Internal leaf anatomy and photosynthetic resource-use efficiency: interspecific and intraspecific comparisons

Tree Physiology 21, 251–259 © 2001 Heron Publishing—Victoria, Canada

Internal leaf anatomy and photosynthetic resource-use efficiency: interspecific and intraspecific comparisons S. MEDIAVILLA,1 A. ESCUDERO1 and H. HEILMEIER2 1

Departamento de Ecologia, Universidad de Salamanca, 37071 Salamanca, Spain

2

Lehrstuhl Pflanzenökologie, Universität Bayreuth, D-95440 Bayreuth, Germany

Received April 8, 1999

Summary Leaf mass per unit area (LMA) and internal leaf anatomy often affect net gas exchange because of their effects on internal CO2 conductance to the site of carboxylation, internal shading, competition for CO2 among carboxylation sites, nitrogen concentration and its partitioning. To evaluate effects of LMA and leaf anatomy on CO2 assimilation, water-use efficiency (WUE) and nitrogen-use efficiency (NUE), we measured LMA, leaf thickness, the thickness of mesophyll components, and gas exchange rates at ambient CO2 concentration in leaves of six woody deciduous and evergreen species with different leaf life spans. In two species, CO2 assimilation was also estimated at saturating CO2 concentrations. There were interspecific differences in all morphological variables studied. Long-lived leaves had higher LMA and were thicker than short-lived leaves. Species with high LMA had low assimilation rates and NUE, both in ambient and saturating CO2 concentrations. Thus, in species with high LMA, assimilation was reduced by non-stomatal limitations, possibly because of a lower allocation of N to the photosynthetic machinery than in species with low LMA. Within a species, thicker leaves tended to have a lower tissue density. In intraspecific comparisons under field conditions, increasing internal air volume had positive effects on WUE, probably because of enhanced internal CO2 conductance to the site of carboxylation. We conclude that, in interspecific comparisons, different patterns of N partitioning strongly influence NUE, whereas in intraspecific comparisons, internal leaf anatomy is a key factor regulating resource-use efficiency. Keywords: Crataegus monogyna, deciduous, evergreen, leaf mass per area, leaf thickness, leaf density, nitrogen, nitrogen-use efficiency, Pyrus bourgaeana, Quercus faginea, Quercus pyrenaica, Quercus rotundifolia, Quercus suber, water-use efficiency.

Introduction Leaf mass per unit area (LMA) has been proposed as an index of plant productivity (Kallis and Tooming 1974, Jurik 1986). Strong negative correlations have been reported between LMA and growth rates (Lambers and Poorter 1992, Nielsen et al. 1996) and between LMA and gas exchange rates, both in

terms of absolute assimilation rates and in terms of nitrogen-use efficiency (NUE) (Reich et al. 1992). A high LMA has been interpreted as an adaptation to drought (Salleo and Lo Gullo 1990) or an antiherbivore defense (Turner 1994), and a high LMA is normally associated with a long leaf life span. Accordingly, the negative effects of LMA on gas exchange may be the consequence of a compromise between productivity and leaf longevity (Reich et al. 1991a). Although effects of LMA are often comparable with those of leaf thickness, the two variables are not always correlated (Witkowski and Lamont 1991). The effects of leaf thickness and LMA on gas exchange can arise from effects of internal anatomy on the conductance of CO2 from the substomatal cavity to the chloroplasts (Nobel 1977, Niinemets 1999). In particular, several authors have considered the effects of the internal cell surface available for CO2 diffusion (Nobel 1977, Romero-Aranda et al. 1997) and the internal air volume (Parkhurst 1986, Evans and von Caemmerer 1996, Roderick et al. 1999a). A high density of photosynthetic material per unit area can lead to enhanced internal competition for CO2 and increased internal shading (Poorter et al. 1990). There may also be secondary effects resulting from species-specific differences in foliar N concentration or N allocation patterns, or both (Evans 1989), and from differences in stomatal conductance between high- and low-LMA leaves. The effects of internal anatomy on internal transfer conductances have been studied in laboratory measurements (Lloyd et al. 1992, Syvertsen et al. 1995), but no estimates have been made under field conditions, where differences in stomatal conductance and environmental influences can mask the effects of other leaf traits. Within a species, differences in LMA and their effects on gas exchange have been studied in relation to the light environment during leaf development (Walters and Field 1987) and in relation to changes associated with growth and senescence (Reich et al. 1991b). Drought-tolerant tree species are known to exhibit great anatomical plasticity (Dickson and Tomlinson 1996). A variety of leaf morphologies can be found even within the same individual (Barbero et al. 1992). Analyzing relationships between leaf morphology and gas exchange rates within a species may help elucidate the effects of leaf anatomy on net gas exchange in the absence of differences

