## Abstract

Aims

Recent work has identified a worldwide ‘economics’ spectrum of correlated leaf traits that mainly reflects the compromises between maximizing leaf longevity and short-term productivity. However, during the early stages of tree growth different species tend to exhibit a common strategy, because competition for soil water and nutrients forces the maximization of short-term productivity owing to the need for rapid growth during the most vulnerable part of the tree’s life cycle. Accordingly, our aim here was to compare the variations that occur during ontogeny in the different leaf traits (morphology and leaf chemical composition) of several coexisting Mediterranean woody species differing in their leaf life spans and to test our hypothesis that tree species with a long leaf life span should exhibit larger shifts in leaf characteristics along ontogeny.

Methods

Six Mediterranean tree species differing in leaf life span, selected from three plots located in central-western Spain, were studied during three growth stages: seedlings, juveniles and mature trees. Leaf life span, leaf morphology (leaf area, dry weight, thickness and mass per unit area) and chemical composition (N and fibre concentrations) were measured in all six species. The magnitude of the ontogenetic changes in the different traits was estimated and related to the mean leaf longevity of the different species.

Important Findings

Along ontogeny, strong changes were observed in all variables analysed. The early growth stages showed lower leaf thickness, leaf thickness and mass per unit area and N, cellulose and hemicellulose concentrations than mature trees, but a higher lignin content. However, these changes were especially marked in species with a longer leaf life span at maturity. Interspecific differences in leaf life span, leaf morphology and chemical composition were stronger at the mature stage than at the seedling stage. We conclude that greater plasticity and more intense strategy shifts along ontogeny are necessarily associated with long leaf life span. Our results thus provide a new aspect that should be incorporated into the analysis of the costs and benefits associated with the different strategies related to leaf persistence displayed by the different species. Accordingly, the intensity of the alterations in leaf traits among different growth stages should be added to the suite of traits that change along the leaf economics spectrum.

## INTRODUCTION

In recent years, considerable research has focussed on ontogenetic changes in several leaf traits in different woody species. Several studies have reported changes in leaf morphology (England and Attiwill 2006; Ishida et al. 2005), leaf phenology (Mediavilla and Escudero 2009; Seiwa 1999) and photosynthesis and water-use strategies (Cavender-Bares and Bazzaz 2000; Drake et al. 2011; Mediavilla and Escudero 2004) as trees age. Describing such changes along ontogeny is indispensable for understanding differences in productivity between species and clarifying their competitive relationships (Ryan et al. 1997). Accordingly, research on ontogenetic changes is essential in order to be able to predict the responses of tree species and forest ecosystems to anthropogenic environmental change (Phillips et al. 2008).

The various studies that have emphasized the importance of the ontogenetic changes in woody species have focussed on a single species (Cavender-Bares and Bazzaz 2000; Drake et al. 2011; Hanson et al. 1994) or only a very few species of similar leaf longevity (Abdul-Hamid and Mencuccini 2009; Thomas 2010). However, no detailed analysis has been made of whether the morphological and chemical changes taking place in leaves among different growth stages occur equally in species of contrasting leaf characteristics. Leaf life span is a pivotal trait in the carbon fixation ‘strategy’ of a species (Wright et al. 2002). Species with a long average leaf life span tend to have high leaf thickness and mass per unit area (LMA), low nutrient concentrations and slow maximum photosynthetic rates (Poorter et al. 2009; Reich et al. 1999). Also, the presence of higher concentrations of lignin, cellulose and hemicellulose in leaves with a long life span has often been proposed (Chabot and Hicks 1982; Freschet et al. 2010; Takashima et al. 2004). Together, these traits form a spectrum of ‘leaf economics’, which has been analysed in depth by many researchers (Wright et al. 2004; Wright and Sutton-Grier 2012). However, to our knowledge, differences in ontogenetic shifts between species along the ‘leaf economics spectrum’ have not yet been analysed.

Our aim here was to compare the variations that occur during ontogeny (seedlings, juveniles and mature trees) in different leaf traits (morphology and leaf chemical composition) of several coexisting Mediterranean woody species differing in their leaf life spans. There are numerous studies that provide data on traits such as LMA or nutrient content, both in seedlings and in adult specimens of different woody species (England and Attiwill 2006; Ishida et al. 2005; Mediavilla and Escudero 2003a; Wright et al. 2004). However, despite the crucial role of fibres in mechanical defences (Lucas et al. 2000) and leaf persistence, data on the leaf fibre contents are much scarcer (Castro-Díez et al. 1997; Damesin et al. 1997; Mediavilla et al. 2008), particularly regarding the early stages of the life of the tree (Staudt et al. 2001; Villar et al. 2006). Moreover, to our knowledge, no authors have analysed how changes in these traits during ontogeny occur in species with different leaf longevities.

In previous studies working with Quercus species, we reported that the greatest differences between adults and seedlings are observed in evergreen species (Mediavilla and Escudero 2003a, 2004). Adult trees of the evergreen Quercus ilex showed a conservative water-use strategy, while this same species was closer to the water-spending strategy (sensuLevitt 1980) in the seedling stage. In contrast, deciduous species maintained more similar leaf traits and a similar wasteful strategy in both growth stages. The result is that species differing in leaf habits were seen to be more similar in the seedling stage than in the mature stage. Accordingly, our hypothesis is that the changes in leaf traits with tree maturity should differ in magnitude among co-occurring species differing in leaf life span, and that the ontogenetic transition in leaf characteristics would be faster and more marked in species with higher leaf duration in the adult growth stage.

## MATERIALS AND METHODS

### Study species and area

Six woody species with contrasting leaf life spans at the adult stage were included in this study: the deciduous Quercus faginea Lam. and Quercus pyrenaica Willd., the evergreen Quercus suber L. and Q. ilex ssp ballota (Desf.) Samp., with a mean leaf life span at the adult stage above ~1 and 2 years, respectively, and the conifers Pinus pinea L. and Pinus pinaster Aiton, with a mean leaf life span above ~3 and 4 years, respectively. These species were selected because of their abundance under Mediterranean climatic conditions and because they are widely used in afforestation activities in the study region.

