Plant Ecology

, Volume 195, Issue 2, pp 273–285

Light response in seedlings of a temperate (Quercus petraea) and a sub-Mediterranean species (Quercus pyrenaica): contrasting ecological strategies as potential keys to regeneration performance in mixed marginal populations


  • Jesús Rodríguez-Calcerrada
    • Unidad de Anatomía, Fisiología y Genética Forestal, Escuela Técnica Superior de Ingenieros de MontesUniversidad Politécnica de Madrid
    • Unidad Mixta INA-UPM
  • Jose Alberto Pardos
    • Unidad de Anatomía, Fisiología y Genética Forestal, Escuela Técnica Superior de Ingenieros de MontesUniversidad Politécnica de Madrid
    • Unidad Mixta INA-UPM
  • Luis Gil
    • Unidad de Anatomía, Fisiología y Genética Forestal, Escuela Técnica Superior de Ingenieros de MontesUniversidad Politécnica de Madrid
    • Unidad Mixta INA-UPM
  • Peter B. Reich
    • Department of Forest ResourcesUniversity of Minnesota
    • Unidad Mixta INA-UPM
    • Centro Nacional de Investigación Forestal (CIFOR)Instituto Nacional de Investigación Agraria y Alimentaria (INIA)

DOI: 10.1007/s11258-007-9329-2

Cite this article as:
Rodríguez-Calcerrada, J., Pardos, J.A., Gil, L. et al. Plant Ecol (2008) 195: 273. doi:10.1007/s11258-007-9329-2


In order to understand better the ecology of the temperate species Quercus petraea and the sub-Mediterranean species Quercus pyrenaica, two deciduous oaks, seedlings were raised in two contrasting light environments (SH, 5.3% full sunlight vs. HL, 70% full sunlight) for 2 years, and a subset of the SH seedlings were transferred to HL (SH–HL) in the summer of the second year. We predicted that Q. pyrenaica would behave more as a stress-tolerant species, with lower specific leaf area (SLA), allocation to leaf mass, and growth rate and less responsiveness to light in these metrics, than Q. petraea, presumed to be more competitive when resources, especially light and water, are abundant. Seedlings of Q. petraea had larger leaves with higher SLA, and exhibited a greater relative growth rate (RGR) in both SH and HL. They also displayed a higher proportion of biomass in stems (SMF), and a lower root to shoot ratio (R/S) in HL than those of Q. pyrenaica, which sprouted profusely, and had higher rates of photosynthesis (An) and stomatal conductance (gwv), but lower whole-plant net assimilation rate (NAR). On exposure to a sudden increase in light, SH–HL seedlings of both species showed a short period of photoinhibition, but fully acclimated photosynthetic features within 46 days after transference; height, main stem diameter, RGR and NAR all increased at the end of the experiment compared to SH seedlings, with these increases more pronounced in Q. petraea. Observed differences in traits and responses to light confirmed a contrasting ecology at the seedling stage in Q. petraea and Q. pyrenaica in consonance with differences in their overall distribution. We discuss how the characteristics of Q. petraea may limit the availability of suitable regeneration niches to microsites of high-resource availability in marginal populations of Mediterranean climate, with potential negative consequences for its recruitment under predicted climatic changes.


Acclimation to lightEcological requirementsCompetitive abilityMarginal populations


The existence of a transitional area between the Oceanic and Mediterranean regions in the Iberian Peninsula favours the occurrence of mixed stands of species with contrasting ecological requirements. There are many woody temperate trees in Mediterranean habitats in a relict situation (e.g. Pinus sylvestris, Ilex aquifolium, Frangula alnus, Fagus sylvatica, Quercus petraea) which is partly explained by their being more sensitive to local environmental stress conditions than co-occurring species that are at an optimum of their distribution. Usually drought arises as the principal limiting factor for these populations, operating either at the formation and dispersal of seeds (Hampe 2005) or the establishment of seedlings (Aranda et al. 2000; Castro et al. 2004; Valladares et al. 2005). The result is a patchy distribution at a landscape scale where individuals are constrained to relatively humid microsites. Nonetheless, the identification of key ecological characteristics in seedlings through the study of responses to light (e.g. shade-tolerance, acclimation potential and competitive ability) can provide valuable insights about niche segregation and recruitment potential even in drought-prone habitats. Comparative ecophysiological studies of closely related species carried out in controlled environments are useful for this purpose (Ashton and Berlyn 1994; Valladares et al. 2002a; Miyazawa and Lechowicz 2004).