252

MEDIAVILLA, ESCUDERO AND HEILMEIER

in stomatal conductance and leaf N concentrations, which are common among species differing in LMA. We evaluated effects of LMA and internal leaf anatomy on CO2 assimilation and resource-use efficiency within and between six Mediterranean woody species. We tested four hypotheses. (1) Changes in LMA are caused by variations in internal anatomy and leaf tissue density, and are not simply a consequence of changes in leaf thickness (Witkowski and Lamont 1991). In turn, variations in internal anatomy affect N concentrations, the density of photosynthetic tissues (Garnier and Laurent 1994) and gas exchange rates. (2) There is a negative correlation between LMA and photosynthetic nitrogen-use efficiency (PNUE) because of the effects of enhanced internal shading and competition for CO2 in the mesophyll of high-LMA leaves. Across species, variation in LMA is associated with large differences in N concentration per unit mass and in percent allocation to the photosynthetic machinery (Poorter and Evans 1998). These differences contribute to a further reduction in PNUE in high-LMA leaves. (3) Because internal leaf anatomy affects assimilation through its effects on internal CO2 conductance to the site of carboxylation, leaf anatomy plays an important role in determining water-use efficiency (WUE) for a given stomatal conductance. (4) Under field conditions, all of the above relationships can be masked by differences in stomatal conductances among leaves such that differences in N concentrations and internal leaf anatomy have only a marginal effect on the efficiency of assimilation of CO2. Materials and methods Study species and area The species, which were selected to exhibit a range of leaf longevities and leaf morphologies, included an evergreen oak species with leaf longevity greater than 2 years (Quercus rotundifolia Lam.), an evergreen oak species with leaf longevity only slightly greater than 1 year (Q. suber L.), two deciduous oak species with leaf abscission in autumn (Q. faginea Lam. and Q. pyrenaica Willd.) and two deciduous shrubs with leaf abscission in late summer (Pyrus bourgaeana Decne. and Crataegus monogyna Jacq.). The oak species and P. bourgaeana are typical overstory species, whereas C. monogyna can also behave as an understory species. All specimens selected for study were fully sun-exposed. The species were studied in two plots situated near Salamanca in central-western Spain. Altitudes range from 850 to 912 m a.s.l. Climate in the study area is cold Mediterranean. Mean annual temperatures are around 11–12 °C. Annual rainfall ranges from 400 to 500 mm, and there is a severe summer drought. The plots consisted of sparse populations (about 50 trees ha –1) of isolated mature trees with open pasture areas among them. Trunk diameters at 1.3-m height ranged from 20 to 60 cm for oak trees and about 5 cm for shrubs. Mean heights were about 6–10 m for trees and 2–3 m for shrubs. Leaf area index of the areas covered by the oak crowns ranged between 1.4 and 3.4.

Sampling methods On each measurement date, beween three and five individual trees of each species were randomly selected per plot. Leaf samples were taken from sunlit branches at mid-height in the canopy. Sampling was performed weekly throughout late spring and early summer over 4 years (1994–1997) for Q. rotundifolia and Q. faginea, and from 1995 through 1996 for the other species. Leaf samples were immediately taken to the laboratory for determination of leaf area (with a Delta-T Image Analysis System, Delta-T Devices Ltd., Cambridge, U.K.), thickness (with vernier calipers and with microscope sections), volume (thickness × area), dry mass (after desiccation at 80 °C for 24 h), LMA (dry mass/area) and tissue density (dry mass/volume) on at least 20 leaves of each species each week, taken at random after mixing the samples from the different trees and plots. The air space in the leaves was estimated as described by Yokoi and Kishida (1985) and Koike (1988). The amount of air space per leaf (Va, cm3) was calculated as: Va = Vf − (wf − wd) / rw − w d / rd,

(1)

where Vf (cm3 ) is volume of fresh leaves; wf (g) is fresh weight; wd (g) is dry weight; rw (g cm –3 ) is density of water and rd is density of dry matter. According to the results of Yokoi and Kishida (1985), rd ranges from 1.4 to 1.5 g cm –3 (we used rd = 1.45). Percentage air space was calculated as Vr = 100(Va /Vf). Nitrogen concentrations in all samples were determined by the micro-Kjeldahl method and with a CE-Instruments NA2100 autoanalyzer (ThermoQuest Italia, Milano, Italy). Seasonal trends of all variables were monitored and all data corresponding to immature leaves (less than 4 weeks old) were removed in order to restrict comparisons to completely mature leaves. Anatomical measurements About 50 mature leaves of each species, sampled at random from different individuals in June 1997, were used for microscopic observation. Small pieces from the middle region of the lamina were fixed in 5% glutaraldehyde for 2 h and postfixed with 1% OsO4 for 1.5 h, followed by dehydration and embedding in Epon resin (Tousimis Research Corp., Rockville, MD). Semi-thin sections were examined with the aid of a light microscope, and total thickness (excluding hairs) and thickness of palisade, spongy mesophyll, cuticle and epidermal layers were measured. Scanning microscope photographs of the abaxial epidermis were used to estimate mean distances between neighboring stomata. Gas exchange measurements Net photosynthesis was measured with an LI-6200 portable photosynthesis system (Li-Cor Inc., Lincoln, NE) at ambient CO2 concentrations (about 360 µl l –1), air temperatures (between 20 and 35 °C), relative humidities (20–50%) and saturating photosynthetic photon flux density (PPFD). Measure-

