| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
What's this? |
Anatomy and Morphology |
University of Maine, Department of Forest Ecosystem Science, 5755 Nutting Hall, Orono, Maine 04469-5755 USA
Received for publication January 20, 2005. Accepted for publication December 2, 2005.
ABSTRACT
The dual function provided by longitudinal tracheids in conifers has led to a generally held trade-off concept that increasing wall thickness and/or volume of latewood tracheids improves mechanical support, while increasing cell diameter and/or volume of earlywood tracheids enhances conductive potential. Yet, some conifers have either uniform cell structure across the growth ring or, at most, a small amount of latewood. How do these trees accomplish the needs for increasing support and conduction with height growth? We examined Metasequoia glyptostroboides, a species that we previously demonstrated improves its mechanical properties with increasing age without a change in specific gravity or secondary wall microfibril angle. In this paper, we showed that lignin and extractive contents are not contributing factors, and through composite structure analysis, we eliminated a role for tracheid length. Using micromorphometric analysis, we demonstrated that as cell diameter increases, total primary wall decreases, secondary wall increases, and strength and conductive capacity increase with no change in specific gravity. Meta-analysis using other species of Cupressaceae, Podocarpaceae, and Araucariaceae provided strong corroborative evidence for this design strategy.
Key Words: Araucariaceae • conifers • Cupressaceae • hydraulics • mechanics • Metasequoia • Podocarpaceae • xylem
In his classic review paper, Bailey (1953)
noted that the two major functions of xylem tracheids in conifers—support and conduction—"are to a considerable degree mutually antagonistic." This concept of xylem trade-offs has been discussed by others (Long et al., 1981
; Carlquist, 1988
; Niklas, 1992
; Gartner, 1995
; Jagels et al., 2003
; Taneda and Tateno, 2004
; McCulloh and Sperry, 2005
). Recently, Hacke et al.
(2001) found a correlation between hydraulic implosion resistance and wood density. They suggested that conducting cells have large hoop stresses that would lead to cell collapse if neighboring cells did not provide mechanical support, especially under drought conditions. In a subsequent paper (Jagels et al., 2003
), we suggested that for low-density conifers with high microfibril angles in tracheids, hoop stresses may, through Poisson's ratio, be at least partially converted to longitudinal strain (Boyd, 1950
), thus reducing the dependency on collaborative cell support in low-density conifers.
Perhaps more critical for many conifers that have the potential to develop into very tall trees on sites where water is not limiting, is how to maintain or increase xylem strength concomitant with improving conductive efficiency. Many conifer species, particularly members of the Pinaceae, have evolved from the more primitive, uniform manoxylic type to the more specialized pycnoxylic wood structure where larger diameter, thin-walled, conductively efficient earlywood tracheids are sandwiched between layers of smaller diameter, thicker-walled, stronger, but less conductively efficient latewood tracheids; a strategy that allows for a partial disconnect between support and conduction (Bailey, 1953
; Boatwright and Garrett, 1983
; Marshall, 1998
; Uggla et al., 2001
; Domec and Gartner, 2002
).
Yet, this explanation is inadequate to account for many conifers, notably many members of the Cupressaceae, Araucariaceae, and Podocarpaceae that have wood with either uniform structure or only develop an insignificant latewood (Wardrop and Addo-Ashong, 1963
; Greguss, 1955
). Barbour and Whitehead (2003)
partially addressed this issue by examining xylem sap velocity in Dacrydium cupressinum (Podocarpaceae), a species with nearly uniform cell structure across the growth ring (Greguss, 1955
). Testing a model proposed by Roderick and Berry (2001)
, they found a negative correlation between average sap velocity and wood density. Although wood density is not correlated with tree height (Greenhill, 1881
; McMahon and Kronauer, 1976
), it is considered the best estimate of mechanical properties where reaction wood is absent (deZeeuw, 1965
; Niklas, 1992
; U.S. Forest Products Laboratory, 1999
). Therefore, a key question is how do conifers with relatively uniform xylem structure adapt to increasing mechanical stresses and hydraulic demands as the tree increases in height? In dicotyledonous angiosperms, the issue has been resolved by the separation of the roles of conduction and support to different cell types, with the corollary that these functions are likely decoupled (Woodrum et al., 2003
).
