SEARCH

SEARCH BY CITATION

Keywords:

  • dry matter content;
  • dry matter concentration;
  • mass concentration of dry matter;
  • specific leaf area;
  • tissue density

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  •  This study compared the predictive ability of dry matter content (DMC, dry mass per fresh mass) of leaves, stems, roots and entire plants in relation to dry matter concentration (D, dry mass per volume of plant organ). Data came from 28 species of field-collected plants (woody and herbaceous) and 17 species of herbaceous plants grown in hydroponic sand culture. Specific leaf areas were also measured.
  •  Dry matter content of the herbaceous plants grown in sand culture varied more between tissue types than did dry matter concentration but the correlation among plant parts was stronger when using DMC. Means and standard errors for DMC (g g−1) were 0.212 ± 0.009 (leaves), 0.176 ± 0.012 (support tissues) and 0.170 ± 0.021 (roots); for D (g cm−3) the values were 0.158 ± 0.010 (leaves), 0.168 ± 0.017 (support tissues) and 0.153 ± 0.013 (roots).
  •  Leaf DMC provided approximate estimates of leaf D (r = 0.76) for the field-collected plants but sclerophyllous leaves from shrubs restricted to acidic bogs proved to be outliers. The relationship between these two variables was stronger in the herbaceous species grown in sand culture, especially so for whole plant estimates (r = 0.91).
  •  Dry matter content and dry matter concentration were equally good predictors of specific leaf area.

Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Specific leaf area (SLA, projected leaf area per dry mass) has become an important variable in comparative plant ecology because it is associated with many critical aspects of plant growth and survival. For instance SLA is often positively correlated with seedling potential relative growth rate (Muller & Garnier, 1990; Poorter & Remkes, 1990) and leaf net photosynthetic rate (Field & Mooney, 1986; Reich et al., 1997; Shipley & Lechowicz, 2000); it is negatively correlated with leaf life span (Reich et al., 1992) and palatability to herbivores (Lucas & Pereira, 1990). Unfortunately, as pointed out by Wilson et al. (1999), this leaf attribute is largely restricted to species whose leaves have typical planar surfaces and therefore largely excludes species with needle-shaped leaves or (especially) leafless species such as cacti or many rushes. Furthermore, there are enough exceptions to the general empirical trends described above that increasing numbers of workers have begun to concentrate instead on dry matter concentration, also called the mass concentration of dry matter, which is one of the two underlying determinants of SLA. Because it is difficult to measure the volume of small irregularly shaped leaves, most of these researchers have not actually measured leaf dry matter concentration (D, leaf dry mass per volume of leaf). Rather, they measure the more easily measured ‘leaf dry matter content’ (LDMC, leaf dry mass per unit leaf fresh mass) because it is generally assumed that LDMC will be a good estimate of D. The objective of this paper is to determine the degree to which this assumption is reasonable.

Leaves are not the only part of the plant whose tissue density has received attention. Roderick and colleagues (Roderick et al., 1999a,b; Roderick, 2000; Roderick et al., 2000) have developed a series of functional relationships relating tissue density to many plant attributes of ecological importance and have pointed out the relative lack of information on comparative densities of different tissues across species. The second objective of this paper is to compare estimates of DMC and of D for leaves, support tissues and roots across a broad range of species.

It is necessary to compare some closely related terms. ‘Density’ is the mass of an object divided by its volume. The density of a plant organ is therefore its mass divided by its volume. The density of the dry matter of an organ is its (dry) mass divided by the volume occupied by its dry matter. The dry matter concentration of an organ (sometimes incorrectly called its dry matter density) is the mass of its dry matter divided by the volume of the organ itself. The dry matter content also called the mass fraction of dry matter in the SI system) is the ratio of dry mass to fresh mass of an organ. The dry matter concentration (D) and the dry matter content (DMC) of a plant or plant organ are therefore defined as:

  • image(Eqn 1)
  • image(Eqn 2)

(MDM, mass of tissue dry matter; MW, mass of water (which is equal to its volume since the density of water is 1 g cm−3); and V, plant or organ volume.) Noting that any mass (M), volume (V) and density (ρ) is related as

  • M = ρV(Eqn 3)

We can rearrange equations (1) (2) and (3) to deduce:

  • D =ρDMC(Eqn 4)

From these equations one can see that D and DMC would only be equivalent if the density of the plant or organ was 1 g cm−3. However, for the purposes of comparative ecology it is only necessary that the two be proportional and this would only be strictly true if leaf or organ density were constant across the species. The degree to which these last two attributes vary between species would determine the degree to which dry matter content is a useful imperfect measure of dry matter concentration.

Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We present data from both plants growing in the field and plants grown under controlled conditions in hydroponic sand culture. Taxonomy follows Marie-Victorin (1964). The field data set consisted of leaves from 28 species, of which 12 were herbaceous, eight were trees, five were forest understory shrubs and three were sclerophyllous shrubs limited to acidic bogs (Table 1). These leaves were collected in September, just before visible leaf senescence began. For each species, a section of stem containing a number of leaves was collected from 10 individuals, placed between wet paper on ice, and returned immediately to the lab. The stem was then placed in water in the dark at 5°C for approx. 12 h. This method was found to be best for measuring specific leaf area and leaf dry matter content in field collected samples (Garnier et al., 2001).

Table 1.  Species means of specific leaf area (SLA), leaf dry matter concentration (LD) and leaf dry matter content (LDMC) of 28 species of varying growth forms growing in the field
SpeciesTypeSLA (cm2 g−1)LD (g cm−3)LDMC (g g−1)
  1. Data come from a total of two leaves per individual and 10 individuals per species.

Acer negrundoTree152.720.350.39
Acer saccharumTree396.470.180.38
Aster lanceolatusGrassland herb205.070.150.28
Betula alleghaniensisTree455.820.180.26
Betula populifoliaTree158.620.290.4
Cassandra calyculataBog shrub100.260.190.5
Centaurea nigraGrassland herb182.240.130.24
Coptis groenlandicaUnderstory herb2060.160.36
Cornus stoloniferaUnderstory shrub195.120.240.35
Fragaria americanaGrassland herb339.780.130.31
Fraxinus americanaTree462.020.180.35
Geum macrophyllumGrassland herb408.690.120.26
Hieracium aurantiacumGrassland herb402.110.10.19
Impatiens pallidaForest herb807.950.070.22
Kalmia augustifoliaBog shrub92.790.280.48
Ledum groenlandicumBog shrub97.970.070.49
Lythrum salicariaGrassland herb172.910.170.32
Ostrya virginianaTree529.810.160.36
Polygonum sagittatumGrassland herb667.760.070.17
Populus tremuloidesTree132.180.420.44
Rhamnus fragulaForest shrub562.290.130.24
Rhus typhinaOpen shrub164.520.250.37
Rubus allegheniensisOpen shrub441.780.170.32
Rubus idaeusOpen shrub359.510.120.34
Rudbeckia hirtaGrassland herb170.270.120.23
Solidago canadensisGrassland herb137.170.20.35
Taraxacum officinaleGrassland herb504.020.080.16
Trifolium pratenseGrassland herb219.430.120.27

The second data set comes from 96 plants of 17 species of herbaceous plants grown from seed for 30 d under controlled conditions in a growth room (Table 2). Temperatures were 25°C during the day and 20°C during the night. Eight plants per species were planted but this number was reduced in some cases due to mortality. Approximately half of the plants of each species were harvested at day 15 and the others were harvested at day 30. Each plant grew in a 1.3-l container with washed silica sand and this sand was filled to field capacity with a modified Hoagland nutrient solution every 6 h during the entire experiment; see Shipley (2000) for more details of hydroponic system and nutrient concentrations. Plants were separated into leaf blades, structural tissues (stems and leaf petioles) and roots. In this data set tissues were measured immediately upon harvesting the plants rather than placing them in the dark at 5°C for 12 h.