The study was carried out at three plots located in the provinces of Salamanca and Zamora, in central-western Spain, between latitudes 41°10′ and 41°15′N and longitudes between 5°42′ and 5°52′W. Each species was studied in three growth stages: seedlings (3–5 years old), the juvenile stage (at an age of 10–12 years) and mature trees. However, an insufficient number of juveniles of Q. suber was found, and for this species only seedlings and mature trees were studied. The plots (Table 1) consisted of sparse populations of isolated mature trees (~50 specimens ha−1) over 100 years old, with open pasture areas among them where seedlings and juveniles were found. All tree age classes were represented in each plot. The height of the individuals selected for study was around 30cm for seedlings, between 1 and 2 m for juveniles and 5–8 m for mature trees. All specimens selected for the study were fully sun exposed.

Table 1:

sites characteristics

Characteristics Plot A Plot B Plot C
Elevation a.s.l. (m) 760 834 985
Climate
Average annual temperature (°C) 12.7 12.5 12.2
Average July–August temperature (°C) 21.8 21.3 22.0
Average December–January temperature (°C) 4.29 3.88 3.96
Average annual precipitation (mm) 404 462 518
Average summer precipitation (mm) 95 76 88
Soil
Sand content (%) 65.0 68.3 69.6
Clay content (%) 22.1 19.5 21.0
Silt content (%) 12.9 12.2 9.40
Organic matter (%) 3.35 2.37 2.60
pH 6.90 4.55 4.62
N (%) 0.044 0.039 0.052
Species Quercus ilex Quercus ilex Quercus ilex
Quercus faginea Quercus suber Quercus suber
Pinus pinea Quercus faginea Quercus pyrenaica
Pinus pinaster Quercus pyrenaica Pinus pinaster
Pinus pinaster
Characteristics Plot A Plot B Plot C
Elevation a.s.l. (m) 760 834 985
Climate
Average annual temperature (°C) 12.7 12.5 12.2
Average July–August temperature (°C) 21.8 21.3 22.0
Average December–January temperature (°C) 4.29 3.88 3.96
Average annual precipitation (mm) 404 462 518
Average summer precipitation (mm) 95 76 88
Soil
Sand content (%) 65.0 68.3 69.6
Clay content (%) 22.1 19.5 21.0
Silt content (%) 12.9 12.2 9.40
Organic matter (%) 3.35 2.37 2.60
pH 6.90 4.55 4.62
N (%) 0.044 0.039 0.052
Species Quercus ilex Quercus ilex Quercus ilex
Quercus faginea Quercus suber Quercus suber
Pinus pinea Quercus faginea Quercus pyrenaica
Pinus pinaster Quercus pyrenaica Pinus pinaster
Pinus pinaster

Climate data were obtained from a digital climatic atlas of the Iberian Peninsula (Ninyerola et al. 2005): a set of maps of mean air temperature, precipitation and solar radiation with a resolution of 200 m using data from climate stations and a combination of geographical variables (altitude, latitude, continentality, solar radiation and terrain curvature). Soil characteristics were analysed at the laboratory according to the methods described in Chapmann and Pratt (1973) and Walkley and Black (1934). Two soil samples were taken at each plot: one at the surface (excluding the forest floor) and the second at around 50-cm depth. Each sample was a composite of 12 subsamples collected at random. The environmental characteristics of the whole study area are fairly homogeneous (Table 1). The climate is cold Mediterranean, most precipitation falling during the winter and spring. As a consequence, water limitation is usually absent during spring and early summer. However, a period of summer drought occurs each year. The soils are poor in organic matter and in nutrient contents, having a medium/low pH (Table 1).

### Leaf measurements: demography, morphology and chemical composition

Several specimens of each species and growth stage were selected randomly at each of the plots, affording a total of 15 specimens per species and growth stage. From these individuals, a sampling of branches with leaves was performed during the autumn of 2012. The branches were taken from the top of the canopy to ensure that they were fully sun exposed. The samples of the evergreen species were separated into annual segments (shoots) of different age classes. Only one flush of leaf growth was observed in all species. Accordingly, all the leaves born in one particular year were considered to belong to the same age class. The number of leaves or needles per shoot was counted for each age class, and the mean age of the leaf population was estimated from the data. Data on leaf life span for the adults were taken from Mediavilla and Escudero (2003b). The leaf life span of juveniles and seedlings of the evergreen species was estimated by the following equation:

$(mean leaf age in seedlings and juveniles × leaf life span of mature trees)(mean leaf age in mature trees)$

The leaf life span of deciduous species was determined as the time elapsed between leaf emergence and the time when half of the leaves had been shed.

Fifteen specimens per species and per growth stage were also selected for leaf morphology and chemical composition analyses. A composite sampling of sun-exposed branches with leaves from different crown positions was undertaken for each individual selected. Samples were also taken in early autumn of 2012, when the current year leaves had fully expanded and the oldest surviving leaves were non-senescent. The samples were taken immediately to the laboratory and the leaves were separated into different age classes. To facilitate comparisons, in the evergreen species, only the youngest leaf cohort was included in the study. Young pine trees bear two kinds of needles: adult leaves (in paired fascicles) and juvenile leaves, which are single and smaller in size. The two types of needles were studied separately. For juveniles and mature trees, a total of 50 leaves of each of the 15 individuals selected for each species were used for the measurements. Owing to their small size, in seedlings we were forced to reduce the number of leaves selected for each individual to 25.

On each leaf or needle, thickness was measured with a digital micrometer (Digimatic micrometer, Mitutoyo, Japan) as a mean of three measurements taken at random positions, avoiding the main ribs in flat leaves. The total projected leaf and needle areas were determined by image analysis (Delta-T Devices LTD, Cambridge, UK). For Pinus, we also measured needle length with the digital micrometer. The samples were then oven dried at 70°C to constant mass and the total dry mass was determined. From the data thus obtained, we calculated the LMA. Leaf N concentrations were determined with a CE-Instruments NA-2100 autoanalyser (ThermoQuest, Milan, Italy). After N analysis, the remaining material was used to analyse the fibre contents (hemicellulose, cellulose and lignin), with an Ankom Analyser (A220; New York), following the Goering and Van Soest (1970) method. Owing to the high burden of time and money involved in determining fibre contents, in this case we limited the analysis to only two growth stages: seedlings and mature trees. For the Q. faginea, finally we did not have enough material for an analysis of fibre contents in the leaves of the seedlings. Both the N content and fibres content were expressed per unit mass (as nutrient or fibre milligrams per gram dry weight of leaf). The N content was also expressed per unit leaf area, obtained as the nutrient content multiplied by LMA.