Interspecific differences under common-environment conditions may indicate different selection pressures in their original habitats. Plant functional traits differ among species from a wide range of climate and habitat types in relation to the variation of the factors that determine the outcome of regeneration (Westoby et al. 2002; Reich et al. 2003; Wright et al. 2005). Species with high-specific leaf area, high-net assimilation rate, high partitioning of biomass to shoots and a rapid growth rate are likely to exhibit a high competitive potential in sites where interspecific competition for light is a key factor for regeneration (Poorter and de Jong 1999). In contrast, species adapted to arid- or semi-arid areas usually have leaves with high mass per unit area, a great partitioning of biomass to roots, a high root to shoot ratio and a conservative growth strategy (Chapin et al. 1993; Reich et al. 1999). This suite of traits provides a greater resistance to water stress, but may result in a less competitive ability to intercept light than that of fast-growing trees.

Responsiveness to light is also related with adaptation to the environmental conditions existing in the habitat. For instance, late successional species show lower phenotypic plasticity to light (Chazdon et al. 1996; Ribeiro et al. 2005) and lower acclimation potential to increasing light (Fetcher et al. 1983; Strauss-Debenedetti and Bazzaz 1991) than early successional species. Similarly, responsiveness to resource availability is relatively low in some Mediterranean woody species, as a result of an adaptation to limiting stressful environments (Valladares et al. 2002b; Chambel et al. 2005), so that one might expect a different response to light between seedlings of a sub-Mediterranean and a temperate species of similar successional status. The objective of this study was to identify ecological characteristics in seedlings of two co-occurring oaks differing markedly in their distribution, namely Q. petraea and Q. pyrenaica, through the leaf- to whole-plant examination of short- to long-term acclimatory responses to light. The possible implications of such responses in relation to low-latitude marginal populations of Q. petraea are discussed herein. Based on Grime’s classification of plant functional types (1977), we hypothesized that seedlings of the temperate Q. petraea would show features of a more competitive species while those of the sub-Mediterranean Q. pyrenaica would reflect a more stress-tolerant character. Particularly we anticipated a more conservative growth pattern, and a lower responsiveness to light (both lower phenotypic plasticity and acclimation potential to a sudden increase in light availability) in Q. pyrenaica than in Q. petraea.

Materials and methods

Study species and experimental design

Q. petraea and Q. pyrenaica are two deciduous white oaks (section: Lepidobalanus) forming late-successional forests in Atlantic and sub-Mediterranean areas, respectively. Q. pyrenaica extends from the southwest of France to northern Morocco while Q. petraea is distributed from the southern Scandinavian Peninsula to the northern Iberian Peninsula, where it coexists with Q. pyrenaica in a number of scattered stands.

In the autumn of 2003 acorns of both species were harvested in a small population of Q. petraea near the southern edge of its distribution (“El Hayedo de Montejo” forest, 41°7′ N, 3°30′ W, 1,300 masl). The seeds were kept at 3°C and 55% relative humidity until the next spring. Acorns of similar size from five different trees were then sown in 400-cm3 plastic pots of 35 cm depth, filled with a mixture of peat and sand (3:1, v/v) containing slow-release fertilizer (5 g dm−3). Plants were cultivated inside a glasshouse under two light treatments: shading (SH; double layer of neutral shadecloth, 5.3% of full sunlight) and high light (HL; no shadecloth, 70% of full sunlight, due to the opacity of fibreglass claddings). Photosynthetically active radiation (PAR) was measured using a LICOR LI-185B irradiance meter equipped with a LI-190SB quantum sensor. Average values of midday PAR were 80 ± 7 μmoles m−2 s−1 in SH and 1050 ± 28 μmoles m−2 s−1 in HL. Although midday temperature in HL was slightly higher than in SH (33.7 ± 0.3°C vs. 31.6 ± 0.2°C) it was assumed that 13-fold differences in relative irradiance were more important in explaining seedling responses.

At the end of the first growing season 80 seedlings of similar size were transplanted singly to 3,000-cm3 plastic cylinders of 40 cm depth. Each pot contained 1.9 kg of the same type of substrate used the first year and was again enriched with slow-release fertilizer. A total of 20 individuals per species were maintained in shade, distributed under four adjustable metal-framed shading structures placed on two benches of the glasshouse. Plants of the HL treatment were distributed between the shading structures at a sufficient distance as to avoid shading. About 90 days after leaf emergence, six plants per species were moved out of the shade frames (SH–HL seedlings) and mixed together with HL seedlings. Plants of all treatments were periodically moved on their benches and regularly watered over the course of the experiment. Bud burst was synchronous in both species (first week of March 2005), which precluded any difference in the length of the growth period. Temperature and relative humidity into the glasshouse were in the range of 15–39°C and 50–90%, respectively.