In a recent study (Jagels et al., 2003
), we noted what seemed to us an anomalous finding in Metasequoia glyptostroboides Hu et Cheng, a species with low density and narrow, poorly defined latewood. We measured modulus of rupture (MOR) and modulus of elasticity (MOE) during the bending of green wood and found that these mechanical properties increased (but only the MOE change was significant at 
= 0.01) between wood sampled near the center of the tree and wood nearer the bark—a finding consistent with that reported for other conifer species (Pansin and deZeeuw, 1980
; Easterling et al., 1982
). However, inconsistent with most reports was our finding that both specific gravity (SG) and microfibril angle (MFA) remained unchanged. Several studies have shown a strong relationship between MFA and wood strength (see references in Butterfield, 1998
). Because we found no evidence of compression wood, we could offer no rational explanation, but suggested that the strength change might be related to a change in tracheid length (Carlquist, 1975
; Kaya and Smith, 1993
) or to our observation of a localized plastic deformation (Jagels et al., 2003
). We will examine the tracheid length question later, but can discard the plastic deformation explanation because careful microscopic examination of the test specimens revealed no inelastic strain.
Because the greatest mechanical stresses occur near the base of the stem, any structural adjustments made as a tree ages will be manifested most prominently in this region (Archer, 1986
; Niklas, 1992
). In most conifers, MFA decreases while SG increases, leading to increases in strength with increasing distance from the pith (Dadswell, 1958
; McMillin, 1973
; Bendtsen, 1978
; Butterfield, 1998
). Metasequoia provides an example of a conifer for which MFA and SG do not change, at least for the first two to three decades, yet strength increases (Jagels et al., 2003
). As such, it is an ideal model for examining other factors that could influence changes in mechanical properties. Increases in theoretical conductive efficiencies in conifers, as manifested in increases in tracheid length and diameter, occur as a progression from pith to bark at tree base, and decline with tree height (Bailey, 1958
; Bannan, 1965
; Carlquist, 1975
; Rundel and Stecker, 1977
). Therefore, the analyses undertaken in this study are focused on the chemical and structural changes in this basal region.
Within conifers, in addition to the factors of density and MFA, ring width, tracheid length and the chemical properties conferred by the relative percentages of lignin and extractives have been suggested to influence wood properties, such as strength. Holocellulose can also affect wood strength, but primarily as a consequence of its (1) structure (crystalline or amorphous), (2) orientation (i.e., MFA of the secondary wall), and most importantly, (3) its distribution (primary and S1 layers vs. secondary walls). Compression wood can also affect strength negatively (Anderson, 1951
; Zobel and McElwee, 1958
; deZeeuw, 1965
; Côté et al., 1966
; Larson, 1966
; Arganbright, 1971
; Uprichard, 1971
; Schniewind, 1972
; Schimleck et al., 2003
; Singleton et al., 2003
; Grabner et al., 2005
). In the case of Metasequoia, we have previously shown that density and MFA are not correlated with an increase in mechanical properties and that compression wood was absent from test samples (Jagels et al., 2003
). In this study, we investigated the potential influences of (1) ring width, (2) percentage latewood, (3) lignin content, (4) extractive content, (5) tracheid length, and (6) relative proportions of wall layers to assess the possible relationship of any of these factors with mechanical property change. To assess point (6), we developed a geometric model, and by stereological analysis, estimated volumetric change in wall layers with age. We also measured changes in tracheid diameter and determined relative changes in conductive efficiency per unit area. Utilizing published strength and tracheid diameter parameters in a meta-analysis, we test the validity of our developed model for other conifers that have similar xylem structure. Finally, we compared the relative adaptive features of the manoxylic and pycnoxylic structure plans in extant conifers.