TREE PHYSIOLOGY VOLUME 21, 2001

INTERNAL LEAF ANATOMY AND PHOTOSYNTHESIS

ments were taken weekly from 0700 to 1000 h solar time on sunny days during late spring and early summer (before the onset of the summer drought), from 1994 to 1997 (68 sampling dates) on completely expanded sunlit current-year leaves on at least two trees of each species per plot. Leaves used for gas exchange measurements were transported to the laboratory for measurement of area, fresh and dry mass and N concentration. All samples taken during 1996 and 1997 (about 700 leaves in total) were also used for internal air space measurements. With this procedure the morphology of each individual leaf could be related to the gas exchange rates of the same leaf. We calculated leaf mass per area (LMA), N concentration per unit mass (Nmass), N concentration per unit area (Narea), assimilation per unit mass (Amass), assimilation per unit area (Aarea) and instantaneous nitrogen-use efficiency (PNUE). The ratio Aarea /g (assimilation per unit area/stomatal conductance) was taken as an estimate of water-use efficiency for a given vapor pressure deficit (intrinsic water-use efficiency). A leaf disc oxygen electrode system (Hansatech Ltd., Kings Lynn, U.K.) was used to measure assimilation at elevated CO2 concentrations under controlled laboratory conditions. During spring and early summer 1996, current-year leaves of Q. rotundifolia and Q. faginea were collected at weekly intervals and immediately taken to the laboratory. The leaf chamber was supplied with air containing 15% CO2 (cf. Quick et al. 1992). Air entering the chamber was previously humidified to saturation to prevent leaf desiccation during measurements. Temperature of the water-jacketed leaf chamber was maintained at 25 °C. Leaf discs were illuminated at gradually increasing PPFD until a maximum rate of O2 evolution was achieved (at about 800 µmol m –2 s –1). All the leaf samples were then used to determine LMA, leaf thickness and nitrogen concentration. Statistical analyses Because relationships between stomatal conductance and CO2 assimilation were curvilinear, nonlinear regressions were performed with the SPSS statistical package (SPSS Inc., Chicago, IL). The equations used for curve fitting were the rectangular hyperbola and saturating functions (Landsberg 1977). Goodness of fit was checked by examination of the residuals (Draper and Smith 1966). Residuals were further analyzed by correlating them with various leaf traits to determine the effects of these traits on resource-use efficiencies. Correlation analyses between the values of the residuals and those of other leaf traits can help elucidate the variation remaining after fitting the rectangular hyperbola and saturating functions (Sullivan et al. 1996). Species differences were analyzed by one-way ANOVA and the Fisher PLSD multiple range test (P < 0.05), after applying the Levene test to check for homogeneity of variances. Correlations between traits were estimated as Pearson’s correlation coefficients. Results Most leaf tissues increased in thickness with leaf longevity

253

(Table 1), although Q. suber leaves had a thicker palisade mesophyll layer than Q. rotundifolia leaves, despite the longer life span of Q. rotundifolia leaves. Deciduous shrubs (P. bourgaeana and C. monogyna) had a thicker epidermis and cuticle than leaves of the deciduous trees Q. pyrenaica and Q. faginea. Palisade mesophyll accounted for more than 50% of total leaf thickness in the deciduous oaks and Q. suber (Table 1). The cuticle constituted a much higher percentage of leaf thickness in Q. rotundifolia than in the other species. Mean distance between stomata was significantly higher in the deciduous shrubs C. monogyna and P. bourgaeana than in the oaks and was lowest in Q. suber. Leaf thickness was higher in species with long leaf life spans than in species with short leaf life spans. Within the deciduous species, oaks had thinner leaves than shrubs. Differences in the percentage of air volume (Vr) among the leaf types indicated that variation in leaf thickness among species was not paralleled by differences in LMA (Table 2). For example, deciduous shrubs (P. bourgaeana and C. monogyna) had leaves that were relatively thick, with low LMA. These findings imply that tissue density (dry mass/volume) varied among the species. Estimated tissue density decreased from: Q. rotundifolia > Q. faginea > Q. suber > P. bourgaeana > Q. pyrenaica > C. monogyna. Tissue density was highest in leaves with high proportions of palisade or cuticle, or both (cf. Witkowski and Lamont 1991). Within a species, the proportions of the different leaf tissues changed with leaf thickness (data not shown), and leaf tissue density in all species decreased as thickness increased, whereas the percentage of air volume inside the leaf increased with thickness (Table 3). During a year, LMA increased in the first weeks before leaf expansion was completed. Even after LMA had approached maximum values, there was a wide range of LMA values within a canopy position. On one day in June 1994, LMA for mature south-exposed leaves of a single tree of Q. faginea sampled before midday ranged from 142 to 195 g m –2. This suggests that there was intraspecific variability in LMA that was not related to any measured environmental parameters. Because differences in LMA were associated with shifts in most other measured leaf traits, we studied the effects of these traits on gas exchange. Among species, Nmass was generally significantly greater in low-LMA leaves than in high-LMA leaves (Table 4), whereas the reverse was found for Narea. The same tendencies were observed in intraspecific comparisons (data not shown). In most leaf types, LMA was negatively correlated to Nmass, whereas Narea and LMA were positively correlated, indicating that the total amount of nitrogen per leaf increased with LMA and that the reduction in Nmass was caused by a dilution effect and not by N limitation. All six species responded to drought with a large decrease in stomatal conductance (g). To examine the effects of leaf morphology on carbon assimilation it is necessary to take into account the dominant effects of the different conductances. In all species and leaf age classes and for the whole growth season, the relationship between g and Aarea showed a clear curvilinearity, with a strong reduction in the slope as g increased

TREE PHYSIOLOGY ONLINE at http://heronpublishing.com

254

MEDIAVILLA, ESCUDERO AND HEILMEIER

Table 1. Mean ± SE (n = 50) thickness (µm) and percentage of total thickness (in parenthesis) of the different tissues in cross sections of the different leaf types and mean distances (µm) between neighboring stomata. Means followed by the same letters within a variable are not significantly different (P < 0.05). Species

Leaf Palisade longevity (days)

Spongy mesophyll

Cuticle

Epidermis

Stomatal spacing

Q. rotundifolia Q. suber Q. faginea Q. pyrenaica C. monogyna P. bourgaeana

735 464 208 164 163 137

114 ± 2.08 a (38.8%) 92 ± 1.93 c (32.7%) 66 ± 2.05 e (34.1%) 56 ± 1.92 f (30.0%) 81 ± 3.69 d (34.1%) 101 ± 3.44 b (36.8%)

