JXB Advance Access originally published online on August 23, 2005
Journal of Experimental Botany 2005 56(420):2637-2649; doi:10.1093/jxb/eri257
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
RESEARCH PAPER |
Storage versus substrate limitation to bole respiratory potential in two coniferous tree species of contrasting sapwood width
1Department of Forest Science, Oregon State University, Corvallis, OR 97331-5752, USA
2Department of Wood Science and Engineering, Oregon State University, Corvallis, OR 97331-5751, USA
* To whom correspondence should be addressed. Fax: +1 541 737 3385. E-mail: Michele.Pruyn{at}oregonstate.edu
Received 21 January 2005; Accepted 4 July 2005
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Two coniferous tree species of contrasting sapwood width (Pinus ponderosa L., ponderosa pine and Pseudotsuga menziesii Mirb., Douglas-fir) were compared to determine whether bole respiratory potential was correlated with available storage space in ray parenchyma cells and/or respiratory substrate concentration of tissues (total nitrogen content, N; and total non-structural carbohydrate content, TNC). An increment core-based, laboratory method under controlled temperature was used to measure tissue-level respiration (termed respiratory potential) from multiple positions in mature boles (>100-years-old). The most significant tissue-level differences that occurred were that N and TNC were two to six times higher for inner bark than sapwood, TNC was about two times higher in ponderosa pine than Douglas-fir and there was significant seasonal variation in TNC. Ray cell abundance was not correlated with sapwood respiratory potential, whereas N and TNC often were, implying that respiratory potential tended to be more limited by substrate than storage space. When scaled from cores to whole boles (excluding branches), potential net CO2 efflux correlated positively with live bole volume (inner bark plus sapwood), live bole ray volume, N mass, and TNC mass (adjusted R2
0.4). This relationship did not differ between species for N mass, but did for live bole volume, live bole ray volume, and TNC mass. Therefore, N mass appeared to be a good predictor of bole respiratory potential. The differences in net CO2 efflux between the species were largely explained by the species' relative amounts of whole-bole storage space or substrate mass. For example, ponderosa pine's inner bark was thinner than Douglas-fir's, which had the greater concentration of ray cells and TNC compared with the sapwood. This resulted in ponderosa pine boles having 30–60% less ray volume and 10–30% less TNC mass, and caused ponderosa pine net CO2 efflux/ray volume and net CO2 efflux/TNC mass to be 20–50% higher than Douglas-fir. In addition, because inner bark respiratory potential was 2–25 times higher than that of sapwood, ponderosa pine's thinner inner bark and deeper sapwood (relative to Douglas-fir) caused its bole net CO2 efflux/live bole volume to be 20–25% lower than that of similarly-sized Douglas-fir trees. Key words: Bole respiration, nitrogen, non-structural carbohydrates, ray parenchyma, sapwood width, xylem anatomy
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Autotrophic respiration is a key physiological process. Along with gross primary production, it determines the net input of carbon to plants (net primary production). The controls for autotrophic respiration are not completely understood. One possible respiratory determinant in trees is the amount and vitality of living wood in the bole (i.e. inner bark and sapwood). Tree boles with greater volumes of living tissue will possibly have more respiring parenchyma cells, providing more sites for storage of respiratory substrate (e.g. nitrogen, proteins, and total non-structural carbohydrates) and potentially greater respiratory losses from the main bole. However, the frequency of parenchyma cells alone does not indicate respiratory activity, because it also depends on their vitality and longevity, which may be related to their function (e.g. growth, storage, phloem loading, ion uptake/exchange, and wound response) and/or the concentration of available respiratory substrate.
Ponderosa pine (Pinus ponderosa L.) has wide sapwood (often about 15–20 cm in the radial direction with 50–200 annual rings), whereas Douglas-fir (Pseudotsuga menziesii Mirb.) has relatively narrow sapwood (often about 5 cm in the radial direction with 15–40 rings). Because ray parenchyma occupies 6.8% and 7.4% of the sapwood volume in ponderosa pine and Douglas-fir, respectively (Panshin and de Zeeuw, 1980
), ponderosa pine may have 2–4 times more sites for respiration and storage than a similarly-sized Douglas-fir tree. Thus, if sapwood amount correlates positively with respiratory loss, ponderosa pine's losses should be greater than those of Douglas-fir. Instead, the opposite was shown (Pruyn et al., 2003): in a species sample including Douglas-fir and ponderosa pine, sapwood respiratory potential (a woody tissue-based, laboratory estimate of respiration) was 1.3–3 times greater in species with small sapwood volumes than in those with large volumes for large, old-growth trees. Those results suggested that sapwood respiration was not limited by sapwood quantity alone, but by other factors, such as frequency, vitality, and content of the sapwood parenchyma.
Literature results on patterns of parenchyma size and abundance within conifer sapwood vary considerably, showing ray homogeneity in the sapwood; pith-to-bark increases in ray size, but decreases in ray frequency; or increases/decreases in only specific annual rings near the bark or pith (e.g. pith-to-bark trends in Douglas-fir, Gartner et al., 2000; and in other conifer species, Bannan, 1937, 1954; Gregory and Romberger, 1975; Lev-Yadun, 1998; tree base-to-tree top trends in conifers, Jaccard, 1915; Bannan, 1965; Gartner et al., 2000). Thus, the general impression of conifer sapwood is that storage sites are available throughout boles and not particularly concentrated to specific areas. Despite this apparent uniformity in sapwood storage site availability, ray metabolism varies with bole position. On a dry mass basis, respiratory potential of the outer sapwood of P. ponderosa was 30–60% higher than that of the middle or inner sapwood, depending on which heights within the tree and/or which tree ages were being compared (Pruyn et al., 2002a). In the same study, sapwood rings produced in the same calendar year released over 50% more CO2 in treetops than in tree bases. Similar within-tree patterns were found within boles of P. menziesii (Pruyn et al., 2002b). One possible explanation for these patterns is that ray cell vitality may vary within the bole. In fact, ray cell vitality (indicated by changes in nuclear morphology) has been shown to decline from outer to inner sapwood in various conifer species (Frey-Wyssling and Bosshard, 1959; Yang, 1993; Gartner et al., 2000). Ray vitality was also found to be highest at the base of the live crown and to decrease from there towards both the tree base and the tree top (Yang et al., 1994).