Once all measurements had been made, a single average value for each of the leaf traits considered was obtained for each of the 15 specimens selected for each species and tree age class (average of the 50 or 25 leaves in the seedlings, selected for that age class). Finally, a single value for each leaf trait was obtained for each species and growth stage as the average for the 15 individuals, after checking that there were no significant differences in the values obtained for the same species and growth stage at the different sites where they appeared (data not shown). We quantified the relative magnitude of the ontogenetic transition for each site as the difference between seedlings and mature trees in the different leaf traits, according to the equation:

$100 × (trait value in mature trees − trait value in seedlings)(trait value in seedlings)$

Given the low presence of ‘juvenile’ needles after the third year of life in pine seedlings, this index was calculated only for ‘adult’ needle traits.

### Data analysis

Two-way analysis of variance (using species and growth stages as sources of variation) and Fisher’s protected least significant difference test were used to establish significant differences at P ≤ 0.05 between means after applying the Levene test to check for homogeneity of variances. Linear regression analysis was used to check if the magnitude of the ontogenetic changes in the different morphological traits and in N and fibre contents was related to the mean leaf longevity of the different species. The SPSS statistical package (SPSS Inc., Chicago, IL) was used for data analysis.

## RESULTS

The leaf life span of the ‘juvenile’ monophyllous needles of P. pinea and P. pinaster seedlings was much shorter than that of the ‘adult’ needles from the same specimens (Table 2). In fact, the mean duration of juvenile P. pinea needles was similar to that of the deciduous Q. faginea leaves. ‘Adult’ needles also had a shorter life span in seedlings as compared with the two other growth stages. Approximately half of the leaf biomass in P. pinea and P. pinaster seedlings was composed of juvenile needles during their first 3 years of life, giving an average leaf longevity of ~485 days in P. pinea and 663 days in P. pinaster; i.e. ~50% less than the mature trees of the same species. However, after the third year the proportion of monophyllous needles produced by the pine seedlings was much smaller. The differences in leaf life span between adult and juvenile pines were less intense, although still significant for P. pinea. Similar trends were obtained for Q. ilex and Q. suber, whereas in the deciduous Q. faginea the differences among the three growth stages were less pronounced, and the leaf life span in seedlings was even slightly greater than in the two other age classes (Table 2).

Table 2:

average leaf traits for the different leaf types studied (standard error in parentheses, n = 15)