13.4 ± 0.19 a (4.5%) 8.0 ± 0.12 b (2.9%) 5.9 ± 0.12 d (2.8%) 5.7 ± 0.14 d (3.1%) 7.0 ± 0.18 c (3.2%) 7.1 ± 0.15 c (2.8%)

41 ± 0.60 b (14.0%) 21 ± 0.35 d (7.7%) 26 ± 0.74 c (12.2%) 24 ± 0.53 c (12.7%) 49 ± 1.20 a (21.9%) 40 ± 0.95 b (15.2%)

38.6 ± 8.12 a 28.7 ± 1.22 a 36.0 ± 9.31 a 30.9 ± 6.47 a 83.0 ± 16.2 b 77.3 ± 11.4 b

125 ± 1.61 b (42.6%) 158 ± 1.61 a (56.7%) 114 ± 1.96 c (53.9%) 102 ± 2.40 d (54.2%) 97 ± 4.34 d (40.7%) 123 ± 3.13 b (45.2%)

Table 2. Mean ± SE (n in parenthesis) thickness (µm), leaf mass per area (LMA, g m –2), percentage air volume (Vr, %) and mean density (mg cm –3) of the different leaf types. Means followed by the same letters within a variable are not significantly different (P < 0.05). Species

Thickness

LMA

Vr

Density

Q. rotundifolia Q. suber Q. faginea Q. pyrenaica C. monogyna P. bourgaeana

337 ± 5.16 a (256) 326 ± 17.50 a (26) 201 ± 3.02 c (330) 215 ± 6.51 bc (154) 233 ± 5.16 b (109) 221 ± 3.27 b (103)

242 ± 4.82 a (260) 201 ± 16.6 b (26) 141 ± 2.71 c (343) 124 ± 3.07 d (158) 121 ± 3.72 d (109) 135 ± 3.17 c (103)

24.6 ± 1.86 a (256) 26.1 ± 1.62 a (25) 34.6 ± 1.85 b (323) 45.9 ± 2.43 c (150) 52.7 ± 1.20 d (109) 42.6 ± 1.64 c (103)

723 ± 13.72 ab (256) 621 ± 30.01 c (26) 705 ± 14.31 b (330) 594 ± 18.23 c (154) 521 ± 12.70 d (109) 617 ± 15.06 c (103)

(Figure 1), suggesting that non-stomatal limitations, such as Rubisco activity and rate of electron transport (Farquhar and Sharkey 1982), operate at high g. The relationship between g and Aarea showed a good fit to a rectangular hyperbola of the form: Aarea = ( BCg / B + Cg) + D,

(2)

where B, C and D are constants. Regression parameters were calculated for each species, and the effects of g on Aarea quantified. The values of the residuals (observed – predicted assimilation) were used to determine the effects of variables other than g on Aarea (Sokal and Rohlf 1981). A positive residual for a measurement indicates a high intrinsic WUE (Aarea /g) for a given g in that particular leaf compared with a “standard” leaf

Table 3. Correlation coefficients between total leaf thickness and leaf mass per area (LMA), mean density and percentage air volume (Vr). Species

LMA

Density

Vr

Q. rotundifolia Q. suber Q. faginea Q. pyrenaica C. monogyna P. bourgaeana

+0.48**1 +0.39* +0.37** +0.18* +0.36** +0.28**

–0.49** –0.85** –0.60** –0.87** –0.79** –0.75**

+0.22** +0.93** +0.41** +0.71** +0.79** +0.57**

1

* = P < 0.05; ** = P < 0.01.

with identical g, because Aarea for the specified g is higher than expected. A positive effect of a trait on the residuals can be interpreted as a favorable effect in terms of increasing Aarea for a given g. Stomatal conductance also had a highly significant positive effect on PNUE. The relationship between PNUE and g was curvilinear and showed a good fit (Figure 2) to a saturation function of the form: PNUE = β(1 − exp(Γg)) + ∆,

(3)

where β, Γ and ∆ are constants. The residuals around the regression line fitted for each species (observed – predicted PNUE) were used to measure the effects of variables other than g on PNUE, as well as the differences in PNUE of the different leaves independently of their g values. (A positive effect of a trait on the residuals suggests that leaves possessing the trait achieve a higher PNUE than expected for the g they exhibit.) Leaf thickness and LMA were correlated with many aspects of gas exchange, both under field conditions and in a controlled environment in the leaf chamber of the oxygen electrode system. Species with thicker leaves or higher LMA, or both, had significantly lower maximum Amass, Aarea and g (Table 4), although leaves of Q. suber had a relatively high mean Aarea. Photosynthetic nitrogen-use efficiency was also lower in high-LMA leaves than in low-LMA leaves, whereas Aarea /g (intrinsic WUE) was usually higher in species with high LMA than in species with low LMA.