If parenchyma abundance is homogenous throughout boles, then differences in vitality among sapwood parenchyma locations suggest that bole respiration may instead be controlled by the availability of substrates for respiration. Woody tissue nitrogen and total non-structural carbohydrate (TNC) levels serve largely as a proxy for proteins, and for sugars and starches, respectively. Thus, both nitrogen and TNC may be used to predict local respiration in plant tissues. As a caveat, however, their abundance may not relate to respiration if they are stored and then used at other times or locations, or in different processes. Tissue-level, respiratory activity (O2 uptake) correlated positively with stem nitrogen concentration in Fraxinus nigra L. (Goodwin and Goddard, 1940) and with respiratory enzyme activity in Pinus radiata D. Don (Shain and Mackay, 1973). Studies on the organ-level have shown that leaves, stems, and root respiration are closely related to their Kjeldahl nitrogen contents (Ryan, 1991, 1995; Ryan et al., 1996; Pregitzer et al., 1998).
In contrast to nitrogen, there are few direct links between TNC concentrations and actual tree respiration in the literature. Reduced soluble carbohydrates from bark to pith were identified for Pinus sylvestris L. (Saranpää and Höll, 1989) and significant seasonal variation in the storage and mobilization of starch, sugars, protein, and fat occurred in the sapwood of Populusxcanadensis Moench ‘robusta’ (Sauter and van Cleve, 1993, 1994), but the possible correlation to respiration was not investigated. Strong correlations between respiration and carbohydrate concentration have been identified for growing tissue (Ryle et al., 1976; Penning de Vries et al., 1979), but not for mature tissues or organs, especially those involved in storage (Coggeshall and Hodges, 1980; Farrar, 1985): an observation that is consistent with theoretical predictions from respiration models (Dewar, 2000). More research is restricted to herbaceous species (e.g. Glycine spp., Lolium multiflorum, Zea spp., citations in previous sentence), woody shrubs (e.g. Larrea tridentata, Cunningham and Syvertsen, 1977) and leaves (Festuca arundinacea Schreb, Moser et al., 1982), suggesting a need for tissue-level studies on the correlation between respiration and TNC contents in tree boles, especially since tree bole tissues are devoted to multiple functions, e.g. translocation (phloem), growth (cambium), storage (sapwood), and development (transition from sapwood to heartwood).
The objectives of the current study were first to investigate the relationships between respiration and abundance of ray cells (as a proxy for sites and storage for respiration), nitrogen and/or TNC content (as a proxy for respiratory substrate) on the tissue level; and second, to scale-up to determine whether the tissue level trends persisted on the whole bole level. This research was undertaken in ponderosa pine and Douglas-fir to compare the results in coniferous species with differing quantities of live bole (inner bark plus sapwood) and to test the following two hypotheses. (i) Tissue-level respiration is not limited by physical storage capacity of the bole tissue (rays), but rather by enzyme (nitrogen) and metabolic substrate availability (total non-structural carbohydrates). (ii) At the whole-bole level, there is a strong positive correlation between respiratory demand (the total respiration expenditure for the tree's bole tissue, or whole-bole, potential net CO2 efflux) and ray and substrate abundance; inter-species differences in this relationship indicate different storage and metabolic strategies for the live bole.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Study areas and species characteristics
The ponderosa pine trees were located just east of the Cascade Range in central Oregon, near Gilchrist (43° 28' N, 121° 41' W, elevation 1355 m). Samples were collected from trees more than 240 years old in early September of 1999. The Douglas-fir trees were located just east of the Coast Range in southern Oregon, near Riddle (42° 57' N, 123° 22' W, elevation 215 m). Samples were collected from trees more than 100 years old in September of 1998. The size and age characteristics of the trees are presented in Table 1. Measurements were not taken during the growing season (March–September) to ensure that measured respiration rates represented only maintenance respiration and not growth respiration. Growth respiration is more likely than maintenance respiration to depend on hormonal or other stimuli and on carbohydrate supply from outside the immediate xylem stores (McCree, 1970
; Thornley, 1970). Ponderosa pine and Douglas-fir trees in the current study were selected from sites used in previous studies (Pruyn et al., 2002a, b), however, different trees were used and none of the previously published data appears here.
|
Cores were also collected once monthly at breast height from four mature trees of both species (100–250-years-old, with stem diameter-at-breast-height and height similar to the harvested individuals) from 1998–2000 at the same sites as the harvested trees. Respiratory potential was measured for each month, and a subsample of months (five) was chosen for chemical analysis. Months that were selected for the subsample were those with significantly higher respiratory potentials and the months before and after the higher-respiration month for comparison.
Tree felling and sampling
Twenty trees were selected randomly for each species that were free of broken tops, bole deformities, or visible disease. Six trees of each species (from the randomly selected 20) were chosen for study to avoid individuals that were clumped or would be hard to fell later. The diameter at 1 m was recorded (Table 1). After felling the ponderosa pine trees, 20-cm-tall discs were sawed from boles at node 240 years from the treetop (base), and just above (to avoid branch whorls within the crown) nodes 65 (lower), 50 (upper), and 15 (top). The same method was used for the Douglas-fir trees, except that bole discs were sampled from nodes 100 (base), 95 (lower), 35 (upper), and 15 (top). Tree height measurements were taken from tree base to base of the live crown (first bole position above ground level with three live branches), to treetops (Table 1), and to each sampled node. Discs for respiration measurements were transported to the laboratory wrapped in extra-strength black plastic bags with moist paper towelling inside to reduce desiccation, and then stored in these bags at 4 °C. Within 1 week after harvesting, three 12 mm diameter increment cores were extracted in the radial direction from each node, wrapped in plastic bags and returned to cold storage.
Tree leaf area above each sampled node was determined as described by Pruyn et al. (2002a). Each tree was divided into four sections. Twenty-five per cent (one of every four leaf bundles) of the leaves were hand-clipped (including attached woody material) from large crown sections (nodes 35–240), and 100% of the leaves from small sections (node 15). The fresh mass of the clipped material was then recorded. Fresh subsamples of the clipped material were weighed, dried, and sorted, leaves from woody twigs and branches, and dry mass for each was recorded. Ten needle fascicles from the clipped material were collected and stored at –10 °C in the laboratory, until leaf dimensions were measured to calculate leaf area. These leaves were then dried and the mass recorded. The proportion of leaf area to dry leaf mass to fresh leaf mass gave total leaf area for each tree.