Table 2.  Species means of specific leaf area (SLA), leaf, structural, and root dry matter concentration (D) and leaf, structural, and root dry matter content (DMC) of 96 plants from 17 herbaceous species grown in hydroponic sand culture
SpeciesSLA (g cm−2)Dry matter concentration (g cm−3)Dry matter content (g g−1)
LeafStructuralRootPlantLeafStructuralRootPlant
Achillea millefolium183.660.160.170.150.170.210.180.150.18
Acorus calamus433.970.100.080.080.080.130.100.110.11
Agrostis alba293.280.170.190.140.160.250.200.150.18
Bromus inermis240.550.200.190.210.190.260.210.170.19
Chrysanthemum leucanthemum219.690.170.150.200.200.230.200.430.22
Danthonia spicata227.100.250.140.070.130.420.300.300.32
Echinocloa crus-galli394.660.100.060.140.080.140.090.090.10
Epilobium glandulosum244.700.170.110.100.110.270.310.350.23
Galium palustre292.220.160.130.100.120.260.160.240.24
Melilotus alba164.910.170.260.090.130.230.250.070.15
Oenothera biennis192.000.140.100.070.100.180.130.070.13
Phleum pratense377.180.130.190.220.170.230.200.160.19
Plantago major180.340.190.100.090.140.190.140.080.14
Rumex longifolius330.330.090.070.150.090.110.090.080.09
Silene cucubalus178.440.160.180.250.170.170.130.480.20
Tanacetum vulgare223.840.120.170.160.130.220.210.180.20
Trifolium pratense218.010.140.120.180.120.210.160.160.15

For the field-collected plants, two leaves (laminas only) were then chosen and measured separately. For the controlled-growth plants all leaves per plant were taken and treated together. Leaf fresh weights, projected surface areas normal to the leaf surface and their volumes were measured. This was accomplished by submerging the leaf blade (or all leaves for the controlled-growth plants) in a small volume of deionized water within a thin rectangular plexiglas container. The change in volume of water (and thus the leaf volume) was estimated using a digital micrometer accurate to 10 µm, keeping the ratio of the surface area of the water to leaf volume sufficiently small that an appreciable increase of the height of the water was observed. Leaves were then dried at 70°C for at least 48 h, after which their dry weights were recorded. Roots and structural tissues were treated in the same way except that surface areas were not taken. Whole plant values were obtained by summing the relevant parts.

Preliminary tests using 12 pieces of plastic of different known volumes (VT, 0.07–2.6 cm3) were used to determine the accuracy of our method of estimating leaf volume (VE). The regression equation was VE = −0.008 + 1.028VT, r2 = 0.998; the residual standard error was 0.037, the intercept was not significantly different from zero and the slope was not significantly different from unity at the 5% significance level. Accurate volumes could not be measured for some tissues of the controlled-growth plants at the first harvest due to their small sizes.

All statistical tests were conducted using the SPLUS program (SPLUS, 1999).

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Table 3 presents the variance components (maximum likelihood estimates) of the measured variables obtained from the field-collected plants, based on a nested model. Since the vast majority of the variance was contributed by interspecific differences, we concentrate mostly on this level. Table 1 presents the means for each of the measured variables for all species of field-collected plants.

Table 3.  Variance components (%, restricted maximum likelihood) of specific leaf area (SLA, cm2 g−1), leaf dry mass (MD, g), Leaf hydrated mass (MH, g), leaf volume (VL, cm3), leaf dry matter concentration (LD, g cm−3) and leaf dry matter content (LDMC, g g−1) of 560 leaves from 280 individuals of 28 different species
LevelSLAMDMHVLLDLDMC
Species94.876.471.266.091.293.9
Plant 0.3 0.0 0.0 0.0 0.8 0.6
Leaf 4.823.628.834.0 8.0 5.5

Field-collected plants

The interspecific relationship between LDMC and leaf dry matter concentration (LD) is shown in Fig. 1 (left, closed circles and stars). The solid line indicates strict equality, that is, a leaf density of 1 g cm−3. The correlation between these two variables, using the 28 species means, is 0.61 (P < 0.001). One species (Ledum groenlandicum, shown by arrow) is an obvious outlier in this figure and if it is excluded the correlation increases to 0.76. In fact all three species typical of acidic bogs (stars) have leaf dry matter concentrations that are much lower than leaf dry matter contents. The regression equation is LD = −0.05 + 0.72LDMC.

image

Figure 1. Interspecific relationships between leaf dry matter concentration (LD), leaf dry matter content (LDMC) and specific leaf area (SLA) involving 17 species grown in hydroponic sand culture (open circles), 25 field-grown species of nonbog plants (trees, shrubs and herbs, closed circles) and three field-grown species of shrubs restricted to acidic bogs (stars). The arrow indicates one of the bog species (Ledum groenlandicum). The line in the left plot shows strict equality (1 : 1) between the variables. The lines in the central and right plots show the empirical regression equation of Garnier et al. (2001). Note that the central and right plots have logarithmic axes.