Physiological variables—gas exchange and chlorophyll fluorescence

Physiological acclimation was explored over the 46 days that followed the transference of SH plants (from 6 June [day 0] to 21 July [day 46]). Gas exchange and chlorophyll a fluorescence measurements were periodically conducted (more thoroughly immediately after transference) on one leaf per plant and five individuals per species and treatment. Leaves were selected from the first flush of growth and maintained for further analysis, if possible. Gas exchange and chlorophyll fluorescence was also measured on second-flush leaves of HL and SH–HL seedlings on day 46.

Photosynthesis (An) and stomatal conductance (gwv) were measured with an IRGA (LCpro Analytical Development Corporation, UK), at saturating light (1,000 μmol m−2 s−1 for HL seedlings and 700 μmol m−2 s−1 for SH and SH–HL) and 365 ppm CO2. A red/blue light emitting diode supplied measuring light. Temperature into the leaf chamber was set to leaf temperature being around 25°C in all cases. In order to avoid excessive temperature inside the glasshouse, gas exchange measurements started at 08:00 h and continued until 11:00 h local time. Prior to measurements, shade plants were allowed to acclimate to a level of artificial light similar to the measuring light for a period of 10–15 min.

Chlorophyll fluorescence was measured with a pulse-modulated fluorometer (FMS 2, Hansatech Instruments Ltd., UK). Maximum photochemical efficiency of photosystem II (PS II) (Fv/F= [Fm − Fo]/Fm) was calculated after measuring maximal (Fm) and minimal fluorescence (Fo) at dawn. After measuring leaf gas exchange, effective photochemical efficiency of PS II (ΦPSII) was calculated as (Fm′ − Fs)/Fm′, where Fm′ is the maximal fluorescence in light and Fs is the steady-state light-adapted fluorescence. Dark- and light-adapted fluorescence was measured in the same area to estimate non-photochemical quenching (NPQ = [Fm − Fm′]/Fm′). Since sunlight was used as the actinic light, measurements were made on cloudless days at roughly the same hour, avoiding either shading the sample area or modifying the leaf angle.

Specific leaf area and chlorophyll content

Chlorophyll content was estimated on one first-flush leaf from five individuals per species and treatment in three dates (days 0, 37 and 61 from transference). For chlorophyll analysis, one leaf disc was cut and immersed in a tube containing 5 ml of dimethyl sulfoxide. Tubes were bathed in darkness for 5 h at 60°C and the absorbance of dimethyl sulfoxide was further measured at 648.2 and 664.9 nm wavelength with a spectrophotometer, determining chlorophyll content per unit of leaf mass (Chlmass). Values of specific leaf area (SLA) were used to calculate chlorophyll content on a leaf area basis (Chlarea).

Leaf morphology and plant architecture

The SLA was determined on one fully exposed leaf from the first flush of growth in five individuals per species and treatment. Discs of known area were cut, oven-dried at 70°C for 48 h and then weighed to estimate SLA as the ratio between the area and its dry mass. At day 61, SLA of second-flush leaves was also determined in HL and SH–HL seedlings. Internode length (IL) was calculated by dividing the length between the lowest leaf and the tip of the main stem by the number of nodes. Mean leaf size (LA) was calculated by dividing the total leaf area of the plant (TLA) at the end of the experiment by the number of leaves produced. In order to estimate the self-shading in a seedling (SS) a rough index was calculated by dividing TLA by the shoot height.

Growth and biomass distribution

Height and stem diameter at the root collar were measured at the time of transplanting (end of the first year), and in 3 days along the second growing season (days 0, 38 and 67).