Core segment preparation
Cores were divided into four segments: inner bark (phloem and cambium) and outer, middle, and inner sapwood. Any green (photosynthetic phelloderm) tissue that was visible just underneath the bark surface on the inner bark samples was removed with a razor blade. Sapwood was defined as the woody tissue extending from the outermost growth ring to the innermost growth ring before the heartwood. Transition zone rings (one or two lighter coloured rings at the sapwood/heartwood boundary) were not included in the samples. Outer, middle, and inner sapwood samples were obtained by dividing sapwood into three equal radial lengths. The exception was in cores extracted from ponderosa pine trees more than 240 years old, where sapwood was greater than 100 growth rings wide. In these cases, core segments of 10–15 growth rings in length were taken within the sapwood from the outer and inner sapwood boundaries and from the centre of the sapwood. Fresh mass and number of rings (to determine age) were recorded for each core segment. Core segment fresh volume was estimated as the mass (1 g H2O=1 cm3) of the water displaced when samples were submerged (D2395, ASTM 2001). Oven-dried masses (60 °C for 48 h) were recorded.
Ray size and distribution
One of the three oven-dried core segments from each sapwood radial and vertical bole position (used in the respiratory analysis) was selected randomly for microscopic analysis for six trees of each species. Inner bark core segments were not analysed because the tissue was needed for the chemical analyses. Tangential sections were made with a sliding microtome in the middle of the earlywood from two (younger tissues) or four growth rings throughout the core segment. Eight tangential sections were made per core segment (two or four per ring, depending on number of rings sampled), stained in safranin, and permanently mounted.
Average ray area (average of individual rays in the tangential view), ray frequency (the number of rays mm–2 tangential section) and ray proportion (proportion of sapwood tangential area occupied by rays=sum of all ray areas divided by the sum of the area viewed) were calculated. Assuming that the same ray proportion would be found at any nearby depth, ray volume within the bole was calculated by multiplying the ray proportion by the corresponding tissue volume. Area, frequency, and ray proportion were estimated using an image analysis system and following the protocol of Gartner et al. (2000). A compound microscope (Nikon Labophot-2, Nikon Microscopy, Melville, NY, USA) with a x10 objective lens and a colour video camera (Sony CCD/RGB, Sony Electronics Inc., Park Ridge NJ, USA) were used to view the slides. The microscopic image of the tangential section was projected onto a colour monitor (33 cm diagonal distance, Sony Trinitron, Sony Electronics Inc., Park Ridge NJ, USA) and onto a computer monitor (Apple Macintosh Quadra 800) by way of a digitizing card. One or two fields of view (1.00x0.80 mm each) were randomly selected from each tangential section made, for a total of eight fields per core segment. All rays within each field of view were highlighted using the paint tool in NIH Image v. 1.60 (Rasband, 1996). Entire rays (both simple and fusiform) were highlighted, including cell walls, ray tracheids, and for fusiform rays, epithelial cells. The resin canal itself was excluded from fusiform rays.
Chemical analyses
Oven-dried core segments were ground to pass a 20-mesh screen and then analysed for nitrogen using a LECO CHN-1000 Analyser (LECO, 3000 Lakeview Ave., St Joseph, MN 49085-2396, USA) that measured tissue total carbon content (%) and total nitrogen content (%) via combustion and subsequent gas analysis. Total non-structural carbohydrate content (%) of the ground tissue was determined using a sulphuric acid solution to extract and hydrolyse carbohydrates to reducing sugars, which included the simple sugars, dextrin, starch, and fructusan, but not the structural carbohydrates such as hemicellulose and cellulose (AOAC, 1965). The Nutritional Analysis Laboratory in the Rangeland Ecosystem Science Department at Colorado State University conducted all chemical analyses (Natural Resource Bldg. Rm. 240, Fort Collins, CO 80523-1478, USA).
Tissue chemical contents were compared within and between the ponderosa pine and Douglas-fir trees. The analysed tissue was from a combination of two or three replicates of the oven-dried, core segments from the respiratory analyses. Inner bark, outer sapwood, and inner sapwood core segments were analysed from three bole heights (base, lower or upper, and top) of the ponderosa pine trees and from two heights (base and top) of mature Douglas-fir. To calculate the moles of nitrogen in each sample, core segment dry mass was multiplied by its nitrogen percentage and divided by nitrogen's atomic weight.
Respiratory measurements
Prior to anatomical or chemical analysis, core segment respiration was measured using a previously developed protocol (Pruyn et al., 2002a, b). After core extraction and segment preparation, core segments were wrapped in plastic and stored at 4 °C until the night prior to analysis. Segments were then transferred to a 25 °C incubator and stored overnight to allow metabolic activity in core segments to stabilize (Goodwin and Goddard, 1940; Levy et al., 1999). Immediately prior to measurement, core segments were reweighed and placed in vials, which were then sealed with gas-tight rubber septa. To determine a rate of CO2 production, the CO2 concentration within vials was measured with a Hewlett-Packard (5700A) gas chromatograph (GC, Hewlett Packard, Rt. 41, Avondale, PA 19311, USA) immediately after closing the vials and again after an incubation period. Core segments were incubated at 25 °C for 6 h between GC measurements. Details of GC analysis and calculation of respiratory potential (nmoles CO2 s–1, on a core segment volume, core segment dry mass, or moles nitrogen basis) in Pruyn et al. (2002a, b). Immediately following the GC analysis, core segments were weighed a third time. Changes in the three successive wet masses verified that water loss was low (1–3%) between sampling and measurement. Respiratory potential on each basis (mass, volume, ray volume, and moles nitrogen) was plotted against growth ring number from bark. The reported values are referred to as respiratory potential, rather than respiration rate, because measurement conditions of these excised samples were probably different from those within the tree.