Download figure to PowerPoint

Fig. 1 (centre, right) shows the relationship between SLA and each of LDMC and LD. The observed correlations are 0.68 and 0.52, respectively; Ledum groenlandicum is again an obvious outlier in the SLA vs LD plot and if it is excluded the correlation increases to 0.69. The centre plot of Fig. 1 also shows the empirical regression equation of Garnier et al. (2001) which is Ln(SLA) = 11.64 – 1.58 Ln(LDMC). There are a group of points in the top right, mostly forest trees, for which the observed LDMC is higher than expected relative to SLA but the rest of the species appear to be consistent with the published regression equation. Interestingly, the empirical regression equation of Garnier et al. (2001) appears to describe the relationship better when using LD rather than LDMC.

Controlled-growth plants

The interspecific relationships between LDMC, LD and SLA for the species means of the controlled-growth plants are shown in Fig. 1 (open circles). The correlation between LD and LDMC was 0.86. Correlations between SLA vs LD and LDMC, using Ln-transformed data, were −0.58 and −0.34, respectively. Fig. 2 (left) shows the relationships between DMC and D for leaves, roots and support tissues, respectively, for each of the 96 plants. Fig. 2 (right) shows the relationship between DMC and D for entire plants. Correlations between DMC and D for leaves, stems, roots and entire plants were 0.66, 0.90, 0.71 and 0.91, respectively; all values were significant at P < 0.001. In order to detect any bias in using DMC to predict D we regressed DMC on D separately for the three tissue types and tested for significant deviations of the slopes from unity and of the intercepts from zero. Only for leaves did the slope (0.71 ± 0.088) differ from 1 (P = 0.001) and only for roots did the intercept differ from zero (0.04 ± 0.013).

image

Figure 2. Relationships between dry matter concentration and dry matter content of 96 plants from 17 species of herbs growing in hydroponic sand culture. Left: open circles indicate leaves; closed circles indicate roots and stars indicate structural tissues (stems, petioles). Right: data based on whole plants. Lines show strict equality (1 : 1) between the variables.

Download figure to PowerPoint

Dry matter content of the plants grown in sand culture varied more between tissue types than did dry matter concentration. Means and standard errors for dry matter content (g g−1) were 0.212 ± 0.009 (leaves), 0.176 ± 0.012 (support tissues) and 0.170 ± 0.021 (roots); for dry matter concentration (g cm−3) the values were 0.158 ± 0.010 (leaves), 0.168 ± 0.017 (support tissues) and 0.153 ± 0.013 (roots). Table 4 shows the Spearman rank correlations between dry matter contents, and between dry matter concentrations, among the tissue types. Correlations between plant parts were stronger when using dry matter content than when using dry matter concentration.

Table 4.  Spearman correlation coefficients between plant parts in relation to dry matter concentration (D, upper triangle matrix) and dry matter content (DMC, lower triangle matrix)
 LeafStructuralRootPlant
  1. Data are based on 96 plants from 17 herbaceous species grown in hydroponic sand culture.

Leaf1.0000.5270.1980.548
Structural0.8171.0000.5030.875
Root0.4650.5501.0000.674
Plant0.170.8480.7861.000

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Specific leaf area was equally strongly correlated with LDMC and with LD in the field data once one obvious outlier was excluded based on log–log plots. In the controlled-growth data LD was a better predictor of SLA but this may be because variation in LDMC was more constrained in this data set. In Garnier et al. (2001), based on field data, the correlation between SLA and LDMC was −0.96. As a simple predictor of SLA, there does not appear to be much of a statistical advantage of either LDMC or LD. Dry matter content has the practical advantage of being easier to measure accurately but would be more variable if the leaf was not rehydrated first. LDMC also appears to be a good, but approximate, measure of LD itself but there appear to be species for which the relationship breaks down. For instance, LDMC explains 58% of the variance in LD in the field data when Ledum groenlandicum was excluded but only 37% when it was included in the analysis. The correlation was stronger in the controlled-growth data. Part of the reason why LDMC overestimated LD to a greater extent in the field data (Fig. 1 right) may be because these leaves had already began to senesce and were thus less able to re-hydrate, although no visible signs of this were seen. For entire plants DMC was a very good predictor of D (Fig. 2 right).