At the moment of transplanting, six seedlings per species from the HL and SH treatments were harvested. At the end of the experiment (day 67), another six seedlings per species were harvested in the HL, SH and SH–HL treatments. The leaves, stems and roots were dried to a constant weight for 3–6 days, and further weighed separately. All of the leaves on each plant were separated as pertaining to the main stem versus the sprouts, as well as to the first-flush of growth versus the following flushes. Leaf area ratio (LAR; total leaf area/total plant mass), leaf mass fraction (LMF; leaf mass/total plant mass), stem mass fraction (SMF; stem mass/total plant mass), root mass fraction (RMF; root mass/total plant mass) and root to shoot ratio (R/S; root mass/shoot mass) were calculated. The relative growth rate (RGR) for each species and light treatment combination was determined using the Eq. 1:
$$ {\text{RGR}} = \frac{{\overline{{\ln ({\text{M}}{}_{2})}} - \overline{{\ln ({\text{M}}{}_{1})}} }} {{{\left( {{\text{D}}_{2} - {\text{D}}_{1} } \right)}}}, $$
where M2 was the total plant-dry mass at the final harvest, M1 was the total plant dry mass at the time of transplanting, D2 was the day of the final harvest and D1 was the time of transplanting. The net assimilation rate (NAR) was calculated using the Eq. 2:
$$ {\text{NAR}} = \frac{{{\left[ {{\text{ln(}}\overline{{{\text{TLA}}_{{\text{2}}} }} {\text{)}} - {\text{ln(}}\overline{{{\text{TLA}}_{{\text{1}}} }} {\text{)}}} \right]}}} {{{\left( {\overline{{{\text{TLA}}_{{\text{2}}} }} - \overline{{{\text{TLA}}{}_{{\text{1}}}}} } \right)}}}\frac{{{\left( {\overline{{{\text{M}}_{2} }} - \overline{{{\text{M}}_{1} }} } \right)}}} {{{\left( {{\text{D}}_{{\text{2}}} - {\text{D}}_{{\text{1}}} } \right)}}}, $$
where TLA2 was the total leaf area at the final harvest, obtained by measuring the area of all the leaves with an image analyser (Delta-T Devices LTD, UK), and TLA1 was the total leaf area at the time of transplanting, which was estimated as the product of LMF by SLA.

Statistical analysis

Effects of species and light were tested by analysis of variance (ANOVA). A repeated measures approach was followed for all variables except those measured only at the end of the experiment. Despite variability among dates of measurement, no seasonal trend was observed in any variable in HL or SH seedlings (date effect at least P > 0.1). A significant interaction between species and light factors was considered as indicative of interspecific differences in plasticity (Schlichting 1986). Acclimation of the various traits studied on each species was evaluated comparing the means of the light treatments (HL, SH and SH–HL) by Tukey’s HSD test. The impact of transference on each species was examined comparing the maximum variation of each SH–HL plant upon exposure to light with respect to the mean value for SH plants. Logistic or lineal models were fitted to individual plants to evaluate the velocity of acclimation of physiological variables; independent parameters of each curve (n = 5) were then compared using ANOVA. Variables were arcsine or log-transformed to enhance homoscedasticity. Significance level was 5%.


Gas exchange parameters

Both the rate of photosynthesis (An) and stomatal conductance (gwv) were higher in Q. pyrenaica than in Q. petraea in HL, but not in SH (Table 1; Fig. 1a, b). Gas exchange of SH–HL plants declined below values of SH plants for the week following transference (Fig. 1c, d). The impact of the sudden increase in light did not significantly differ among species. Photoinhibition was temporary; An and gwv gradually increased in shade-formed leaves so that they did not differ significantly from those of HL plants at the end of the measuring period. Leaves of SH–HL plants developed after transference had gas exchange rates not significantly different from second-flush HL leaves (16.6 in HL vs. 17.7 μmol m−2 s−1 in SH–HL for An [both species combined; P > 0.1] and 402 in HL vs. 364 mmol m−2 s−1 in SH–HL for gwv [both species combined P > 0.1]).
Table 1

Two-way analysis of variance aimed to test the effects of light, species and the interaction of both factors on leaf and plant traits





Species × light






















Chlorophyll content









Morphology and architecture





















Growth and biomass distribution









Total plant mass
























A repeated measures approach (ANOVAR) was followed for those variables repeatedly measured throughout the experiment

*** P < 0.001; ** P < 0.01; *P < 0.05; n.s. P ≥ 0.05
Fig. 1

Plasticity to light (a, b) and acclimation of shade-grown seedlings to high light (c, d) on Q. pyrenaica (grey symbols) and Q. petraea (white symbols). (a, b) Mean values (+1 SE) of photosynthesis (An) and stomatal conductance (gwv) of high light (HL) and shade (SH) seedlings. (c, d) Mean values (±SE) of An and gwv of transferred seedlings (SH–HL), expressed as a percentage of HL mean value (thick line). The thin solid line and the thin dotted line indicate mean percent values of Q. petraea and Q. pyrenaica in SH, respectively