MATERIALS AND METHODS
In Jagels et al. (2003)
, two Metasequoia trees from closed canopy stands in China (designated JPC) and New Jersey, USA (designated PNJ) were sampled for tests of mechanical properties, analysis of decay resistance, and determination of specific gravity, MFA, and tracheid length. These samples are used in the present study. In a subsequent paper (Visscher and Jagels, 2003
), a procedure was developed to digitally determine cells per unit area (CPA) in thin transverse sections of Metasequoia. Here, thin sections (18–-22 µm) of rings along two opposing radii were prepared from breast height disks of trees JPC and PNJ, using a sliding microtome (model 860, American Optical, Buffalo, New York, USA). Sections were stained overnight in 1% Bismark brown and mounted in a low-viscosity medium (Cytoseal 60, Richard-Allan Scientific, Kalamazoo, Michigan, USA). CPAs were measured in black and white images of transverse sections at magnification of 100x, using a Zeiss Axioskop (Oberkochen, Germany), as previously described (Visscher and Jagels, 2003
). Percentage cell wall per unit area (CWA) was measured at six randomly chosen places in the earlywood of rings 2, 6, 8, 12, 16, 18, 22, and 26. Images captured with a SPOT-RT digital camera (Diagnostic Instruments, Sterling Heights, Michigan, USA) were converted to threshold images in WinSeedle program (version 5.1A; Regent Instruments, Quebec, Canada), using Photo-Paint (Corel, Ottawa, Ontario, Canada). To adjust for slight variations in section staining or image quality, threshold levels were manually adjusted to maximize the amount of cell wall area converted to black pixels. Images were analyzed both as dark objects on a pale background (to obtain CWA) and as light objects on a dark background (to obtain percentage lumen area, CLA). The combined values represent the total pixels in each image (1 920 000). The ratio of pixels per millimeter was used to convert values to percentage wall (or lumen) per square millimeter. Ring width was measured for each annual increment along both radii and averaged. Cell shape and wall thickness in earlywood and latewood were examined microscopically and assessed for latewood according to Mork's index (Mork, 1928
) or the circularity index (Jagels and Dyer, 1983
; Jagels et al., 1994
). For determining tracheid length/diameter ratios, wood was macerated, stained, and measured from digital images (Jagels et al., 1982
).
Due to an insufficient quantity of wood from the JPC tree, only the the PNJ tree was chemically analyzed. Small blocks, approximately 2.5 cm3, were cut from each end of beams previously used for mechanical testing (Jagels et al., 2003
). Sapwood had been removed during slabbing of logs, so all samples consisted of heartwood. Blocks from inner heartwood near the center of the tree (n = 16) and outer heartwood (n = 16) were each separately pooled and analyzed as paired batches, using TAPPI (Technical Association of Pulp and Paper Institute) standard methods (T264-om 88 and T222-om 88) for extractives and lignin (Jagels et al., 1988
; Polman et al., 1999
) using the less toxic ethyl alcohol-cyclohexane rather than benzene for extractive removal (Singleton et al., 2003
). Tukey's HSD test was used to test for mean separations.
To test for the relative proportions of combined primary (P) and S1 layers in comparison to secondary, S2, wall layers, model grids were created and scaled to the cross-sections of rings 6 and 26 in PNJ, each with the same CWA. From these grids, relative changes in proportion of P-S1 to S2 were calculated, using a linear measure of total cell wall perimeter (CWP) as an estimate of aerial density, and by stereological extrapolation, volume density (Weibel and Bolander, 1973
).