TREE PHYSIOLOGY VOLUME 21, 2001

INTERNAL LEAF ANATOMY AND PHOTOSYNTHESIS

255

Table 4. Mean stomatal conductance ± SE (mol H2O m –2 s –1), net photosynthetic capacity at ambient CO2 concentration on area (Aarea, µmol m –2 s –1), mass (Amass, nmol g –1 s –1) and nitrogen (PNUE, µmol g –1 N s –1) bases, and N concentration per unit area (Narea, g m –2) and per unit mass (Nmass, mg g –1) for the species studied during the period of the year when photosynthetic activity is at a maximum. For each variable, means not followed by the same letter are significantly different (P < 0.05). Species

n

g

Aarea

Amass

A/g

Narea

Nmass

PNUE

Q. rotundifolia Q. suber Q. faginea Q. pyrenaica C. monogyna P. bourgaeana

21 17 43 38 41 27

0.16 ± 0.006 ab 0.23 ± 0.016 bc 0.22 ± 0.014 bc 0.34 ± 0.018 d 0.21 ± 0.014 bc 0.27 ± 0.018 c

11.2 ± 0.47 ab 14.0 ± 0.70 cd 14.8 ± 0.47 de 16.7 ± 0.46 e 12.8 ± 0.52 bc 14.6 ± 0.49 cde

50.2 ± 3.2 a 81.5 ± 4.7 b 109.0 ± 3.8 c 141.0 ± 4.0 d 104.1 ± 5.9 c 111.3 ± 5.3 c

70.1 ± 2.6 a 62.0 ± 3.0 abc 71.5 ± 2.5 a 52.4 ± 1.9 c 65.0 ± 2.9 ab 56.3 ± 2.2 bc

3.09 ± 0.11 ab 3.37 ± 0.11 a 3.18 ± 0.09 ab 3.32 ± 0.11 a 2.46 ± 0.10 c 2.85 ± 0.11 bc

13.7 ± 0.54 a 19.5 ± 0.64 bc 23.8 ± 0.69 d 27.4 ± 0.64 e 19.6 ± 0.78 bc 22.1 ± 0.76 cd

3.65 ± 0.19 ab 4.19 ± 0.21 bc 4.61 ± 0.15 cd 5.21 ± 0.17 d 5.20 ± 0.31 d 5.07 ± 0.23 cd

There were no correlations between LMA or leaf thickness and g in most intraspecific comparisons, but PNUE tended to decrease as LMA increased, as indicated by the negative correlation between LMA and the residuals derived from the regression of PNUE on g (Table 5). In most intraspecific comparisons, Narea correlated positively with g (Table 5), and also showed positive effects on Aarea (data not shown). In many cases, Narea was also correlated with the residuals of the assimilation–conductance response model (Table 5), indicating direct positive effects of N concentration on the biochemical efficiency of assimilation. However, the increase in assimilation rate with Narea did not compensate for the increase in g, so that PNUE for a given conductance tended to be lower in high-N leaves than in low-N leaves (Table 5). Internal air volume generally was not significantly correlated with g (Table 5) or Aarea (data not shown). However, when compared to the residuals around the regression line between g and Aarea, relative air concentration (Vr) was strongly positively correlated with the residuals in most leaf types (Ta-

ble 5). Similar results were obtained for leaf thickness, whereas leaf tissue density generally had negative effects on the residuals (data not shown). Thus, although leaves with a high internal air volume did not exhibit higher absolute assimilation rates, the efficiency of the mesophyll for assimilating carbon seemed to be significantly improved by a high internal air volume when estimated for a given g. Differences between mean values of gas exchange of Q. faginea and Q. rotundifolia measured with an oxygen electrode were similar to those found in field measurements. The species with higher values for LMA and leaf thickness showed significantly lower values for Aarea, Amass and PNUE (Table 6).

Discussion Leaf thickness and LMA have often been reported to covary. However, different leaves may differ in tissue density, causing a lack of correlation between thickness and LMA (Witkowski and Lamont 1991). Differences in tissue density arise from

Figure 1. Relationship between stomatal conductance (g) and CO2 assimilation (Aarea) in leaves of Q. faginea (CO2 assimilation = ((31.7 × 145g)/(31.7 + 145g)) – 0.77; R 2 = 0.91 (n = 140)) and in leaves of Q. rotundifolia (CO2 assimilation = ((38.6 × 108g)/(38.6 + 108g)) – 0.61; R 2 = 0.80 (n = 129)).

TREE PHYSIOLOGY ONLINE at http://heronpublishing.com

256

MEDIAVILLA, ESCUDERO AND HEILMEIER

Figure 2. Relationship between stomatal conductance (g) and photosynthetic nitrogen-use efficiency (PNUE) in leaves of Q. faginea (PNUE = 5.97 (1 – exp – 9.16g) – 0.42; R 2 = 0.79 (n = 127)) and in leaves of Q. rotundifolia (PNUE = 4.8 (1 – exp –12.3g) – 0.8; R 2 = 0.78 (n = 129)).

high-LMA leaves. There were significant differences in Aarea /g between species (Table 4), suggesting that CO2 concentrations in intercellular air spaces tended to be lower in high-LMA leaves than in low-LMA leaves. In high-LMA leaves, increased competition for CO2 in the mesophyll could reduce CO2 partial pressure at the carboxylation sites (Poorter et al. 1990). However, if competition for CO2 were the only reason for the negative effects of LMA on Amass and PNUE, such effects should disappear under conditions of elevated CO2 concentrations. In the oxygen electrode study carried out at a CO2 concentration of 15% (Table 6), PNUE was lower in Q. rotundifolia, with the higher LMA, than in Q. faginea. The decrease in PNUE and Amass associated with a high LMA was of similar magnitude to that observed in the field, suggesting that other differences linked to high LMA caused the decrease in CO2 assimilation, such as a lower proportion of leaf N in the photosynthetic machinery (Evans 1989, Niinemets 1999, Roderick et al. 1999b). A comparison of the internal leaf anatomy of both species supports this conclusion. Quercus rotundifolia had much higher percentages of epidermis and