Chlorophyll fluorescence parameters

Light regime had a strong effect on the non-photochemical quenching (NPQ) and the effective quantum yield (ΦPSII), but not on the maximum quantum yield of PS II at dawn (Fv/Fm) (Table 1; Fig. 2a–c), indicating a daily build-up of photoinhibition reversible overnight. There were no statistically significant differences between Q. petraea and Q. pyrenaica in any chlorophyll fluorescence parameter. Transference of shade-acclimated plants to HL affected the activity of PS II (Fig. 2d–f). Immediately after transference, Fv/Fm of SH–HL plants declined with respect to SH plants, the reduction being slightly higher in Q. pyrenaica (0.76 vs. 0.80 on day 2; P < 0.05). ΦPSII also decreased with respect to SH plants (0.24, minimum value in Q. pyrenaica on day 2, and 0.31 the minimum value in Q. petraea on day 1; P < 0.05), while a nearly 3-fold increase over HL values occurred on NPQ in both species. By the second to fifth day after transference all chlorophyll fluorescence parameters began to recover, so that 46 days later shade-developed leaves reached values not significantly different from leaves continuously subjected to high light. The acclimation of chlorophyll fluorescence parameters was slightly more rapid in Q. pyrenaica (P < 0.05 for Fv/Fm and ΦPSII; P = 0.11 for NPQ) (Fig. 2d–f). High-light formed leaves of SH–HL plants had also values of ΦPSII and NPQ not significantly different from those of HL plants (0.54 in HL vs. 0.57 in SH–HL for ΦPSII [both species combined; P > 0.1] and 0.97 in HL vs. 0.79 in SH–HL for NPQ [both species combined P > 0.1]).
Fig. 2

Plasticity to light (ac) and acclimation of shade-grown seedlings to high light (df) on Q. pyrenaica (grey symbols) and Q. petraea (white symbols). (ac) Mean values (+1 SE) of predawn maximum photochemical efficiency of PS II (Fv/Fm), effective photochemical efficiency of PS II (ΦPSII) and non-photochemical quenching (NPQ) of high light (HL) and shade (SH) seedlings. (df) Mean values (±SE) of Fv/Fm, ΦPSII and NPQ of transferred seedlings (SH–HL), expressed as a percentage of HL mean value (thick line). The thin solid line and the thin dotted line indicate mean percent values of Q. petraea and Q. pyrenaica in SH, respectively

Chlorophyll content

Chlorophyll content (either Chlmass as Chlarea) was higher in Q. petraea in both environments, but especially marked was the difference in SH (Tables 1 and 2A). After transference, Chlmass declined in SH–HL plants to HL values (Table 2A), earlier in Q. pyrenaica (day 37, not shown). Chlarea was lowest in SH–HL leaves on day 61.
Table 2

Mean values (±SE) of variables related with chlorophyll content (A), leaf morphology and plant architecture (B), and growth and biomass distribution (C) 2 months after moving SH-grown plants to HL (SH–HL plants)





Q. pyrenaica

Q. petraea

Q. pyrenaica

Q. petraea

Q. pyrenaica

Q. petraea


Chlmass(mg g−1)

17.4 (1.0)b**

24.2 (0.6)b

10.5 (0.4)a**

13.1 (0.5)a

10.6 (0.7)a*

12.6 (0.6)a

Chlarea (g m2)

0.71 (0.10)ab**

0.90 (0.04)c

0.60 (0.05)a

0.67 (0.07)a

0.74 (0.09)b

0.80 (0.09)b


SLA (cmg−1)

244 (9)c*

268 (8)c

175 (9)b

196 (7)b

143 (8)a

157 (1)a

LA (cm2)

15.8 (2.1)a*

22.8 (2.6)a

19.2 (2.0)a**

31.3 (3.8)b

21.1 (2.9)a*

27.6 (1.2)ab

TLA (m2)

0.02 (0.001)a

0.03 (0.001)a

0.06 (0.01)b

0.1 (0.02)b

0.14 (0.01)c**

0.21 (0.01)c

IL (mm)

4.8 (2.1)

6 (2.0)

No data

No data

5.8 (0.6)**

12.8 (1.6)

SS (cmcm−1)

8.7 (1.2)a

12.1 (1.6)a

18.3 (3.4)b

16.5 (1.9)b

42.9 (3.5)c**

27.6 (1.7)c


Height (cm)