Species Age class Leaf life span (days) Leaf size (cm2, mm) Leaf dry mass (mg) LT (μm) LMA (g·m−2Nmass (mg·g−1Narea (g·m−2
Pinus pinaster 1516 (21)a 133 (0.2)a 124 (4)a 951 (19)a 371 (4.4)a 9.01 (0.21)a 3.63 (0.15)a
1253 (60)a 92 (0.1)b 74 (0.5)b 792 (13)b 334 (3.0)b 8.22 (0.02)b 2.72 (0.12)b
Sa 943 (29)b 66 (0.5)c 44 (4)c 665 (24)c 288 (5.5)c 7.64 (0.28)c 2.29 (0.08)c
Sj 383 (22)c 32 (0.1)d 22 (0.8)d 420 (12)d 196 (2.8)d 7.10 (0.25)d 1.41 (0.14)d
Pinus pinea 1049 (15)a 102 (1.1)a 95 (2.6)a 556 (6)a 296 (4.5)a 10.6 (0.23)a 3.18 (0.09)a
842 (16)b 81 (0.7)b 68 (3.6)b 486 (8)b 271 (4.0)b 10.8 (0.14)a 3.16 (0.06)a
Sa 765 (51)c 56 (0.6)c 38 (0.7)c 413 (8)c 242 (2.7)c 9.43 (0.13)b 2.44 (0.07)b
Sj 235 (32)d 21 (0.5)d 18 (0.6)d 322 (8)d 190 (3.8)d 8.79 (0.22)c 1.67 (0.05)c
Quercus ilex 730 (18)a 4.58 (0.15)a 101 (3.2)a 333 (1.6)a 219 (3.2)a 13.9 (0.16)a 2.99 (0.05)a
746 (17)a 4.18 (0.14)a 86 (3.2)a 291 (2.2)b 211 (2.4)a 14.6 (0.15)a 3.06 (0.04)a
551 (23)b 3.26 (0.10)b 61 (2.0)b 252 (3.6)c 189 (2.5)b 12.6 (0.21)b 2.35 (0.04)b
Quercus suber 450 (16)a 5.90 (0.14)a 110 (2.7)a 267 (2.2)a 189 (1.7)a 15.4 (0.26)a 2.72 (0.05)a
406 (14)b 2.13 (0.05)b 35 (0.5)b 209 (3.3)b 167 (1.5)b 14.1 (0.12)b 2.32 (0.02)b
Quercus faginea 195 (16)a 7.75 (0.19)a 101 (3.1)a 217 (2.1)a 136 (1.6)a 19.1 (0.17)a 2.57 (0.04)a
206 (15)a 7.09 (0.23)a 93 (3.0)a 198 (1.8)a 137 (1.9)a 19.9 (0.19)a 2.69 (0.03)a
274 (20)b 5.20 (0.23)b 63 (2.6)b 175 (1.8)b 124 (1.8)b 17.7 (0.26)b 2.22 (0.04)b
Quercus pyrenaica 174 (5)a 22.2 (0.94)a 257 (9.9)a 192 (5.0)a 117 (1.3)a 19.6 (0.34)a 2.30 (0.04)a
170 (4)a 10.0 (0.38)b 112 (4.8)b 181 (4.2)a 116 (1.5)a 20.4 (0.29)a 2.32 (0.04)a
165 (6)a 9.29 (0.54)b 96 (4.2)b 160 (1.6)b 106 (1.6)b 18.6 (0.37)b 2.00 (0.05)b
Species × age class interaction <0.0001 NS NS <0.0001 <0.0001 <0.0001 <0.0001
Species Age class Leaf life span (days) Leaf size (cm2, mm) Leaf dry mass (mg) LT (μm) LMA (g·m−2Nmass (mg·g−1Narea (g·m−2
Pinus pinaster 1516 (21)a 133 (0.2)a 124 (4)a 951 (19)a 371 (4.4)a 9.01 (0.21)a 3.63 (0.15)a
1253 (60)a 92 (0.1)b 74 (0.5)b 792 (13)b 334 (3.0)b 8.22 (0.02)b 2.72 (0.12)b
Sa 943 (29)b 66 (0.5)c 44 (4)c 665 (24)c 288 (5.5)c 7.64 (0.28)c 2.29 (0.08)c
Sj 383 (22)c 32 (0.1)d 22 (0.8)d 420 (12)d 196 (2.8)d 7.10 (0.25)d 1.41 (0.14)d
Pinus pinea 1049 (15)a 102 (1.1)a 95 (2.6)a 556 (6)a 296 (4.5)a 10.6 (0.23)a 3.18 (0.09)a
842 (16)b 81 (0.7)b 68 (3.6)b 486 (8)b 271 (4.0)b 10.8 (0.14)a 3.16 (0.06)a
Sa 765 (51)c 56 (0.6)c 38 (0.7)c 413 (8)c 242 (2.7)c 9.43 (0.13)b 2.44 (0.07)b
Sj 235 (32)d 21 (0.5)d 18 (0.6)d 322 (8)d 190 (3.8)d 8.79 (0.22)c 1.67 (0.05)c
Quercus ilex 730 (18)a 4.58 (0.15)a 101 (3.2)a 333 (1.6)a 219 (3.2)a 13.9 (0.16)a 2.99 (0.05)a
746 (17)a 4.18 (0.14)a 86 (3.2)a 291 (2.2)b 211 (2.4)a 14.6 (0.15)a 3.06 (0.04)a
551 (23)b 3.26 (0.10)b 61 (2.0)b 252 (3.6)c 189 (2.5)b 12.6 (0.21)b 2.35 (0.04)b
Quercus suber 450 (16)a 5.90 (0.14)a 110 (2.7)a 267 (2.2)a 189 (1.7)a 15.4 (0.26)a 2.72 (0.05)a
406 (14)b 2.13 (0.05)b 35 (0.5)b 209 (3.3)b 167 (1.5)b 14.1 (0.12)b 2.32 (0.02)b
Quercus faginea 195 (16)a 7.75 (0.19)a 101 (3.1)a 217 (2.1)a 136 (1.6)a 19.1 (0.17)a 2.57 (0.04)a
206 (15)a 7.09 (0.23)a 93 (3.0)a 198 (1.8)a 137 (1.9)a 19.9 (0.19)a 2.69 (0.03)a
274 (20)b 5.20 (0.23)b 63 (2.6)b 175 (1.8)b 124 (1.8)b 17.7 (0.26)b 2.22 (0.04)b
Quercus pyrenaica 174 (5)a 22.2 (0.94)a 257 (9.9)a 192 (5.0)a 117 (1.3)a 19.6 (0.34)a 2.30 (0.04)a
170 (4)a 10.0 (0.38)b 112 (4.8)b 181 (4.2)a 116 (1.5)a 20.4 (0.29)a 2.32 (0.04)a
165 (6)a 9.29 (0.54)b 96 (4.2)b 160 (1.6)b 106 (1.6)b 18.6 (0.37)b 2.00 (0.05)b
Species × age class interaction <0.0001 NS NS <0.0001 <0.0001 <0.0001 <0.0001

Only data from the dominant leaf age class in the evergreen species are included. For each leaf type, means with different letters indicate significant differences between growth stages at P = 0.05. M = mature trees, J = juveniles, S = seedlings. For Pinus seedlings, Sa = ‘adult’ leaves, Sj = ‘juvenile’ monophyllous needles. LT = leaf thickness, Nmass = N concentration per unit mass, Narea = N content per unit area.

The seedlings always had smaller and lighter leaves than mature trees (Table 2). For the deciduous Quercus species, the leaves of juvenile and mature trees showed similar leaf thicknesses and LMA, with significantly higher values relative to seedlings (Table 2). In contrast, in Q. ilex leaf thickness was significantly different between adults and juveniles. For Pinus, the differences in leaf thickness and LMA between mature trees and seedlings were higher than for Quercus species, and the juveniles showed an intermediate degree of maturity between the two other age classes. Leaf thickness and LMA were especially low in the monophyllous needles of Pinus seedlings in comparison with both the ‘adult’ needles of the same specimens and the needles from the two other growth stages (Table 2).

In all species, the initial ontogenetic stage showed the lowest leaf N contents (especially the monophyllous needles of Pinus). There were no significant differences between juveniles and adults, with the exception of P. pinaster, for which N contents increased with tree age across all age classes (Table 2). The leaves in the early stages also showed a lower content of cellulose and hemicellulose than in the adult stage (Table 3). In Pinus, again the monophyllous needles reached the lowest contents for both fibre types, significantly lower than those of other age classes (Table 3). In contrast, the seedling leaves showed the highest lignin content, although finally the mature trees in general had a higher total fibre content (Table 3).