From first principles one can deduce that the accuracy of approximation will decrease as the dry matter concentration of a given plant tissue deviates more from the general interspecific trend. The outliers in our field data were the three species of shrubs typical of acidic bogs, especially Ledum goenlandicum (Leatherleaf). As the common English name implies, this species is extremely sclerophyllous and it is presumably aberrant because of its unusually high concentration of very dense tissues and large intercellular spaces. The three bog species also have leaf morphologies (leaves with rolled edges, dense hairs) that could have introduced errors in the measurement of leaf volume since such attributes could trap air, over-estimating volume and underestimating LD.

We conclude that dry matter content is a reasonable indirect measure of dry matter concentration when used in broad interspecific comparisons, but a direct measure of D might be needed when working at an intraspecific level or when more precise estimates of this variable are required. However, it is more difficult to accurately measure volumes than dry masses, especially if the plant part is very small. For this reason, one might prefer dry matter content even at an interspecific level when dealing with small volumes.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

This research was financed by the Natural Sciences and Engineering Research Council of Canada. M. Roderick and an anonymous reviewer improved this paper with their suggestions.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Field C, Mooney HA. 1986. The photosynthesis-nitrogen relationship in wild plants. In: Givnish TJ, ed. On the economy of plant form and function. Cambridge, UK: Cambridge University Press, 2555.
  • Garnier E, Shipley B, Roumet C, Laurent G. 2001. A standardized protocol for the determination of specific leaf area and leaf dry matter content. Functional Ecology 15: 688695.
  • Lucas PW, Pereira B. 1990. Estimation of the fracture toughness of leaves. Functional Ecology 4: 819822.
  • Marie-Victorin F. 1964. Flore laurentienne. Montreal, Quebec: Les Presses de l’Université de Montreal.
  • Muller B, Garnier E. 1990. Components of relative growth rate and sensitivity to nitrogen availability in annual and perennial species of Bromus. Oecologia 84: 513518.
  • Poorter H, Remkes C. 1990. Leaf area ratio and net assimilation rate of 24 species differing in relative growth rate. Oecologia 83: 553559.
  • Reich PB, Walters MB, Ellsworth DS. 1992. Leaf life-span in relation to leaf, plant, and stand characteristics among diverse ecosystems. Ecological Monographs 62: 365392.
  • Reich PB, Walters MB, Ellsworth DS. 1997. From tropics to tundra: global convergence in plant functioning. Proceedings of the National Academy of Sciences, USA 94: 1373013734.
  • Roderick ML. 2000. On the measurement of growth with applications to the modelling and analysis of plant growth. Functional Ecology 14: 244251.
  • Roderick ML, Berry SL, Noble IR. 2000. A framework for understanding the relationship between environment and vegetation based on the surface area to Volume ratio of leaves. Functional Ecology 14: 423437.
  • Roderick ML, Berry SL, Noble IR, Farquhar GD. 1999a. A theoretical approach to linking the composition and morphology with the function of leaves. Functional Ecology 13: 683695.
  • Roderick ML, Berry SL, Saunders AR, Noble IR. 1999b. On the relationship between the composition, morphology and function of leaves. Functional Ecology 13: 696710.
  • Shipley B. 2000. Plasticity in relative growth rate and its components following a change in irradiance. Plant, Cell & Environment 23: 12071216.
  • Shipley B, Lechowicz MJ. 2000. The functional coordination of leaf morphology and gase exchange in 40 wetland plant species. Ecoscience 7: 183194.
  • SPLUS. 1999. S-PLUS 2000 User’s Guide. Seattle, WA, USA: MathSoft.
  • Wilson PJ, Thompson K, Hodgson JG. 1999. Specific leaf area and leaf dry matter content as alternative predictors of plant strategies. New Phytologist 143: 155162.