Scaling core segment measurements to the whole-bole level
To understand whole-bole level (excluding branches) trends and the effect of the metric used to scale-up, net CO2 efflux was scaled to whole boles in the September-sampled ponderosa pine and Douglas-fir trees. Core segment respiratory potential was scaled to boles using four cylinders (one for each vertical position sampled) to represent bole segment volumes along the bole heights, as described previously in Pruyn et al. (2002a). In addition to scaling to whole-bole live volume (inner bark plus sapwood), respiratory potential was also scaled to whole-bole ray volume, nitrogen mass, and total non-structural carbohydrate mass. Ray volume in boles was calculated by multiplying the ray proportion (measured from core segments) to the corresponding volume of tissue at each radial and vertical bole position. Because the ray proportion in the inner bark was not measured, an estimate of 92% was used to represent the living cells of the inner bark for both species. This estimate was based on the knowledge that non-living elements of inner bark are primarily sclereids (Mauseth, 1988), which in Douglas-fir inner bark approximated 7.5% (Ross and Krahmer, 1971). Total nitrogen and non-structural carbohydrate contents of whole boles were determined by multiplying each radial position's volume at each bole vertical position by its density (oven-dry mass/green volume, determined from core segments) to obtain the oven-dry mass. The proportions of total nitrogen and non-structural carbohydrate from each radial position were multiplied (separately) by the corresponding mass for each bole radial and vertical position. Each radial position's volume or mass for each bole height was then multiplied by the respective respiratory potential on a core segment volume or mass basis to calculate its net CO2 production. Whole-bole net CO2 efflux was calculated by summing the volumetric rates of each radial and vertical bole position.
Statistical analyses
All data were analysed in Statistical Analysis Systems software, release 8.0 (SAS Institute Inc., 1998). The Shapiro–Wilk W-test was used to determine whether the response variables were distributed normally. A transformation (square-root or logarithmic) was performed when necessary to meet assumptions of normality and constant variance. Least squares means (LSMEANS), generated from the various SAS procedures described below, are reported ±pooled SE, or confidence interval (CI) for transformed variables. Within a specific table or figure, if the CI was required for one variable, it was presented for all. To compare ray anatomical characteristics, tissue chemical contents, and respiratory potential between ponderosa pine and Douglas-fir trees, least squares means (LSMEANS) were generated using PROC MIXED, with a randomized block design and strip-plot (split-block) treatments (Little and Hills, 1978; Milliken and Johnson, 1984). To account for correlation within trees, individual trees were assigned as blocks, with random effects of block and the interactions among block, bole radial and vertical positions. For core anatomy and tissue chemical content, effects of bole radial position and vertical position, and their interaction were tested separately for each species. Significant differences (P <0.05) were identified via pair-wise comparisons (t-tests) among bole positions and between species using Fisher's Protected Least Significant Difference (FPLSD) procedure (Fisher, 1966). A comparison of regression lines was used with the random effects accounting for within-bole variation to compare the relationship between average ray area and ray frequency between species. To compare respiratory potential and tissue chemical contents among sampling dates, a repeated measures analysis was used to test each variable separately for each species, with a subject of tree by radial position and an unstructured covariance matrix.
For core segment respiratory potential per unit volume, dry mass, ray volume or nitrogen, the effects of tissue age (growth ring number from bark), species and their interaction were tested using a comparison of regression lines, including random effects to account for within-bole variation. The comparison of regression lines was also used to understand the relationship of core respiratory potential on a mass basis to tissue anatomy or chemical content for each species. These regressions were performed for the inner bark and sapwood combined, and for the sapwood alone. The effects of ray proportion, total nitrogen content, or total non-structural carbohydrate content were tested separately, along with the interaction between species and ray proportion or tissue chemical content. Parameter estimates for these regressions should be interpreted with care because of the lack of independence between inner bark, outer sapwood, and inner sapwood. Although their correlation was accounted for in the regression model, the specified correlation matrix may not entirely reflect the natural variation within trees.
To compare whole-bole level trends in net CO2 efflux, a comparison of regression lines was used to test the significance of species, scaling index (i.e. live bole volume, ray volume, nitrogen mass, and total non-structural carbohydrate mass), and their interaction. Standard errors were not scaled up with the means from core to whole-bole level for two reasons. (i) Because the ray anatomy and tissue chemical content analysis were only conducted on one core per radial and vertical position, there were no estimates for within tree variation for these explanatory variables. (ii) The error in the response variable was incorporated into the residual error and thus the assumption was that the error was approximately the same for all of the trees. The whole-bole values presented are merely estimates of in situ activity, and the variance among the sample size of 12 trees should be interpreted as an approximation for the confines of whole-bole net CO2 efflux.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Variation in ray anatomy
The most notable significant differences in average ray area (µm2, average of individual rays in the tangential view), ray frequency (number mm–2), and ray proportion (% of the sapwood volume occupied by rays) occurred with respect to bole vertical position at a specific radial position (Fig. 1). For example, in Douglas-fir trees, ray frequency was higher yet average ray area was smaller in the inner sapwood near treetops than lower in the bole (significant effects of bole vertical position and the interaction of bole radial and vertical positions, P <0.01 from strip-plot in PROC MIXED and FPLSD, Fig. 1a, b). Ponderosa pine sapwood ray frequency was higher near treetops than in the middle and inner sapwood of the lower bole positions (Fig. 1b), and ray area was smaller near treetops than at the base, but the trend was not as strong as in Douglas-fir (Fig. 1a).
|
There was no difference between either the frequency or ray proportion near the tree base or treetops between ponderosa pine and Douglas-fir (P >0.07, strip-plot in PROC MIXED and FPLSD). Also, the interaction of bole radial position by bole vertical position by species was significant (P <0.03) for the three ray characteristics (i.e. area, frequency, and proportion), indicating that the within-bole variation in ray characteristics differed between species. When the regression of average ray area versus ray frequency was compared between the two species, average ray area was
20% higher in Douglas-fir than in ponderosa pine for each level of ray frequency (Fig. 1c, P <0.02). This regression confirmed the observation that when sapwood rays are more frequent, they will be smaller, a trend that did not differ between species (insignificant interaction between species and ray frequency, P=0.9).
Variation in tissue chemistry
Total carbon content was fairly constant within boles, between species, and among seasons, ranging from 47–52% (data not shown). Significant differences that did occur were often small. Larger differences occurred between the inner bark and sapwood, with the former being 4–8% higher than the latter for most trees. In contrast to tissue carbon content, differences were larger and more frequent for tissue nitrogen and TNC. Both ponderosa pine and Douglas-fir had 1.5–6 times higher nitrogen and TNC contents in the inner bark than in the sapwood (P <0.05; Table 2). At any given vertical position, ponderosa pine sapwood nitrogen and TNC contents were constant with respect to growth ring number. The exception was in September-sampled trees at the lower bole position, which had 20% higher TNC in the inner versus outer sapwood (P <0.05). In Douglas-fir, sapwood nitrogen content near the tree base and treetop was constant from bark to sapwood/heartwood boundary, whereas TNC was significantly higher (14–18%, P <0.05, Table 2) in the inner than the outer sapwood.