Theoretical conductive efficiency per unit area was compared between ring 6 and 26. Mean tracheid diameter, d(di), derived from digital image microscopy, was calculated from CLA, CWA, and CPA. To test whether this value could serve as a reasonable surrogate for hydraulic mean diameter in a relatively uniform structure conifer, we measured cell lumen diameter, d, directly on printed microscopic images (N = 200 per ring). From these d values, we calculated arithmetic means (
d/N), Hagen-Pouseuille mean diameters ([d5/d4]
) (Tyree and Zimmermann, 2002
) and weighted hydraulic mean diameters (d5/d4) as described by Kolb and Sperry (1999)
. Divergence from circularity was ignored because lumen shape for both rings was similar, and we assumed that pitting and tracheid length were scaled to lumen diameter (Hacke et al., 2004
).
Using meta-analysis, data were compiled from several sources (see Table 3 for list) to assess whether other conifers with relatively uniform wood structure fit the observed strategy employed by Metasequoia. Species in the families Cupressaceae, Araucariaceae, and Podocarpaceae were chosen. Criteria for inclusion were (1) mostly a uniform cross-sectional structure, with or without a narrow latewood; (2) one or more cohorts (preferably closely related) with matched green-basis specific gravity (green volume/oven-dry mass); (3) published data on MOR and MOE in bending for green wood; (4) some estimate of average tracheid diameter; and (5) published, non-overlapping maximum tree heights. Meeting all of these criteria limited the sample size; taxonomic divergence was accepted where closely related species could not be found (note that Araucaria and Prumnopitys are in different families, the Araucariaceae and Podocarpaceae, respectively). A group of Picea species (Pinaceae family) with more strongly defined and wider latewood was used for comparison. To test for differences, wood property variation values were taken from Markwardt and Wilson (1935)
because these authors tested the widest range of tree species within a single volume. The variances are listed in Table 3 for green SG, MOR, and MOE. Assuming that sample sizes for published means exceeded eight (a very conservative estimate), our calculations determined that means differing by more than the Markwardt and Wilson (1935)
percentage variation values would be significant at < 0.05. Species comparisons were limited to those cohorts that had non-overlapping maximum tree heights and SG values that differed by less than the Markwardt and Wilson (1935)
average 8% variance. These were then tested for differences that exceeded the variances for MOR (12%) and MOE (16%).
|
Measuring the width of true latewood based on either Mork's index (using double-wall thickness change) or the circularity index (shape change) proved elusive. Along individual radial files of cells, true latewood (Jagels et al., 1994
) ranged from zero to a maximum of 10 cells. Some of this variability is revealed in Fig. 1a. Transition latewood (Fig. 1b), or cells that expand in diameter like earlywood but also reveal some wall thickening (Larson, 1969
), developed in a zone that began from a few to more than 50 cells after the onset of lateral annual growth and extended to the zone of latewood. Thus, with the exception of a very few cells, the wood is relatively homogeneous in transverse view.
|
= 0.05). The CPA difference is illustrated further in Fig. 3. Both trees followed similar patterns from inner to outer wood, but tracheid length changed more in PNJ than JPC, and CPA values were smaller in PNJ than JPC.
|
|
Chemical analysis for content of lignin and hot water and organic reagent soluble extractives revealed no significant differences between samples from inner or outer heartwood (Table 1). These values are similar to those reported by Polman et al. (1999)
. By extrapolation from these data, holocellulose content should also be unchanged.
|
; Panshin and deZeeuw, 1980
; Cave and Walker, 1994
; Walker and Woolens, 1998). Due primarily to its parallel arrangement of cellulose microfibrils into crystalline arrays, the S2 layer can sustain maximum axial stresses 9–17 times greater than the S1 layer (Kretschman et al., 1998
; Mark, 1965
). This provides the theoretical basis for assessing the relative proportions of these wall layers. In Fig. 4 we have created a model with two grids that are closely scaled to the change in CPA between ring 6 and 26 in PNJ. The model shows a decrease of 125% in cells per mm2, while the PNJ tree shows a similar change of 132%. In the model, this leads to a 50% decrease in CWP. Assuming tracheids were also square in sectional view, a CWP decrease of 52% would be predicted between rings 6 and 26 (Fig. 4). Because tracheids have rounded corners in sectional view, this prediction is a slight overestimate. A conservative estimate would likely exceed 40%—still a significant decrease in the aerial density of the weakest wall layers. Because total CWA remains constant, this translates into a similar increase in the stronger S2 layer.