Table 3:

Average leaf fibres content for the different species and growth stages studied (standard error in parentheses, n = 15)

Species Age class Hemicellulose (H) (mg·g−1Cellulose (C) (mg·g−1H + C (mg·g−1Lignin (mg·g−1Total fibres content (mg·g−1
Pinus pinaster 88 (4.24)a 236 (6.37)a 324 (8.97)a 164 (4.90)c 488 (10.8)a
Sa 56 (2.44)b 135 (2.28)b 191 (3.42)b 236 (4.56)b 427 (5.36)b
Sj 28 (0.12)c 68 (0.27)c 97 (0.15)c 273 (5.00)a 370 (5.15)c
Pinus pinea 86 (5.37)a 211 (5.38)a 297 (8.63)a 116 (4.17)c 413 (7.77)a
Sa 61 (1.97)b 157 (5.08)b 218 (4.05)b 162 (6.66)b 380 (4.72)b
Sj 34 (1.63)c 92 (1.91)c 127 (2.24)c 220 (6.48)a 347 (5.69)c
Quercus ilex 81 (6.76)a 217 (3.17)a 298 (7.96)a 122 (1.49)b 420 (8.12)a
51 (2.13)b 194 (2.53)b 245 (3.22)b 145 (3.50)a 390 (4.34)b
Quercus suber 79 (3.13)a 193 (5.09)a 273 (7.51)a 153 (3.97)b 426 (9.46)a
57 (2.39)b 172 (1.62)b 229 (3.58)b 174 (9.54)a 403 (6.57)a
Quercus faginea 98 (6.06) 164 (4.49) 262 (8.54) 116 (3.10) 378 (8.29)
— — — — —
Quercus pyrenaica 101 (6.71)a 160 (2.63)a 261 (8.46)a 98 (3.45)b 359 (8.51)a
87 (3.20)b 146 (4.28)b 233 (5.32)b 109 (3.38)a 342 (3.35)b
Species x age class interaction NS <0.0001 <0.0001 <0.0001 NS
Species Age class Hemicellulose (H) (mg·g−1Cellulose (C) (mg·g−1H + C (mg·g−1Lignin (mg·g−1Total fibres content (mg·g−1
Pinus pinaster 88 (4.24)a 236 (6.37)a 324 (8.97)a 164 (4.90)c 488 (10.8)a
Sa 56 (2.44)b 135 (2.28)b 191 (3.42)b 236 (4.56)b 427 (5.36)b
Sj 28 (0.12)c 68 (0.27)c 97 (0.15)c 273 (5.00)a 370 (5.15)c
Pinus pinea 86 (5.37)a 211 (5.38)a 297 (8.63)a 116 (4.17)c 413 (7.77)a
Sa 61 (1.97)b 157 (5.08)b 218 (4.05)b 162 (6.66)b 380 (4.72)b
Sj 34 (1.63)c 92 (1.91)c 127 (2.24)c 220 (6.48)a 347 (5.69)c
Quercus ilex 81 (6.76)a 217 (3.17)a 298 (7.96)a 122 (1.49)b 420 (8.12)a
51 (2.13)b 194 (2.53)b 245 (3.22)b 145 (3.50)a 390 (4.34)b
Quercus suber 79 (3.13)a 193 (5.09)a 273 (7.51)a 153 (3.97)b 426 (9.46)a
57 (2.39)b 172 (1.62)b 229 (3.58)b 174 (9.54)a 403 (6.57)a
Quercus faginea 98 (6.06) 164 (4.49) 262 (8.54) 116 (3.10) 378 (8.29)
— — — — —
Quercus pyrenaica 101 (6.71)a 160 (2.63)a 261 (8.46)a 98 (3.45)b 359 (8.51)a
87 (3.20)b 146 (4.28)b 233 (5.32)b 109 (3.38)a 342 (3.35)b
Species x age class interaction NS <0.0001 <0.0001 <0.0001 NS

For each leaf type, means with different letters indicate significant differences between growth stages at P = 0.05. M = mature trees, S = seedlings. For Pinus seedlings, Sa = ‘adult’ leaves, Sj = ‘juvenile’ monophyllous needles.

For most of the variables analysed, the analysis of variance showed a significant species by tree age interaction, with mean values that changed in the same direction among tree age classes, but with different intensities for the different species (Tables 2 and 3). Leaf life span (with Q. faginea as an exception), leaf thickness, LMA, leaf N and fibre concentrations increased (or decreased in the case of the lignin content) markedly between seedlings and adults in all species. However, the magnitude of the differences between growth stages increased with mean leaf longevity at maturity of the species of study. The interspecific differences in the magnitude of the ontogenetic changes were mainly due to differences in the juvenile–adult transition, which was in general more pronounced for the species with longer leaf life spans. The magnitude of the ontogenetic transition in leaf traits was positively related to leaf longevity for the whole range of the six species (Figs 1 and 2). The same results were obtained when only the set of Quercus species was included, although in this case the analyses were limited to leaf morphological traits and N (Fig. 1) contents, since we did not have fibre contents for Q. faginea seedlings (Fig. 2).

Figure 1:

relationships between the intensity of the changes in leaf traits along ontogeny and leaf life span at maturity of the different species. The intensity of the ontogenetic change was evaluated as: 100 × (trait value in mature trees − trait value in seedlings)/(trait value in seedlings). Data are means of the values obtained in the different plots where each species was present (n = 1–3). Pp: Pinus pinaster, Pn: P. pinea, Qi: Quercus ilex, Qs: Q. suber, Qf: Q. faginea, Qp: Q. pyrenaica.

Figure 1:

relationships between the intensity of the changes in leaf traits along ontogeny and leaf life span at maturity of the different species. The intensity of the ontogenetic change was evaluated as: 100 × (trait value in mature trees − trait value in seedlings)/(trait value in seedlings). Data are means of the values obtained in the different plots where each species was present (n = 1–3). Pp: Pinus pinaster, Pn: P. pinea, Qi: Quercus ilex, Qs: Q. suber, Qf: Q. faginea, Qp: Q. pyrenaica.

Figure 2:

relationships between the intensity of the changes in leaf fibre concentrations along ontogeny and leaf life span at maturity of the different species. The intensity of the ontogenetic change was evaluated as: 100 × (fibre concentration in mature trees − fibre concentration in seedlings)/(fibre concentration in seedlings). Data are means of the values obtained in the different plots where each species was present (n = 1–3). Pp: Pinus pinaster, Pn: P. pinea, Qi: Quercus ilex, Qs: Q. suber, Qp: Q. pyrenaica. H+C = hemicellulose + cellulose concentrations.