|
There were also significant, yet inconsistent, trends for tissue nitrogen and TNC contents from tree base to treetop, between species and among sampling dates. For the inner bark, ponderosa pine nitrogen and TNC contents decreased by 13–26% from tree base to treetop (Table 3; P <0.05). For sapwood of ponderosa pine and inner bark and sapwood of Douglas-fir, nitrogen and TNC contents significantly increased from tree base to treetop by 20–50% (Table 2). In September, mature ponderosa pine inner bark was higher than mature Douglas-fir in nitrogen (27%) and TNC (1.2–2.2 times) at the tree base, whereas the reverse was true near treetops (Table 2). Sapwood differences between the two species were not evident for nitrogen and small for TNC (if present, 10–25% lower in ponderosa pine). The effect of species alone was not significant for nitrogen or TNC contents (P >0.1), whereas the interaction of species by bole radial and vertical positions was (P <0.0001), indicating that the within-tree trends differed between the two species. Although statistical comparisons were not made between the species for all dates sampled (due to small sample size with respect to number of sampling dates), nitrogen (inner bark) and TNC (inner bark and sapwood) were frequently higher in ponderosa pine than in Douglas-fir (Table 3).
|
Respiratory potential comparisons between the two species
For the September harvests of mature ponderosa pine and Douglas-fir, core segment respiratory potential (on all indices) varied significantly by tissue age (P <0.0001, comparison of regression lines), decreasing from inner bark, throughout the sapwood to the heartwood/sapwood boundary (Fig. 2). For both volume-based indices (core segment and ray volume), the interaction of species by tissue age was significant (P <0.02), although the main effect of species was not significant (P >0.6, comparison of regression lines). This suggested that the relationship between volume-based respiratory potential and tissue age differed for each species (Fig. 2a, c). For example, inner bark volume-based respiratory potential of ponderosa pine was 1.5–3 times higher than Douglas-fir at all bole vertical positions except near treetops, where ponderosa pine inner bark respiratory potential was 30% lower than that of Douglas-fir (Fig. 2a, inset). In contrast to the inner bark, there were few significant differences between the two species in sapwood volume-based respiratory potential (Fig. 2a, c).
|
On a core segment mass basis, ponderosa pine respiratory potential was 1.5–4 times higher (species effect, P=0.0001, strip-plot in PROC MIXED and FPLSD) than Douglas-fir at all bole radial and vertical positions, whereas the relationship between respiratory potential and age did not differ between the two species (P=0.8). When respiratory potential was considered on a core segment nitrogen basis, there was no difference between the two species (P >0.1, Fig. 2d) and respiratory potential varied significantly only by age (P <0.02).
Relation of core segment respiratory potential to ray anatomy and tissue chemical content
The higher concentrations of TNC during the winter months (December, January, February) often corresponded to higher respiratory potentials in ponderosa pine (Table 3). The significant variation of Douglas-fir nitrogen and TNC among sampling dates did not relate consistently to the significant differences of respiratory potential. However, there were isolated examples of positive correlation: TNC increased with respiratory potential from August to September in 1998 at all radial positions, and higher TNC in the outer and inner sapwood in September 1999 may have resulted in a delayed effect of increased respiratory potential from September to October in 1999.
There was a significant, positive relationship between respiratory potential on a mass basis and core segment total nitrogen and TNC contents (Fig. 3b, c, P=0.0001, strip-plot in PROC MIXED and FPLSD), whereas there was no effect of ray proportion (Fig. 3a, P=0.3). The effect of species was significant for both total nitrogen and TNC contents (P <0.0001), whereas the interaction between species and tissue chemical content was significant only for TNC (P <0.02), indicating that its relationship to mass-based respiratory potential differed between the two species. The interpretation of the log–log relationship between core segment respiratory potential and tissue chemical content was that for every doubling of nitrogen content, core respiratory potential increased 4.2-fold (95% CI, 2.1–5.2). Moreover, for each level of nitrogen content, core respiratory potential was 1.9 times larger for ponderosa pine than for Douglas-fir (Fig. 3b). Each doubling of TNC resulted in core respiratory potential increasing by a factor of 2.2 (95% CI, 2.0–2.4) for ponderosa pine and by 2.5 (95% CI, 2.2–2.9) for Douglas-fir. For a given TNC content in the range presented (Fig. 3c), ponderosa pine core respiratory potential was estimated to be 1.2–1.9 times larger than that for Douglas-fir.
|
Scaling to the whole-bole level
Whole-bole level potential net CO2 efflux correlated positively with live bole volume, ray parenchyma volume in the bole, live bole mass of nitrogen, and live bole mass of TNC (Fig. 4, comparison of regression lines). There was no significant difference in whole-bole net CO2 efflux/nitrogen mass between ponderosa pine and Douglas-fir (Fig. 4c). By contrast, whole-bole net CO2 efflux scaled from the three other indices (live bole volume, ray parenchyma volume, and TNC mass) showed significant species effects (P <0.04), but no significant interaction between species and the explanatory variable (P >0.1) (Fig. 4a, b, d). Ponderosa pine whole-bole net CO2 efflux (mmol s–1) was 0.5 (95% CI, 0.3–0.7) or 0.2 (95% CI, 0.1–0.4) units higher than Douglas-fir for each unit increase in ray volume (m3) or TNC mass (kg), respectively (Fig. 4b, d), and 0.4 (95% CI, 0.02–0.7) units lower than Douglas-fir for each unit increase in live bole volume (m3, Fig. 4a).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Limitations to tissue-level respiration
As hypothesized, it appeared that tissue-level respiration was not limited by physical storage capacity of the bole tissue (abundance of rays). The ray parenchyma anatomical characteristics in the mature ponderosa pine and Douglas-fir trees were generally constant with respect to bole radial (growth ring) and vertical (node) position, and therefore did not reflect the within-bole respiratory gradients from the current or previous studies (Pruyn et al., 2002a
, b). A notable exception was the high frequency of rays at node 15 compared with the lower nodes at all radial positions in mature ponderosa pine and at the inner radial positions of mature Douglas-fir, which corresponded to sapwood respiratory potentials that were 1.3–2 times higher near the treetop compared with the base (current data not shown and Pruyn et al., 2002a, b). The decrease in respiratory potential on a ray volume basis from outer to inner sapwood with relatively constant ray volume in both species suggested a bark-to-pith decrease in ray vitality, a trend that has been reported previously (Goodwin and Goddard, 1940; Frey-Wyssling and Bosshard, 1959; Shain and MacKay, 1973; Yang, 1993; Gartner et al., 2000). Decreased ray vitality could result from lower numbers of living rays, or decreased vitality in all rays with increasing proximity to the sapwood/heartwood boundary. To resolve this uncertainty, vital staining techniques should be explored (e.g. triphenyl tetrazolium tri-chloride, Ryan, 1990; or Cooamassie Blue, Stockfors and Linder, 1998).