23.9 (3.2)a

22.7 (1.6)a

32.8 (4.4)a**

58.8 (11.3)b

32.6 (1.3)a**

75.9 (4.6)b

Diameter (mm)

4.9 (0.3)a

5.5 (0.3)a

5.7 (0.6)a**

9.2 (0.6)b

8.9 (0.6)b**

13.2 (0.4)c

Total plant mass (g)

5.7 (0.8)a

7.4 (0.7)a

14.9 (2.7)b*

25.0 (4.1)b

39.9 (3.1)c**

64.2 (4.9)c

LAR (cmg−1)

35.6 (2.1)a

36.8 (3.8)a

38.9 (3.1)a

37.7 (1.8)a

35.0 (1.7)a

32.7 (1.7)a

NAR (g m−2 day−1)







RGR (mg g−1 day−1)







Notice that SLA, Chlmass and Chlarea is of first-flush leaves, and that LA is calculated from all leaves (i.e. including those newly formed after transference in SH–HL plants)

Different letters indicate significant differences among light treatments within each species. The symbol indicates a significant difference between Q. pyrenaica and Q. petraea within each light treatment (* P < 0.1 and ** P < 0.05)

Leaf morphology and plant architecture

As expected, specific leaf area (SLA) was greatly increased by shading, and was slightly higher in Q. petraea in both light treatments (Tables 1, 2B). Interspecific difference in total leaf area (TLA) was only significant in HL, while the leaf size (LA) was significantly higher in Q. petraea than in Q. pyrenaica. The ratio between TLA and shoot height, used as an index of self-shading (SS), was lower in Q. petraea in HL. Although this index may give a good estimation of within-plant self-shading in HL, given the huge differences in height between the two species, this may not be the case in SH, where leaf arrangement along the stem should be considered. HL seedlings of Q. petraea had twofold longer leaf internodes than those of Q. pyrenaica (Tables 1, 2B). SLA of shade-developed leaves decreased significantly in both species, but by 61 days after transference SLA did not reach HL values (Table 2B). Newly developed leaves of SH–HL seedlings had similar SLA as second-flush HL ones (125 in HL vs. 131 cmg−1 in SH–HL [both species combined P > 0.1]). SH–HL seedlings had higher TLA, LA and SS than SH seedlings; in the case of LA, the increase in light had no effect in Q. pyrenaica (Table 2B).


All seedlings flushed two or three times in HL and once in SH. Interspecific differences were more evident in HL (Tables 1, 2C). Total plant-dry mass, height and main-stem diameter, and RGR, were all higher in Q. petraea. Differences in growth metrics between species were minimal in shade; hence plasticity in these variables was greater in Q. petraea. The pattern of shoot growth was also different in each species. In Q. pyrenaica new flushes did not emerge from apical or near-apical buds but mostly from basal sprouts near the root-collar, giving most seedlings a spherical-shaped aspect with older leaves shaded by the new ones. The net assimilation rate (NAR) was higher in Q. petraea, while no differences in LAR were apparent between species or light treatments. All SH–HL plants flushed at least once after transference, showing an increase in growth with respect to SH plants, which was more evident in Q. petraea. Differences between species for SH–HL seedlings held similar to those in HL (Table 2C).

Distribution of biomass

Biomass distribution differed more between species than among light environments. In HL, the species had similar biomass fraction in leaves (LMF), but the fraction in stems (SMF) was higher, and in roots (RMF) was lower, in Q. petraea than in Q. pyrenaica (Table 1; Fig. 3); accordingly, the root to shoot ratio (R/S) was greater in Q. pyrenaica (1.55 ± 0.05 vs. 0.94 ± 0.07 g g−1). Proportion of biomass from sprouts was 28% of total aerial biomass in Q. pyrenaica whereas it was negligible in Q. petraea. Light barely modified distribution of biomass in Q. pyrenaica but it did in Q. petraea, SH seedlings having lower LMF and higher RMF. Transference of shade-grown plants modified biomass partitioning. Distribution of biomass in leaves and stems increased at the expense of biomass partitioned to roots, more markedly in Q. petraea (Fig. 3). Number of flushes and leaves newly produced in HL after transference was similar in both species (not shown). However, larger leaves of Q. petraea resulted in higher biomass proportion of new to shade-grown leaves compared to Q. pyrenaica (83% vs. 68.4%; P < 0.01). While SH–HL seedlings of Q. petraea flushed from apical buds, most seedlings of Q. pyrenaica flushed from basal buds (0% vs. 20% above-ground biomass in sprouts in Q. petraea and Q. pyrenaica, respectively).
Fig. 3