|
). Table 2 provides a comparison for relative change in mean tracheid diameter to the fourth power times unit area using arithmetic means (d(di), d/N) and two "hydraulic" means. The Kolb and Sperry (1999)
hydraulic mean (d5/d4) yielded the least change between ring 6 and 26, but this weighted mean is more useful for relating cavitation resistance to conducting efficiency in hardwoods (Kolb and Sperry, 1999
). For the purpose of this study of conifers that limit cavitation with numerous bordered pits rather than conduit diameter, specific conductivity calculated from the Hagen-Pouseuille formula is more appropriate (Tyree and Zimmermann, 2002
). This value was quite similar to that calculated from d(di), both predicting about a two-fold increase in specific hydraulic conductivity.
|
average variance estimate of 12%; but among the MOE comparisons, the species with the larger average tracheid diameter had MOE values that were equal to or exceeded the 16% variance estimate of Markwardt and Wilson (1935)
when compared to the SG cohort. Also, the species with the highest average MOE values and tracheid diameters were, in each case, the tallest trees. The only exception was among the comparison group of three Picea species that all have a well-developed latewood. In this group, no pattern could be discerned and no significant differences were found. DISCUSSION
In Jagels et al. (2003)
, we eliminated compression wood, SG, and MFA as possible factors that could have influenced an increase in mechanical properties between inner and outer heartwood. The current research provides further confirmation for a lack of change in SG by showing no change in CWA (Fig. 2). Lignin and extractive content are similarly eliminated as contributing factors by data in Table 1. A corollary of this analysis is that total carbohydrate content also remains unchanged (Browning, 1967
; Sjöström, 1981
). Approximately 2–4% of this would be nonstructural polysaccharide, such as stored starch, while the remainder is structural holocellulose, comprising more than half the dry weight of wood (Sjöström, 1981
). It should be noted that in this study we did not analyze sapwood because most extractives are not produced until the transition from sapwood to heartwood occurs, with the onset of death of the parenchyma cells (Sjöström, 1981
). The other chemical components and physical parameters are established during cell differentiation in the year the ring was formed, and once these cells are dead, no further changes occur in the sapwood or heartwood (Panshin and deZeeuw, 1980
; Larson, 1994
).
Latewood comprised such a small proportion of the ring that its contribution to strength appears to be negligible because SG was unresponsive to ring width. Similarly, ring width had no relationship with any other parameter, including strength (Jagels et al., 2003
). In one tree (PNJ), the widest rings were near the center of the tree, while in the other (JPC) the widest rings were in later-formed wood. These results confirm the negligible influence of latewood in Metasequoia. By contrast, in strongly pycnoxylic woods, ring width, by affecting the relative proportions of earlywood and latewood, is generally correlated with mechanical properties (Panshin and deZeeuw, 1980
).
Woodrum et al. (2003)
found a positive correlation between ray volume and MOE among five different species of Acer, although the authors point out that this relationship may be "coincidental" because observed changes in fiber density (and hence wood density) could have compensated for the observed increases in parenchyma content. In fact, the data in their paper shows a positive correlation between increasing wood density and increasing ray parenchyma. Evidence that volume of ray and longitudinal parenchyma is decoupled from wood density or strength has been demonstrated for a wide number of hardwoods (see Tables 5–7 in Panshin and deZeeuw, 1980
). Previously, as a test for ray volume variability within a single conifer species, we collected Larix laricina wood from its extensive natural range, from Maine to the western Northwest Territories and from West Virginia to Labrador, and found no differences in ray volume related to age, habitat, or geographical location (R. Jagels and G. Visscher, unpublished data). Ray volume in Metasequoia averages 5.3% with a variance of less than 1% (Visscher, 2002
); this is typical for conifers (Panshin and deZeeuw, 1980
). We conclude that ray parenchyma variance has a very low probability for influencing the mechanical property changes we observed.