Figure 2:

relationships between the intensity of the changes in leaf fibre concentrations along ontogeny and leaf life span at maturity of the different species. The intensity of the ontogenetic change was evaluated as: 100 × (fibre concentration in mature trees − fibre concentration in seedlings)/(fibre concentration in seedlings). Data are means of the values obtained in the different plots where each species was present (n = 1–3). Pp: Pinus pinaster, Pn: P. pinea, Qi: Quercus ilex, Qs: Q. suber, Qp: Q. pyrenaica. H+C = hemicellulose + cellulose concentrations.

## DISCUSSION

Our results confirm the significant differences between adults and seedlings reported by other authors for different woody species. As has been frequently observed (Cavender-Bares and Bazzaz 2000; Kolb and Stone 2000; Palow et al. 2012), in our study the leaves of seedlings also showed lower thickness, LMA and N concentration than the adults of the same species. In the intermediate growth stage, we also observed changes that suggested a maturation of leaf strategies with respect to the seedlings. The early growth stages showed lower cellulose and hemicellulose concentrations than mature trees, but a higher lignin concentration, although the total fibre content was finally higher in the adults. In this case, however, we are unaware of any study addressing ontogenetic changes in fibre contents with which our results can be compared.

These trends in the leaf traits with increasing tree age were repeated in all species studied. However, although the direction of the changes with plant maturation was similar, there were interspecific differences in the magnitude of the ontogenetic shift. The changes between growth stages tended to be more intense with increasing leaf longevity at maturity of the species studied. This trend was seen both at the morphological level as well as in leaf chemical composition. The result was that the species showed relatively similar leaf traits during the early stages, but diverged further as the trees aged. These interspecific differences in ontogenetic drifts were not only a consequence of the evergreen–deciduous dichotomy, since the differences between the evergreen species also revealed a direct effect of the differences in leaf life span. The ontogenetic trends were also similarly related to differences in leaf life span when the comparisons were made only for Quercus spp., which shows that these trends were independent of phylogeny.

The ontogenetic changes in leaf traits observed in the study species may be suitably analysed in the context of the trade-off between rapid resource acquisition vs. greater resource conservation, i.e. embodied in the leaf economics spectrum (Donovan et al. 2010; Heberling and Fridley 2012; Wright et al. 2002, 2004). Most leaf traits analysed revealed that, within the same species, young trees tended to be closer to the ‘fast acquisition’ end of the spectrum (low leaf life span and LMA and low contents of structural carbohydrates), whereas mature specimens were closer to the ‘conservation’ end. Evidently, species with a long leaf life span at maturity need to develop an intense shift in leaf traits to evolve from the fast acquisition strategy during their first stages to the conservation strategy at the adult stage. In contrast, species that are already close to the ‘fast acquisition’ end of the spectrum at the adult stage exhibit fewer changes along ontogeny. Important exceptions were the low N and high lignin concentrations found for the seedling stage, since a rapid acquisition of resources is supposedly correlated with leaf nitrogen, while high leaf lignin contents would instead reflect the resource conservation strategy (Freschet et al. 2010). However, also in these cases ontogenetic changes (increases in N and decreases in lignin concentration) tended to be more intense for species with longer leaf life spans.

These different types of behaviour seem to match the differences in water-use strategies between seedlings and adult trees found in previous studies (Mediavilla and Escudero 2003a, 2004). Mature oak trees showed lower stomatal conductances and higher stomatal sensitivity to drought than seedlings. The seedlings also tended to develop lower minimum water potentials at midday than adult trees (Mediavilla and Escudero 2004). These ontogenetic changes in water-use patterns seem to respond to presumable changes in the ability to access soil resources as a function of the rooting depth of the respective tree age classes. Seedlings have smaller and shallower root systems than mature trees and grow in soil layers that are most susceptible to soil drying, largely due to the transpiration of herbaceous vegetation (Cavender-Bares and Bazzaz 2000; Weltzin and McPherson 1997). Therefore, maintaining high stomatal conductances is the only way to ensure that seedlings can grow quickly before soil water is consumed by the neighbouring vegetation. Accordingly, seedlings tend to maintain fast water acquisition and consumption. Conversely, the roots of mature trees can penetrate into deeper soil layers where competition with the herbaceous vegetation is normally absent (Lyford 1980). This allows adults to develop a water-saving strategy, especially in species with a longer leaf life expectancy. Accordingly, at least with respect to soil water, adult trees tend to be closer to the resource conservation end of the leaf economics spectrum than their counterparts from the seedling stage. The properties of adult leaves thus reveal a change in the priority of resource use from a maximization of water consumption in seedlings to increased persistence in adults, which develop more xeromorphic leaves (higher LMA, thickness and structural carbohydrates content) than seedlings. These xeromorphic traits, as has been shown in numerous studies (Chabot and Hicks 1982; Takashima et al. 2004; Turner 1994), confer the mechanical protection to leaves needed to ensure persistence.

It has also been proposed that a higher content of lignin could be associated with an increase in leaf duration (Chabot and Hicks 1982; Freschet et al. 2010; Wright and Cannon 2001). The higher lignin concentration in seedlings therefore seems in conflict with respect to their shorter leaf duration. Lignin is a major constituent in water conduction. Numerous experiments (Coleman et al. 2008; Kitin et al. 2010; Voelker et al. 2011) have revealed that the reduction of lignin biosynthesis through genetic manipulation results in an increased susceptibility to collapse in the driver system and, therefore, a restriction of water flow through vessels, resulting in a significant effect on the rates of photosynthesis and transpiration. Maintaining high transpiration rates require a driver system that provides the adequate water supply imposed by the ‘wasteful’ strategy in the early stages. High lignin concentrations thus seem to be more related to the fast water consumption strategy typical of seedlings than to the requirements for leaf persistence. In fact, previous studies carried out by our team revealed a lack of correlation between the lignin content in the leaves of different species and their mean leaf longevities (Mediavilla et al. 2008).