In contrast to the lack of correlation between bole storage capacity and respiratory potential, both enzyme (nitrogen, N) and metabolic substrate availability (total non-structural carbohydrates, TNC) had significant, positive relationships with volume-based or mass-based, core segment respiratory potential, which supported the original hypothesis that substrate was more important than parenchyma volume for tissue-level respiration in both of these species. The strongest example of this tendency was in the inner bark relative to the sapwood, where much higher respiration corresponded with much higher nitrogen and TNC levels, which may be explained by the cambium of the inner bark being involved in growth and thus more substrate-dependent than the storage-oriented sapwood (Dewar, 2000). Tree base to treetop trends in bole tissue chemical content in the current study also corresponded positively with core respiratory potential, for example, decreasing from tree base to treetop in ponderosa pine inner bark and the reverse for sapwood of both species (current data not shown: Pruyn et al., 2002a, b). Further, the often-higher levels of nitrogen and TNC in ponderosa pine boles corresponded with higher respiratory potentials on a mass or volume basis, as compared to Douglas-fir, yet when respiratory potential was represented on a nitrogen basis, the species no longer differed. These results indicated that enzyme (nitrogen), or sugar/starch (TNC) concentrations in woody tissue may limit respiration.
The current study's log–log relationship between core segment respiratory potential and core nitrogen content in ponderosa pine contrasted the linear relationship previously estimated for Pinus spp. with an infrared gas analysis system, in that a doubling of tissue N resulted in a 1.3–2-fold increase (as compared to the current, 4.2-fold increase) in bole respiration (for N contents 0.02–0.12 mol kg–1, Vose and Ryan, 2002; Maier et al., 1998). However, the current study's regression for sapwood alone was more comparable: respiratory potential was 1.6 times higher with each doubling of core N, which suggested a different relationship between respiratory potential and nitrogen in the inner bark, possibly related to its growth and transport functions. Further, respiratory potential per unit N in the sapwood of both species (0.5–6 µmol mol–1 N s–1, corrected to 15 °C, assuming a Q10 of 2.0) was higher than values reported for Pinus taeda bole sapwood and was closer to values of coarse roots (1.2 versus 3.4 µmol mol–1 N s–1 at 15 °C, respectively; Vose and Ryan, 2002).
Despite the positive correlation between nitrogen and TNC contents and respiratory potential in Douglas-fir and ponderosa pine boles, this relationship was not identifiable or consistent for all comparisons. Further, the radial constancy of nitrogen and TNC contents with respect to sapwood growth ring number in the current study did not reflect radial trends in respiratory potential (current data and Pruyn et al., 2002a, b) and contrasted results from Pinus sylvestris L. (Saranpää and Höll, 1989; Fischer and Höll, 1992), Quercus petraea (Matt) Liebl. and Fagus sylvatica L. (Barbaroux and Bréda, 2002), where sugars and starches were more highly concentrated in the outer versus inner growth rings. One explanation for the current lack of a sugar/starch gradient across the sapwood is that the previous studies divided sapwood into five to six zones and analysed sugars and starches separately. By contrast, the current study analysed sugars and starches combined (TNC content) in only two zones, potentially causing subtle differences in specific rings to go unnoticed. For example, summing the sugar and starch values from Saranpää and Höll (1989) for each zone and then averaging the zones into outer and inner sapwood reduced the sugar/starch difference between outer and inner sapwood from 70–100% to 20–60%.
Interpreting the respiratory significance of tissue chemical content is problematic because sites of high respiration and of high carbohydrate content may be quite different (Farrar, 1985). For example, it has been shown that the concentration of the unloaded sucrose outside the phloem (i.e. in the differentiating xylem) of P. sylvestris was not related to annual growth ring width. This suggested that xylem production (a respiration-dependent process) along the stem was not limited by substrate unloading or local availability, but rather by greater specific mass transfer of sucrose (higher phloem and cambium activity) (Sundberg et al., 1993). Such observations have led researchers to conclude that only a better understanding of the sugar fluxes, between tissues and between subcellular compartments, will reveal the true nature of the relationship between carbohydrate status and respiration rate (Farrar, 1985; Cannell and Thornley, 2000; Dewar, 2000).
Whole-bole CO2 efflux, parenchyma, N, and TNC
On the whole-bole level, there was a significant correlation between respiratory demand (whole-bole, potential net CO2 efflux) and ray and substrate abundance. This correlation was shown by the positive, linear relationships between whole-bole net CO2 efflux and live bole volume, the ray volume in the live bole, the live bole mass of N, and the live bole mass of TNC, all with an adjusted R2 of 0.41–0.77. This relationship did not differ between species for live bole N mass, implying that it was a good estimator of CO2 efflux. The relationship between net CO2 efflux and the other scaling factors (live bole volume, ray volume in the live bole, and live bole mass of TNC) were similar between the two species in that they both shared the same slope for each factor. The difference was that ponderosa pine was either 20–50% higher (ray volume in the live bole and live bole mass of TNC) or 20–25% lower (live bole volume) than Douglas-fir. Thus, inconsistent, weak and/or complex correlations on the tissue-level scaled up to reveal important trends on the whole-bole level.
The differences in whole-bole net CO2 efflux between the species were largely explained by the species' relative amounts of whole bole storage space or substrate mass. The higher ray proportions and TNC mass of the inner bark (relative to the sapwood) became important on the whole-bole level because ponderosa pine's inner bark volume was only 25% that of Douglas-fir. Scaling the high ray proportions and high TNC mass to ponderosa pine's smaller inner bark resulted in a smaller total ray volume and total TNC mass for ponderosa pine boles relative to Douglas-fir, which caused ponderosa pine's whole-bole net CO2 efflux (per smaller ray volume or smaller TNC mass) to be higher than that of Douglas-fir. Likewise, the lower respiratory potential of the sapwood (relative to inner bark) strongly influenced ponderosa pine's whole-bole net CO2 efflux/live bole volume because it was scaled to ponderosa pine's larger, live bole volume (twice that of Douglas-fir).
