||||||