Lower leaf N concentrations in seedlings also seem to contradict the fast resource acquisition strategy, since they should lead to low photosynthetic rates (Reich et al. 1999; Wright et al. 2004). In this sense, the trend to a fast resource acquisition strategy attributed to the early ontogenetic stages seems to be applicable only to soil water resources. Low N concentrations in seedlings might be a consequence of the need for a sufficient water supply to meet the high transpiration rates of their leaves. In a previous study (Juárez-López et al. 2008), we found that during their first years of life in seedlings of two Quercus species (Q. ilex and Q. faginea), >50% of the root biomass was invested in a single taproot that penetrated to a depth of 50cm, suggesting that they obtained their water mainly from a depth of around 50cm, where soil N concentrations are lower than at the surface. In contrast, many authors have reported that in Mediterranean-climate environments mature trees develop both a superficial and a deep root system, providing both a reliable groundwater supply and good access to nutrients near the soil surface (Dawson and Pate 1996; Filella and Peñuelas 2003). The differential development of root systems might explain the lower leaf N contents found for the seedlings in the present paper, although the evidence for this explanation is only indirect.

In conclusion, although plant species may be adequately classified according to their positions along the leaf economics spectrum (Donovan et al. 2010; Heberling and Fridley 2012; Wright et al. 2002, 2004), according to our results, during the early stages of growth interspecific differences along the leaf life span spectrum seem to be much less intense, as a result of the common water-spending strategy adopted during the seedling stage. One consequence of the interspecific similarity at the seedling stage is that the ontogenetic changes are much more intense for species with longer leaf longevity at maturity. The dual leaf system in the pine seedlings and the extremely short life span of the ‘juvenile’ leaves typical of these species, which disappear in later growth stages, is a clear manifestation of the need for dramatic shifts in strategy along ontogeny in species with a long leaf life span. Accordingly, at least in water-limited environments, intense changes along ontogeny might constitute another fundamental trait of species with long leaf life spans, which should be added to the suite of traits seen to vary along the leaf economics spectrum. It is reasonable to assume that increased plasticity along ontogeny implies additional costs (van Kleunen and Fischer 2005) for the species with longer leaf life span, which obviously needs to be taken into account when analysing the competitive relationships among different tree species differing in leaf life span. The differential behaviour of seedlings observed in the present study is probably a manifestation of a universal propensity of seedlings to suffer higher mortality rates than adults. In consequence, seedlings are also more prone to acquire and utilize resources quicker than adults, although in the present study this trend has been observed only with respect to water resources. More research is needed to confirm that the differences in the ontogenetic shifts associated with differences in leaf life span seen in this study are also present in other sets of species, and, especially, to investigate whether the same trends apply also to less water-limited environments.

## FUNDING

Spanish Ministerio de Ciencia e Innovación—EU-FEDER (Project No. BOS2002-02165, CGL2010-21187); the Regional Government of Castilla-León (Project No. SA040/03).

Conflict of interest statement. None declared.

## REFERENCES

Abdul-Hamid
H
Mencuccini
M
(2009)
Age- and size-related changes in physiological characteristics and chemical composition of Acer pseudoplatanus and Fraxinus excelsior trees
.
Tree Physiol

29
:
27
38
.
Castro-Díez
P
P
Pérez-Rontomé
C
et al
(1997)
Leaf morphology and leaf chemical composition in three Quercus (Fagaceae) species along a rainfall gradient in NE Spain
.
Trees

11
:
127
34
.
Cavender-Bares
J
Bazzaz
FA
(2000)
Changes in drought response strategies with ontogeny in Quercus rubra: implications for scaling from seedlings to mature trees
.
Oecologia

124
:
8
18
.
Chabot
BF
Hicks
DJ
(1982)
The ecology of leaf life spans
.
Ann Rev Ecol System

13
:
229
59
.
Chapmann
HD
Pratt
PF
(1973)
Methods of Analysis for Soils, Plants and Waters
.
Riverside, CA
:
University of California Press
.
Coleman
HD
Samuels
AL
Guy
RD
et al
(2008)
Perturbed lignification impacts tree growth in hybrid poplar—a function of sink strength, vascular integrity, and photosynthetic assimilation
.
Plant Physiol

148
:
1229
37
.
Damesin
C
Rambal
S
Joffre
R
(1997)
Between-tree variations in leaf δ13C of Quercus pubescens and Quercus ilex among Mediterranean habitats with different water availability
.
Oecologia

111
:
26
35
.
Dawson
TE
Pate
JS
(1996)
Seasonal water uptake and movement in root systems of Australian phraeatophytic plants of dimorphic root morphology: a stable isotope investigation
.
Oecologia

107
:
13
20
.
Donovan
LA
Maherali
H
Caruso
CM
et al
(2010)
The evolution of the worldwide leaf economics spectrum
.
Trends Ecol Evol

175
:
211
24
.
Drake
PL
Froend
RH
Franks
PJ
(2011)
Linking hydraulic conductivity and photosynthesis to water-source partitioning in trees versus seedlings
.
Tree Physiol

31
:
763
73
.
England
JR
Attiwill
PM
(2006)
Changes in leaf morphology and anatomy with tree age and height in the broadleaved evergreen species, Eucalyptus regnans F. Muell
.
Trees

20
:
79
90
.
Filella
I
Peñuelas
J
(2003)
Partitioning of water and nitrogen in co-occurring Mediterranean woody shrub species of different evolutionary history
.
Oecologia

137
:
51
61
.
Freschet
GT
Cornelissen
JHC
van Logtestijn
RSP
et al
(2010)
Evidence of the ‘plant economics spectrum’ in a subarctic flora
.
J Ecol

98
:
362
73
.
Goering
HK
Van Soest
PJ
(1970)
Forage Fibre Analysis (Apparatus, Reagents, Procedures and Some Applications)
.
Agricultural Handbook no. 379
, p.
1
20
.
Washington DC
:
ARS-USDA
.
Hanson
PJ
Samuelson
LJ
Wullschleger
SD
et al
(1994)
Seasonal patterns of light-saturated photosynthesis and leaf conductance for mature and seedling Quercus rubra L. foliage: differential sensitivity to ozone exposure
.
Tree Physiol

14
:
1351
66
.
Heberling
JM
Fridley
JD
(2012)
Biogeographic constraints on the world-wide leaf economics spectrum
.
Global Ecol Biogeog

21
:
1137
46
.
Ishida
A
Yazaki
K
Hoe
AL
(2005)
Ontogenetic transition of leaf physiology and anatomy from seedlings to mature trees of a rain forest pioneer tree, Macaranga gigantea
.
Tree Physiol