variations in internal anatomy that determine the volume of intercellular air spaces (Koike 1988), the amount of cuticle, the thickness of the cell walls (Witkowski and Lamont 1991) and the amount of supporting tissue (Salleo and Lo Gullo 1990). Oak species with a deciduous habit (Q. pyrenaica, Q. faginea) and evergreen Q. suber with a relatively short leaf longevity in this study were characterized by a high percentage of palisade mesophyll (Table 1). Species with a long leaf life span, especially Q. rotundifolia, had high percentages of cuticle (cf. Koike 1988). Deciduous shrubs had a large proportion of epidermis and spongy mesophyll, and consequently a significantly lower tissue density and a higher relative intercellular air space volume than the oak leaves, with the exception of Q. pyrenaica (Table 2). Under field conditions and in interspecific comparisons, mean assimilation rates were significantly lower in high-LMA leaves than in low-LMA leaves (Table 4). The differences were more pronounced when assimilation was expressed per unit mass or per unit nitrogen than per unit area (cf. Reich et al. 1992). Several explanations may account for the low PNUE of

Table 5. Correlation coefficients between N concentration per unit area (Narea ), leaf mass per area (LMA), leaf thickness and percentage air volume (Vr ) and: (A) stomatal conductance (g); (B) the residuals around the regression line between g and CO2 assimilation and (C) the residuals around the regression line between g and photosynthetic nitrogen-use efficiency (PNUE). Species

A

B

Narea Q. rotundifolia Q. suber Q. faginea Q. pyrenaica C. monogyna P. bourgaeana 1

+0.31** +0.19 +0.17** +0.17* +0.34** +0.74**

1

C

LMA

Thickness Vr

Narea

LMA

Thickness Vr

Narea

LMA

Thickness Vr

+0.16* –0.13 +0.13* +0.13 +0.15 +0.11

+0.005 +0.46 –0.09 –0.18* +0.008 +0.47**

+0.07 +0.47** +0.14* +0.18* +0.38** +0.20

+0.08 +0.30 +0.01 +0.05 –0.18* –0.51**

+0.07 +0.32 +0.44** +0.23* +0.35** +0.07

–0.40** –0.42** –0.44** –0.54** –0.29** –0.01

–0.29** –0.09 –0.42** –0.35** –0.45** –0.62**

–0.02 –0.68* +0.17* –0.06 +0.36* –0.13

–0.18** –0.10 –0.13 –0.28** +0.15 –0.27

+0.22** +0.21 +0.41** +0.25** +0.48** +0.07

* = P < 0.05; ** = P < 0.01.

TREE PHYSIOLOGY VOLUME 21, 2001

+0.41** +0.44 +0.36** +0.02 +0.28 +0.06

INTERNAL LEAF ANATOMY AND PHOTOSYNTHESIS

257

Table 6. Mean ± SE net photosynthetic capacity in saturating CO2 concentrations measured with an oxygen electrode on area (Aarea, µmol m –2 s –1), mass (Amass, nmol g –1 s –1) and nitrogen (PNUE, µmol g –1 N s –1) bases, and N concentration per unit area (Nmass, g m –2) and per unit mass (Nmass, mg g –1). Species

n

Aarea

Amass

Nmass

Narea

PNUE

Q. faginea Q. rotundifolia P-value

43 33

16.8 ± 0.76 12.9 ± 0.98 0.002

117 ± 4.2 66 ± 4.6 0.0001

17.6 ± 0.33 12.3 ± 0.25 0.0001

2.51 ± 0.06 2.41 ± 0.06 0.2709

6.82 ± 0.27 5.36 ± 0.37 0.0017

cuticle and a lower percentage of palisade mesophyll compared with Q. faginea (Table 1). Even within the same tree in a single period of the growth season, we observed considerable variation in LMA that was linked to changes in leaf tissue density, thickness, internal anatomy, N concentration and gas exchange parameters. At the intraspecific level, relationships between the different leaf traits investigated were similar among species with the same leaf habit, regardless of their phylogenetic affinity (Tables 3 and 5). This suggests that the correlations observed reflect fundamental relationships between morphological traits and physiological performance of the leaves, and that these relationships are not affected by phylogenetic constraints. In most leaf types, a high leaf thickness was positively associated with the residuals of the regression of Aarea on g (Table 5). Although the relationship was not always significant, it revealed a higher efficiency of CO2 assimilation for a given stomatal conductance in thick leaves within a species than in thin leaves. The positive effects of thickness, however, are not a result of the accumulation of photosynthetic tissue per unit area, because LMA did not show any positive effect on CO2 assimilation. Among the factors investigated, relative air volume exhibited the highest correlation coefficients against the residuals for most leaf types (Table 5). This result could be explained in terms of the effects of internal leaf anatomy on the conductance of CO2 from substomatal cavities to the carboxylation sites (Sharkey 1985, Koike 1988). An increase in mesophyll thickness presents a greater cell wall area for CO2 diffusion and so should tend to decrease liquid-phase resistance. On the other hand, an increase in thickness should tend to increase the path length from the stomata to cell wall surfaces, increasing gaseous diffusion resistance. The increase in thickness was normally associated with an increase in the percentage of total leaf volume occupied by air space (Table 3), which should tend to increase the space available for lateral diffusion of CO2. Lateral movement of CO2 can limit net assimilation of CO2 when the distance between stomata is high compared with mesophyll thickness (Parkhurst 1986). Mean distances between stomata ranged between about 83 µm in C. monogyna and 29 µm in Q. suber (Table 1). Although the latter value is rather low compared with the mean mesophyll thickness measured for the same species (Table 1), heterogeneity in stomatal closure can effectively increase the distance between open stomata (Kappen et al. 1995). In the heterobaric leaves of the species studied here, CO2 diffusion between compartments was impeded by bundle sheath extensions.