Percent fraction of biomass distributed to leaves (dotted area), stems (hatched area) and roots (open area) at the end of the experiment on seedlings of Q. pyrenaica (grey bars) and Q. petraea (white bars). Different letters indicate significant differences among light treatments within each species. An asterisk indicates a significant difference between Q. pyrenaica and Q. petraea within each light treatment (P < 0.05)


Species differences in ecophysiological traits and in long-term responses to light

The species differed in their response to light availability, especially regarding growth-related parameters. Seedlings of Q. petraea in HL had a faster growth rate likely associated with its higher NAR and SLA (Veneklaas and Poorter 1998). Lower light-saturated rates of net photosynthesis per unit area in Q. petraea was compensated for by morphological and architectural traits—larger leaves, longer internodes and taller shoots—that likely increased both the plant’s efficiency for light capture and the overall carbon gain (Niinemets et al. 2002; Brites and Valladares 2005). On the other hand, both the sprouting habit and shorter internode length in Q. pyrenaica increased self-shading, which translated into a lower NAR despite its higher instantaneous leaf gas exchange rates in fully exposed leaves (Kamaluddin and Grace 1993; Valladares et al. 2002b). Lower growth rate of Q. pyrenaica could also be related to a higher storage of non-structural carbohydrates in the roots, as is often the case with sprouting species (Kruger and Reich 1997; Pausas et al. 2004; Schwilk and Ackerly 2005), or a greater carbon use in root respiration (Lambers et al. 1998). Provided likely ontogenetic effects, the inherent conservative growth of Q. pyrenaica seedlings point to an adaptive stress-tolerance strategy in the species (Chapin et al. 1993; Reich et al. 2003; Valladares et al. 2005), which is consistent with the hypothesis of a high phenotypic inertia in genotypes specialized to stressful environments (Valladares et al. 2000; Valladares et al. 2002b; Chambel et al. 2005).

Increases of chlorophyll concentration and specific leaf area in response to shade were consistent with an optimizing light-capture strategy reported in most species. At this respect, although an intermediate degree of shade-tolerance has been noted in seedlings of both species (Kelly 2002; Baraza et al. 2004), the higher SLA, Chlmass and LA exhibited by seedlings of Q. petraea in deep shade could reflect a greater capacity to tolerate shade (Niinemets and Kull 1994; King 2003). However, the results at the plant-level (mainly, rapid growth in high light, lower leaf mass fraction and similar leaf area ratio in shade compared to HL) were at odds with this conclusion (Walters and Reich 2000). Two years under deep shade could severely limit the maintenance of a positive carbon balance once carbohydrate supply from the acorns was exhausted. The large root system generated in the first year and the continuous lack of new surplus photosynthate may explain the fewer and smaller leaves produced in shade with respect to high light, and both the higher fractional proportion of root biomass and lower of leaf biomass. Values of RMF in HL in this study were similar to the ones observed for Q. pyrenaica and Q. robur (a closely-related species to Q. petraea) grown under optimal conditions (Antúnez et al. 2001), supporting that RMF was unusually high in our shade treatment. Further, we must consider that individual acclimatory responses to resources do not always match across-species evolutionary trends (see Valladares and Sánchez-Gómez 2006).

Acclimation to a sudden increase in irradiance

Our results indicated a rapid acclimation of shade-developed foliage to a higher light environment, as previously found in seedlings of other deciduous oaks (Naidu and DeLucia 1998). Species’ response was rather similar regarding leaf physiological traits, but clearly distinct in growth-related metrics. On exposure to high light, shade-developed leaves experienced a much higher radiation load that they could utilize in photosynthesis, resulting in a reduction in the maximum quantum yield of PS II at dawn (Fv/Fm), indicative of photoinhibition (Maxwell and Johnson 2000). The rapid onset of photoprotective mechanisms (NPQ) contributed to dissipate part of the excess of energy, preventing more severe damages through a down-regulation of PS II (as indicated by a temporary reduction in ΦPSII). But photosynthetic features rapidly acclimated to the new light intensity. Within days, the effective photochemical efficiency of PS II increased in parallel to a recovery in Fv/Fm and the relaxation of non-photochemical quenching, suggesting a rapid reorganization of PS II (Strauss-Debenedetti and Bazzaz 1991; Yamashita et al. 2000; Cai et al. 2005). Photosynthesis was also gradually increasing in parallel to an increase in stomatal conductance, whereas chlorophyll concentration in shade-grown leaves decreased on exposure to high light in both species as a result of chlorophyll degradation. Acclimation of shade-developed leaves may be particularly important in Mediterranean environments, where summer water stress restricts the probability of forming new flushes. However, transferred plants produced a new flush whose leaves were functionally similar to leaves continuously submitted to high light, which likely contributed to explain overall plant acclimation (Naidu and DeLucia 1998; Yamashita et al. 2000) and the greater acclimation potential in seedlings of Q. petraea under these controlled conditions. The higher proportion of newly formed surface area in high light in Q. petraea than in Q. pyrenaica contributed to reach a higher net assimilation rate and growth after transference. Besides, there was a differential pattern of growth on exposure to high light between the two oaks. The less conservative response of the seedlings of Q. petraea, i.e. both the lower sprouting and root to shoot ratio, reveals a differential shift in carbohydrate allocation of enhanced partitioning to main-stem, which may give a competitive advantage in terms of suppression of neighbouring vegetation and improved light harvesting.