25
:
513
22
.
Juárez-López
FJ
Escudero
A
Mediavilla
S
(2008)
Ontogenetic changes in stomatal and biochemical limitations to photosynthesis of two co-occurring Mediterranean oaks differing in leaf life span
.
Tree Physiol

28
:
367
74
.
Kitin
P
Voelker
SL
Meinzer
FC
et al
(2010)
Tyloses and phenolic deposits in xylem vessels impede water transport in low-lignin transgenic poplars: a study by cryo-fluorescence microscopy
.
Plant Physiol

154
:
887
98
.
Kolb
TE
Stone
JE
(2000)
Differences in leaf gas exchange and water relations among species and tree sizes in an Arizona pine-oak forest
.
Tree Physiol

20
:
1
12
.
Levitt
J
(1980)
Responses of Plants to Environmental Stresses
.
New York
:
.
Lucas
PW
Turner
IM
Dominy
NJ
et al
(2000)
Mechanical defences to herbivory
.
Ann Bot

86
:
913
−–
20
.
Lyford
WH
(1980)
Development of the Root System of Northern Red Oak (Quercus rubra L.)
.
Petersham, MA
:
Harvard University, Harvard Forest, Harvard University Press
.
Mediavilla
S
Escudero
A
(2003a)
Mature trees versus seedlings: differences in leaf traits and gas exchange patterns in three co-occurring Mediterranean oaks
.
Ann For Sci

60
:
455
60
.
Mediavilla
S
Escudero
A
(2003b)
Relative growth rate of leaf biomass and leaf nitrogen content in several Mediterranean woody species
.
Plant Ecol

168
:
321
32
.
Mediavilla
S
Escudero
A
(2004)
Stomatal responses to drought of mature trees and seedlings of two co-occurring Mediterranean oaks
.
For Ecol Manag

187
:
281
94
.
Mediavilla
S
A
B
et al
(2008)
Testing the correlations between leaf life span and leaf structural reinforcement in 13 species of European Mediterranean woody plants
.
Funct Ecol

22
:
787
93
.
Mediavilla
S
Escudero
A
(2009)
Ontogenetic changes in leaf phenology of two co-occurring Mediterranean oaks differing in leaf life span
.
Ecol Res

24
:
1083
90
.
Ninyerola
M
Pons
X
Roure
JM
(2005)
Atlas Climático Digital de la Península Ibérica. Metodología y aplicaciones en bioclimatología y geobotánica
.
Bellaterra
:
.
Palow
DT
Nolting
K
Kitajima
K
(2012)
Functional trait divergence of juveniles and adults of nine Inga species with contrasting soil preference in a tropical rain forest
.
Funct Ecol

26
:
1144
52
.
Phillips
NG
Buckley
TN
Tissue
DT
(2008)
Capacity of old trees to respond to environmental change
.
J Integr Plant Biol

50
:
1355
64
.
Poorter
H
Niinemets
U
Poorter
L
et al
(2009)
Causes and consequences of variation in leaf mass per area (LMA): a meta-analysis
.
New Phytol

182
:
565
88
.
Reich
PB
Ellsworth
DS
Walters
MB
et al
(1999)
Generality of leaf trait relationships: a test across six biomes
.
Ecology

80
:
1955
69
.
Ryan
MG
Binkley
D
Fownes
JH
(1997)
Age-related decline in forest productivity: pattern and process
.

27
:
213
62
.
Seiwa
K
(1999)
Ontogenetic changes in leaf phenology of Ulmus davidiana var. japonica, a deciduous broad-leaved tree
.
Tree Physiol

19
:
793
7
.
Staudt
M
Joffre
R
Rambal
S
et al
(2001)
Effect of elevated CO2 on monoterpene emission of young Quercus ilex trees and its relation to structural and ecophysiological parameters
.
Tree Physiol

21
:
437
45
.
Takashima
T
Hikosaka
K
Hirose
T
(2004)
Photosynthesis or persistence: nitrogen allocation in leaves of evergreen and deciduous Quercus species
.
Plant Cell Environ

27
:
1047
54
.
Thomas
SC
(2010)
Photosynthetic capacity peaks at intermediate size in temperate deciduous trees
.
Tree Physiol

30
:
555
73
.
Turner
IM
(1994)
Sclerophylly: primarily protective?
Funct Ecol

8
:
669
75
.
van Kleunen
M
Fischer
M
(2005)
Constraints on the evolution of adaptive phenotypic plasticity in plants
.
New Phytol

166
:
49
60
.
Villar
R
Robleto
JR
De Jong
Y
et al
(2006)
Differences in construction costs and chemical composition between deciduous and evergreen woody species are small as compared to differences among families
.
Plant Cell Environ

29
:
1629
43
.
Voelker
SL
Lachenbruch
B
Meinzer
FC
et al
(2011)
Transgenic poplars with reduced lignin show impaired xylem conductivity, growth efficiency and survival
.
Plant Cell Environ

34
:
655
68
.
Walkley
A
Black
IA
(1934)
An examination of the Degtjareff method for determining soil organic matter and a proposed modification of the chromic acid titration method
.
Soil Sci

37
:
29
38
.
Weltzin
JF
McPherson
GR
(1997)
Spatial and temporal soil moisture resource partitioning by trees and grasses in a temperate savanna, Arizona, USA
.
Oecologia

112
:
154
64
.
Wright
IJ
Cannon
K
(2001)
Relationships between leaf lifespan and structural defences in a low nutrient, sclerophyll flora
.
Funct Ecol

15
:
351
59
.
Wright
IJ
Sutton-Grier
A
(2012)
Does the leaf economic spectrum hold within local species pools across varying environmental conditions?
Funct Ecol

26
:
1390
98
.
Wright
IJ
Westoby
M
Reich
PB
(2002)
Convergence towards higher leaf mass per area in dry and nutrient-poor habitats has different consequences for leaf life span
.
J Ecol

90
:
534
43
.
Wright
IJ
Reich
PB
Westoby
M
et al
(2004)
The worldwide leaf economics spectrum
.
Nature

428
:
821
7
.