However, if the distribution of open stomata is random (Terashima et al. 1988), nonuniform stomatal opening within a compartment may cause large variation in CO2 supply to the different mesophyll cells of the compartment. Under these conditions, efficient lateral CO2 diffusion mediated by an increased internal air volume could reduce the variation in CO2 supply to the mesophyll cells, leading to a higher assimilation rate than is possible when lateral diffusion is restricted (Lloyd et al. 1992). Absolute assimilation rates are largely limited by stomatal conductance, especially when stomatal conductance responds to the vapor pressure deficit (VPD) between leaf and air (Farquhar and Sharkey 1982). Because stomatal conductance was unrelated to internal air space in most intraspecific comparisons (Table 5), the positive effects of internal air volume on absolute assimilation rates were masked (data not shown). However, there was often a significant positive correlation between percentage air space (Vr) and the residuals around the regression line between conductance and assimilation (Table 5), indicating that increased internal CO2 conductance results in an increased assimilation rate for a given stomatal conductance, rather than an increase in absolute assimilation rate. Therefore, a high internal air volume and efficient stomatal control contribute to increased water-use efficiency. Among the leaf adaptations investigated, those related to mesophyll anatomy had the most positive effects on PNUE and with no apparent costs. Stomatal closure, although necessary to avoid desiccation, had negative effects, including reductions in assimilation rate (Figure 1) and PNUE (Figure 2). Because leaf nitrogen concentration affects the biochemical efficiency of assimilation positively, it is positively related to the assimilation rate achieved for a given stomatal conductance (Table 5). However, increasing N concentration implies higher costs of acquisition, especially in infertile environments (Van der Werf et al. 1989). The need to compensate for these costs may explain why high-N leaves maintained elevated stomatal conductances (Table 5) and transpiration rates. Leaves with high Narea had a lower PNUE for a given stomatal conductance than leaves with low Narea (Table 5). Nevertheless, within a species, leaves with high internal air volume had a higher assimilation rate for a given stomatal conductance than leaves with a low internal air volume. We are unable to explain the strong intraspecific variability in leaf morphology in the studied species. We found no evidence of a disadvantage in an “open” mesophyll design that could explain the propensity of all of the study species to

TREE PHYSIOLOGY ONLINE at http://heronpublishing.com

258

MEDIAVILLA, ESCUDERO AND HEILMEIER

maintain numerous leaves lacking this apparently efficient design. It is possible that leaves with a low tissue density are less able to survive severe drought because of a lower resistance to physical damage by desiccation. Deciduous leaves showed significantly higher percentages of internal air volume and lower tissue densities than evergreen leaves (Table 2). This agrees with the tendency of deciduous species to shed their leaves in response to severe water deficits during summer (Escudero and Del Arco 1987), although other traits of deciduous leaves could explain their lower resistance to desiccation as well (Turner 1994). If these postulated negative effects of high internal air volume are also present at the intraspecific level, the variety of leaf morphologies that we observed could be interpreted as a response of Mediterranean plants to the diversity of stress factors typical in these environments. Acknowledgments This study received financial support from the European Community (Project No. STEP CT90-0037) and the Spanish Ministry of Education (Projects No. FOR89-0845 and AMB95-0800). H. Santiago performed most of the oxygen electrode measurements. References Barbero, M., R. Loisel and P. Quézel. 1992. Biogeography, ecology and history of Mediterranean Quercus ilex ecosystems. Vegetatio 99/100:19–34. Dickson, R.E. and P.T. Tomlinson. 1996. Oak growth, development and carbon metabolism in response to water stress. Ann. Sci. For. 53:181–196. Draper, N.R. and H. Smith. 1966. Applied regression analysis. John Wiley, New York, 543 p. Escudero, A. and J.M. Del Arco. 1987. Ecological significance of the phenology of leaf abscission. Oikos 49:11–14. Evans, J.R. 1989. Photosynthesis and nitrogen relationships in leaves of C3 plants. Oecologia 78:9–19. Evans, J.R. and S. von Caemmerer. 1996. Carbon dioxide diffusion inside leaves. Plant Physiol. 110:339–346. Farquhar, G.D. and T.D. Sharkey. 1982. Stomatal conductance and photosynthesis. Annu. Rev. Plant Physiol. 33:317–345. Garnier, E. and G. Laurent. 1994. Leaf anatomy, specific mass and water content in congeneric annual and perennial grass species. New Phytol. 128:725–736. Jurik, T.W. 1986. Temporal and spatial patterns of specific leaf weight in successional northern hardwood tree species. Am. J. Bot. 73:1083–1092. Kallis, A. and H. Tooming. 1974. Estimation of influence of leaf photosynthetic parameters, specific leaf weight and growth functions on yield. Photosynthetica 8:91–103. Kappen, L., G. Schultz and R. Vanselow. 1995. Direct observations of stomatal movements. In Ecophysiology of Photosynthesis. Eds. E.-D. Schulze and M.M. Caldwell. Springer-Verlag, Berlin, pp 231–246. Koike, T. 1988. Leaf structure and photosynthetic performance as related to the forest succession of deciduous broad-leaved trees. Plant Species Biol. 3:77–87. Lambers, H. and H. Poorter. 1992. Inherent variations in growth rate between higher plants: a search for physiological causes and ecological consequences. Adv. Ecol. Res. 23:187–261. Landsberg, J.J. 1977. Some useful equations for biological studies. Exp. Agric. 13:272–286.