Based on assumptions and theoretical calculations, the size and number of bordered pits may have some influence on wall strength (Boyd, 1985
; Carlquist, 1988
). Sperry and Hacke (2004)
calculated that pit apertures (but not the pit chamber) could reduce implosion resistance. In strongly pycnoxylic conifers, pit diameters are linked to wall thickness by an inverse relationship (Jane et al., 1970
; Carlquist, 1988
), but in woods with little or no latewood (like Metasequoia), pit aperture size varies little (Greguss, 1955
). In this study, we did not determine number of pits per radial sectional area and, therefore, cannot exclude this as an influencing factor. A significant change in number of pits, however, should be reflected in a change in density. Furthermore, resistance to implosion (as calculated by Sperry and Hacke, 2004
) is quite different from the complex of compressive, tensile, and shear stresses imposed in a bending test.
Thus, two prime contenders remain: tracheid length and CPA. Some previous studies have suggested a link between strength or stiffness and tracheid length (Wellwood, 1962
; Carlquist, 1975
; Rundel and Stecker, 1977
). Carlquist (1975)
suggested that an increase in tracheid length might be associated with a greater need for support. Other studies have linked tracheid length and tracheid diameter to hydraulic conductance (Bannan, 1965
; Rundel and Stecker, 1977
; Lewis and Tyree, 1985
; Zobel and Sprague, 1998
; Hacke et al., 2004
). Because an increase in tracheid length is usually associated with a decrease in MFA (Wardrop and Dadswell, 1950
; Hiller and Brown, 1967
; Walker and Woolons, 1998
), tracheid length may simply be a proxy for MFA, the factor actually affecting strength differences. In Metasequoia, however, no relationship between tracheid length and MFA was found. A similar lack of relationship has been reported for root wood of Pinus radiata and P. nigra (Matsumura and Butterfield, 2001
).
By considering conifer wood as a natural composite material consisting of tracheids, that is, as short "fibers," embedded in a matrix of lignin, one can model the effect of fiber length on strength. In short-fiber composites, once a minimum fiber aspect ratio (length/diameter: l/d) exceeds approximately 50 : 1, any further increase in fiber length has no appreciable impact on composite strength (Agarwal and Broutman, 1990
). Bannan (1965)
measured tracheid l/d ratios for approximately 24 conifer species. The minimum average value he measured was 72 : 1 in juvenile stems of Thuja occidentalis. The largest ratio for mature wood was 143 : 1 for Sequoia sempervirens. In Metasequoia we found l/d ratios that varied from 98 : 1 to 107 : 1—well over the minimum of 50 : 1.
Manufactured composites mostly rely on surface bonding of fibers to a matrix, and thus, failure generally involves fiber pullout, in which short fibers separate from the embedding matrix (Agarwal and Broutman, 1990
). In wood, the lignin matrix permeates all layers of tracheid cell walls. Consistent with this, Mark (1967)
noted that failure in wood generally initiates in the S1 layer of the cell wall, not in the middle lamella ("matrix") region between tracheids. Groom et al. (2002)
reported occasional cell separation failure in the middle lamella region, but only in high-density very thick-walled latewood cells in Pseudotsuga menziesii. These observations support the concept that tracheid length, certainly in a low-density wood like Metasequoia, is unlikely to be a direct determinant of mechanical properties, and instead mechanical failure will most likely initiate in the interface between primary and S1 wall layers.