Diverse ecological characteristics at the seedlings stage help to explain varying success in sub-Mediterranean habitats

Seedlings of Q. petraea and Q. pyrenaica differed in ecophysiological traits and responses, which suggests they might have contrasting competitive ability, and a segregation of regeneration niches according to resources’ availability. Such differences could play a role in the differential regeneration success of the species in coexisting sub-Mediterranean populations. Both the rapid growth and acclimation potential showed by Q. petraea are likely to enhance its competitive ability in non-water-limited and productive sites where among-species competition for resources is the driving force of regeneration (Keddy et al. 1997), as they enable overtopping of slower growing competitors (Küppers 1989; Cornelissen et al. 1996). But in Mediterranean ecosystems recruitment relies to a greater extent on the capacity of seedlings to endure the combination of multiple stresses and disturbances, such as nutrient or water shortages, wildfires or herbivore damages. Several features exhibited by seedlings of Q. petraea are less suitable to such conditions (Villar et al. 2004). First, lower root to shoot ratio can be related with either a more limited access to belowground resources or a lower ability to recover from above-ground disturbances. Second, high radiation may be a potential source of stress, especially in periods of strong water deficit and limited photosynthetic activity; even though the leaf area to plant biomass ratio was similar between species, higher self-shading in Q. pyrenaica seedlings minimizes the leaf area exposed to high-intensity radiation, thus maintaining water balance by reducing transpiration at the plant level and avoiding the probability of suffering severe photoinhibition and overheating in open sites (Bragg and Westoby 2002). Further, the sprouting habit of Q. pyrenaica observed in this study at the seedling stage is a typical trait of the species (Calvo et al. 2003), which constitutes an advantage to thrive in fire-prone habitats and unstable hill slopes (Sakai et al. 1995; Kruger and Reich 1997; Bond and Midgley 2003).

These results are consistent with the modest presence of Q. petraea in Mediterranean ecosystems, where it only forms small stands in favourable areas at high altitudes or northern exposures. Small-scale habitat heterogeneity in less suitable, low-latitude areas can thus play an important role in the recruitment of this species. For instance, Pardo et al. (2004) compared two nearby stands of Q. petraea and Q. pyrenaica within a forest in a Mediterranean mid-mountain site (the “Hayedo de Montejo” forest) and found out that the former was at a site of deeper and more humid soil. However, a recent inventory of the whole forest addressing the evolution of vegetation since year 1994 showed that Q. petraea was not able to spread over open areas, and that under the forest canopy the increase in measurable saplings was lower compared to the increase of Q. pyrenaica (Nanos et al. unpublished results). It is very likely that future warming and severity of droughts will accelerate the reduction of suitable regeneration niches for this and likely many other temperate trees near the southern edge of distribution, increasing the probability of extinction of such populations (Willi et al. 2006).


We thank Sven Mutke, Rebecca Montgomery, Alvaro Soto and Ruben Milla for valuable comments and suggestions. We are also grateful to Jesús Alonso for technical assistance and Maria del Carmen del Rey for help in plant measurements. This work was supported by the Consejería de Medio Ambiente y Desarrollo General de la Comunidad Autónoma de Madrid. J. Rodríguez-Calcerrada was supported by a scholarship from the Consejería de Educación de la Comunidad de Madrid (C.M.) and the Fondo Social Europeo (F.S.E.).

Copyright information

© Springer Science+Business Media B.V. 2007