Lloyd, J., J.P. Syvertsen, P.E. Kriedemann and G.D. Farquhar. 1992. Low conductances for CO2 diffusion from stomata to the sites of carboxylation in leaves of woody species. Plant Cell Environ. 15: 873–899. Nielsen, S.L., S. Enríquez, C.M. Duarte and K. Sand-Jensen. 1996. Scaling maximum growth rates across photosynthetic organisms. Funct. Ecol. 10:167–175. Niinemets, Ü. 1999. Components of leaf dry mass per area—thickness and density—alter leaf photosynthetic capacity in reverse directions in woody plants. New Phytol. 144:35–47. Nobel, P.S. 1977. Internal leaf area and cellular CO2 resistance: photosynthetic implications of variations with growth conditions and plant species. Physiol. Plant. 40:137–144. Parkhurst, D.F. 1986. Internal leaf structure: a three-dimensional perspective. In On the Economy of Plant Form and Function. Ed. T.J. Givnish. Cambridge Univ. Press, Cambridge, pp 215–249. Poorter, H. and J.R. Evans. 1998. Photosynthetic nitrogen use efficiency of species that differ inherently in specific leaf area. Oecologia 116:26–37. Poorter, H., C. Remkes and H. Lambers. 1990. Carbon and nitrogen economy of 24 wild species differing in relative growth rate. Plant Physiol. 94:621–627. Quick, W.P., M.M. Chaves, R. Wendler, M. David, M.L. Rodrigues, J.A. Passaharinho, J.S. Pereira, M.D. Adcock, R.C. Leegood and M. Stitt. 1992. The effect of water stress on photosynthetic carbon metabolism in four species grown under field conditions. Plant Cell Environ. 15:25–35. Reich, P.B., C. Uhl, M.B. Walters and D.S. Ellsworth. 1991a. Leaf life-span as a determinant of leaf structure and function among 23 tree species in Amazonian forest communities. Oecologia 86:16–24. Reich, P.B., M.B. Walters and D.S. Ellsworth. 1991b. Leaf age and season influence the relationships between leaf nitrogen, leaf mass per area and photosynthesis in maple and oak trees. Plant Cell Environ. 14:251–259. Reich, P.B., M.B. Walters and D.S. Ellsworth. 1992. Leaf life-span in relation to leaf, plant and stand characteristics among diverse ecosystems. Ecol. Monogr. 62:365–392. Roderick, M.L., S.L. Berry, I.R. Noble and G.D. Farquhar. 1999a. A theoretical approach to linking the composition and morphology with the function of leaves. Funct. Ecol. 13:683–695. Roderick, M.L., S.L. Berry, A.R. Saunders and I.R. Noble. 1999b. On the relationship between the composition, morphology and function of leaves. Funct. Ecol. 13:696–710. Romero-Aranda, R., B.R. Bondada, J.P. Syvertsen and J.W. Grosser. 1997. Leaf characteristics and net gas exchange of diploid and autotetraploid citrus. Ann. Bot. 79:153–160. Salleo, S. and M.A. Lo Gullo. 1990. Sclerophylly and plant water relations in three Mediterranean species. Ann. Bot. 65:259–270. Sharkey, T.D. 1985. Photosynthesis in intact leaves of C3 plants: physics, physiology and rate limitations. Bot. Rev. 51:53–105. Sokal, R.R. and F.J. Rohlf. 1981. Biometry: the principles and practice of statistics in biological research. W.H. Freeman and Co., San Francisco, 859 p. Sullivan, N.H., P.V. Bolstad and J.M. Vose. 1996. Estimates of net photosynthetic parameters for twelve tree species in mature forests of the southern Appalachians. Tree Physiol. 16:397–406. Syvertsen, J.P., J. Lloyd, C. McConchie, P.E. Kriedemann and G.D. Farquhar. 1995. On the relationship between leaf anatomy and CO2 diffusion through the mesophyll of hypostomatous leaves. Plant Cell Environ. 18:149–157.

TREE PHYSIOLOGY VOLUME 21, 2001

INTERNAL LEAF ANATOMY AND PHOTOSYNTHESIS Terashima, I., S.Ch. Wong, C.B. Osmond and G.D. Farqhuar. 1988. Characterisation of non-uniform photosynthesis induced by abscisic acid in leaves having different mesophyll anatomies. Plant Cell Physiol. 29:385–394. Turner, I.M. 1994. Sclerophylly: primarily protective? Funct. Ecol. 8:669–675. Van der Werf, A., T. Hirose and H. Lambers. 1989. Variation in root respiration: causes and consequences for growth. In Causes and Consequences of Variation in Growth Rate and Productivity of Higher Plants. Eds. H. Lambers, M.L. Cambridge, H. Konings and T.L. Pons. SPB Academic Publishing, The Hague, pp 227–240.

259

Walters, M.B. and C.B. Field. 1987. Photosynthetic light acclimation in two rainforest Piper species with different ecological amplitudes. Oecologia 72:449–456. Witkowski, E.T.F. and B.B. Lamont. 1991. Leaf specific mass confounds leaf density and thickness. Oecologia 88:486–493. Yokoi, Y. and A. Kishida. 1985. On the relationship between two indices (“Bulk Density” and “Dry-Matter Content”) of dry matter accumulation. Bot. Mag. Tokyo 98:335–345.

TREE PHYSIOLOGY ONLINE at http://heronpublishing.com