Because the influence of ray volume and substrate mass on whole-bole net CO2 efflux was conveyed via the inner bark/sapwood volume, these two species likely had contrasting strategies for storage and metabolism in the live bole. Ponderosa pine's higher respiratory demand (whole-bole net CO2 efflux) per unit storage space (ray volume) and sugar/starch concentration (TNC content) was possibly attributable to higher maintenance requirements of its proportionally larger sapwood compared with Douglas-fir. Further, for each level of whole-bole net CO2 efflux, the smaller sapwood of Douglas-fir contained more storage space and more sugar/starch, which possibly resulted from its higher leaf area to sapwood area ratio (i.e. increased photosynthetic supply per unit sapwood, relative to ponderosa pine; Table 1). By contrast, nitrogen mass in the bole predicted whole-bole net CO2 efflux independently of species, supporting the previously suggested hypothesis that maintenance respiration is directly or indirectly related to tissue protein content (Cannell and Thornley, 2000). The possibly high cost of sapwood maintenance for ponderosa pine was not evident for net CO2 efflux per unit live bole volume, which was lower than that of Douglas-fir, resulting in the species with a proportionally larger live bole volume (ponderosa pine) having a lower live bole respiratory demand, a trend that has been shown for other conifer species (Pruyn et al., 2003). This result suggested that ponderosa pine's higher respiratory demand per unit storage space (living ray cells) and sugar/starch concentration was balanced out over its larger sapwood, possibly in support of non-living tracheids (e.g. ion uptake, transport, and gradient control) or other sapwood functions.
In the current study, the diverse tissues of the tree boles were analysed separately first to build an understanding toward the mechanistic controls of respiration and it was learnt that core-based respiratory potential may be limited by substrate concentration, but that the correlation was strongly driven by the high concentration of N and TNC in the inner bark relative to the sapwood. When comparing the relationship between core-based respiratory potential and core N content to similar estimates from chamber-based, organ level measurements (via infrared gas analysis), similarities were found in the sapwood and also differences, possibly from the influence of the inner bark measurements. Further research is necessary to determine the relationship between core-based respiratory potential, live ray cells, and chamber-based, organ level measurements, which would clarify the proportion of the live bole that is metabolically active and help bridge the gap between tissue and whole-tree level respiration. Scaling from cores to whole boles indicated the importance of live bole geometry (inner bark and sapwood volume) to the outcome of net respiratory (CO2) efflux, suggesting a need for more tissue-based measurements and scaling estimates to reveal/include the influence of inner bark and sapwood in branches, boles, and roots. However, before whole-tree respiration can be explained in terms of coarse, fine, and substrate-limited control of the respiratory pathway, analytical and integrated studies are required on carbon flux to and partitioning within respiratory metabolism and the related, energy-dependent processes (Farrar, 1985).
|
Acknowledgements |
|---|
We thank Crown Pacific Limited Partnership, Gilchrist Forestry Division, OR for providing the ponderosa pine bole wood sample trees and Roseburg Forest Products in Riddle, OR and the OSU Research Forests for providing the Douglas-fir. Also, we thank Christine Groll and Jonathan Fung for preparing hundreds of tangential wood sections, and Jennifer Secord and Alec Mesdag for measuring the anatomical characteristics of many of those sections. We thank Barbara Oskroba from the Nutritional Analysis Laboratory at Colorado State University for providing useful background information about the chemical analyses and promptly processing our samples. We appreciate the comments of two reviewers. This research was funded by USDA CSREES 96-35103-3832 and 97-35103-5052, and NSF DEB 9632921, and a special grant from the USDA to Oregon State University for wood utilization research.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
AOAC (Association of Official Agricultural Chemists). 1965. Official methods of analysis of the Association of Official Agricultural Chemists/Committee on editing methods of analysis, 10th edn. Washington, DC: The Association, 498–499.
ASTM. 2001. Standard test methods for specific gravity of wood and wood-based materials, Method B: Volume by water immersion (D2395). Annual book of ASTM standards, Section 4, Construction, v. 4.10, Wood. Conshohocken, PA: ASTM, 360–361.
Bannan MW. 1937. Observations on the distribution of xylem-ray tissue in conifers. Annals of Botany 1, 717–726.
Bannan MW. 1954. Ring width, tracheid size, and ray volume in stem wood of Thuja occidentalis L. Canadian Journal of Botany 32, 466–479.[CrossRef][Web of Science]
Bannan MW. 1965. Ray contacts and rate of anticlinal division in fusiform cambial cells of some Pinaceae. Canadian Journal of Botany 43, 487–507.[CrossRef]
Barbaroux C, Bréda N. 2002. Contrasting distribution and seasonal dynamics of carbohydrate reserves in stem wood of adult ring-porous sessile oak and diffuse-porous beech trees. Tree Physiology 22, 1201–1210.
Cannell MGR, Thornley JHM. 2000. Modelling the components of plant respiration: some guiding principles. Annals of Botany 85, 45–54.
Coggeshall BM, Hodges HF. 1980. the effect of carbohydrate concentration on the respiration rate of soybean. Crop Science 20, 86–90.
Cunningham GL, Syvertsen JP. 1977. The effect of non-structural carbohydrate levels on dark CO2 release in creosotebush. Photosynthetica 11, 291–295.[Web of Science]
Dewar RC. 2000. A model of the coupling between respiration, active processes and passive transport. Annals of Botany 86, 279–286.
Farrar JF. 1985. The respiratory source of CO2. Plant, Cell and Environment 8, 427–438.[CrossRef]
Fischer C, Höll W. 1992. Food reserves of Scots pine (Pinus sylvestris L.). II. Seasonal changes and radial distribution of carbohydrate and fat reserves in pine wood. Trees 6, 147–155.
Fisher RA. 1966. The design of experiments, 8th edn. New York: Haffner Publishing Co., 184–213.
Frey-Wyssling A, Bosshard HH. 1959. Cytology of the ray cells in sapwood and heartwood. Holzforshung 13, 129–137.
Gartner BL, Baker DC, Spicer R. 2000. Distribution and vitality of xylem rays in relation to tree leaf area in Douglas-fir. International Association of Wood Anatomists Journal 21, 389–401.
Goodwin RH, Goddard DR. 1940. The oxygen consumption of isolated woody tissues. American Journal of Botany 27, 234–237.[CrossRef][Web of Science]
Gregory RA, Romberger JA. 1975. Cambial activity and height of uniseriate rays in conifers. Botanical Gazette 136, 246–253.[CrossRef]
Jaccard P. 1915. Über die Verteilung der Markstrahlen bei den conifern. Berichte der Deutschen Botanischen Gesellschaft 33, 492–495.
Levy PE, Meir P, Allen SJ, Jarvis SG. 1999. The effect of aqueous transport of CO2 in xylem sap on gas exchange in woody plants. Tree Physiology 19, 53–58.
Lev-Yadun, S. 1998. The relationship between growth ring width and ray density and ray height in cell number in the early wood of Pinus halepensis and Pinus pinea. International Association of Wood Anatomists Journal 19, 131–139.
Little TM, Hills FJ. 1978. Agricultural experimentation; design and analysis. New York: John Wiley & Sons, 118–124.
Maier CA, Zarnoch SJ, Dougherty PM. 1998. Effects of temperature and tissue nitrogen on dormant season stem and branch maintenance respiration in a young loblolly pine (Pinus taeda) plantation. Tree Physiology 18, 11–20.
Mauseth JD. 1988. Plant anatomy. Menlo Park, CA: The Benjamin/Cummings Publishing Co., Inc.
McCree KJ. 1970. An equation for the rate of dark respiration of white clover plants grown under controlled conditions. In: Prediction and measurement of photosynthetic productivity. Proceedings of the IBP/PP technical meeting, Trebon, 14–21 September 1969, Wageningen, Pudoc, 221–229.
Milliken GA, Johnson DE. 1984. Analysis of messy data, Vol. 1. Designed experiments. New York: Van Nostrand Reinhold Co., 315–321.
Moser LE, Volenec JJ, Nelson CJ. 1982. Respiration, carbohydrate content, and leaf growth of tall fescue. Crop Science 22, 781–786.
Panshin AJ, de Zeeuw C. 1980. Textbook of wood technology, 4th edn. San Francisco, CA: McGraw-Hill, Inc., 158, 181.
Penning de Vries FWT, Witlage JM, Kremer D. 1979. Rates of respiration and of increase in structural dry matter in young wheat, ryegrass and maize plants in relation to temperature, to water stress and their sugar content. Annals of Botany 44, 595–609.
Pregitzer KS, Laskowski MJ, Burton AJ, Lessard V, Zak DR. 1998. Variation in sugar maple root respiration with root diameter and soil depth. Tree Physiology 18, 665–670.
Pruyn ML, Gartner BL, Harmon ME. 2002a. Respiratory potential in sapwood of old versus young ponderosa pine trees in the Pacific Northwest. Tree Physiology 22, 105–116.
Pruyn ML, Gartner BL, Harmon ME. 2002b. Within-stem variation of respiration in Pseudotsuga menziesii (Douglas-fir) trees. New Phytologist 154, 359–372.[CrossRef][Web of Science]
Pruyn ML, Harmon ME, Gartner BL. 2003. Stem respiratory potential in six softwood and four hardwood tree species in the central cascades of Oregon. Oecologia 137, 10–21.[CrossRef][Web of Science][Medline]
Rasband W. 1996. NIH Image 1.6, Image analysis software. National Institutes of Health, Bethesda, MD, USA.
Ross WD, Krahmer R. 1971. Some sources of variation in structural characteristics of Douglas-fir bark. Wood and Fiber 3, 35–46.
Ryan MG. 1990. Growth and maintenance respiration in stems of Pinus contorta and Picea engelmannii. Canadian Journal of Forest Research 20, 48–57.
Ryan MG. 1991. Effects of climate change on plant respiration. Ecological Applications 1, 157–167.[CrossRef][Web of Science]
Ryan MG. 1995. Foliar maintenance respiration of subalpine and boreal trees and shrubs in relation to nitrogen content. Plant, Cell and Environment 18, 765–772.[CrossRef]
Ryan MG, Hubbard RM, Pongracic S, Raison RJ, McMurtrie RE. 1996. Foliage, fine-root, woody-tissue and stand respiration in Pinus radiata in relation to nitrogen status. Tree Physiology 16, 333–343.
Ryle GJA, Cobby JM, Powell CE. 1976. Synthetic and maintenance respiration losses of 14CO2 in uniculm barley and maize. Annals of Botany 40, 721–728.
Saranpää P, Höll W. 1989. Soluble carbohydrates of Pinus sylvestris L. sapwood and heartwood. Trees 3, 138–143.
SAS Institute. 1998. Statistical analysis systems software, release 8.0. Cary, NC; SAS Institute Inc.
Sauter JJ, van Cleve B. 1993. Occurrence of a maltose pool and of maltase in poplar wood (PopulusxCanadensis ‘robusta’) during Fall. Journal of Plant Physiology 141, 248–250.[Web of Science]
Sauter JJ, van Cleve B. 1994. Storage, mobilization and interrelations of starch, sugars, protein and fat in the ray storage tissue of poplar trees. Trees 8, 297–304.
Shain L, Mackay JFG. 1973. Seasonal fluctuation in respiration of aging xylem in relation to heartwood formation in Pinus radiata. Canadian Journal Botany 51, 737–741.[CrossRef]
Stockfors J, Linder S. 1998. Effects of nitrogen on the seasonal course of growth and maintenance respiration in stems of Norway spruce trees. Tree Physiology 18, 155–166.
Sundberg B, Ericsson A, Little CHA, Näsholm T, Gref R. 1993. The relationship between crown size and ring width in Pinus sylvestris L. stems: dependence on indole-3-acetic acid, carbohydrates and nitrogen in the cambial region. Tree Physiology 12, 347–362.
Thornley JHM. 1970. Respiration, growth and maintenance in plants. Nature 227, 304–305.[Medline]
Vose JM, Ryan MG. 2002. Seasonal respiration of foliage, fine roots, and woody tissues in relation to growth, tissue N, and photosynthesis. Global Change Biology 8, 182–193.[CrossRef][Web of Science]
Yang KC. 1993. Survival rate and nuclear irregularity index of sapwood ray parenchyma cells in four tree species. Canadian Journal of Forest Research 23, 673–679.[CrossRef]
Yang KC, Chen YS, Benson CA. 1994. Vertical and radial variation of nuclear elongation index of living sapwood ray parenchyma cells in a plantation tree of Cryptomeria japonica. International Association of Wood Anatomists Journal 15, 323–327.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||