This led us to the question of how an increase in cell size while cell-wall volume and SG remain constant leads to an increase in strength. Marshall (1998)
has shown that for honeycomb designs made from uniform, homogeneous materials (such as aluminum), strength remains constant for honeycombs that differ in cell size, so long as density remains the same. However, wood is not a homogeneous material. The tracheid cell wall is multilayered, and each layer has different properties. Using this knowledge, we modeled how the relative proportions of P-S1 and S2 would change as CWA remained constant between rings 6 and 26 (Fig. 4). Based on our model, we conservatively estimated that the weaker P-S1 decreased by more than 40% while the stronger S2 increased by a similar amount. Thus, the most parsimonious interpretation for the increase in strength, without an increase in specific gravity, between inner and outer wood is that in the outer wood, a higher proportion of cell wall is composed of cellulose in parallel, crystalline microfibrils that enhance mechanical properties (Harada, 1965
).
Our test for differences between the calculated change in specific conductivity based on Hagen-Pouseuille tracheid diameter and simple mean diameter revealed more congruence than divergence (Table 3), with both estimates suggesting an approximately two-fold increase in conductive capacity after 20 years of growth. Although we recognize that mean tracheid diameter is not a valid substitute in many cases, the closeness in tracheid diameters revealed by our analysis of a conifer with relatively uniform xylem structure gave us confidence in using average tracheid diameters from the wood anatomy literature for our meta analysis, in which we were looking for relative differences in conductivity, not absolute ones.
The validity and extendibility of the Metasequoia model was tested by meta-analysis using other tree species with similar xylem structure (Table 3), and congruity was found among all group comparisons. The within-tree location of the test samples that form the data set for Table 2 is not known. However, wood testing laboratories generally test wood from trees that have reached commercial size (usually >30 years), and due to a protocol requirement that wood be "clear" (knot-free), test beams are generally taken from outer, mature wood of basal logs (Markwardt and Wilson, 1935
; U.S. Forest Products Laboratory, 1999
). The variance values used in Table 3 are conservatively large because Markwardt and Wilson (1935)
tested both conifers and dicotyledonous woods, including the notoriously variable ring-porous hardwoods (Panshin and deZeeuw, 1980
). For each comparison, the species that has the potential to attain the greatest maximum height had the highest MOE. As is the case for Metasequoia (Jagels et al., 2003
), MOE rather than MOR is the mechanical property that is weakly linked to SG but strongly linked with tracheid diameter. This fits with the concept that MOE is more important in tall trees, reducing sway, which in turn reduces the chance of Euler or Brazier buckling (Niklas, 1992
).
Among the strongly pycnoxylic spruces, in which a true latewood is much better developed, the relationship between tracheid diameter and MOE did not hold (Table 3), and we interpret this to be a consequence of the more often cited design strategy in which a well-defined latewood overwhelms the contribution of earlywood to strength, particularly with increasing age (Wellwood, 1962
; Panshin and deZeeuw, 1980
).
The possible presence of radially differentiated mechanical stresses in the living tree xylem, as noted by Boyd (1950)
and Gillis and Hsu (1979)
, and reviewed by Niklas (1992)
, was not investigated in this study and therefore cannot be eliminated as a contributor to the differential mechanical properties observed. However, indirect evidence suggests that significant radial differences are unlikely in Metasequoia. Because we observed no change in MFA with radial position, the usual increased tensional stresses associated with diminution of MFA is not likely (Boyd, 1950
), and for the meta-analysis of similar conifers, in which selection of test samples was presumably random, the relationship between cell diameter and MOE still held.
In summary, contrary to intuitive reasoning and prevailing theory, some conifers have the ability to synchronously improve mechanical stiffness and hydraulic efficiency in the same cells with no change in specific gravity of the xylem. This strategy creates the opportunity for simultaneously improving xylem strength and conductivity in the same cells. This more primitive manoxylic design plan also has advantages for improving overall resistance to hydraulically induced implosion. It should be noted that a uniform "honeycomb" structure is one commonly adopted in other natural (honeybees) and manmade structures.
Yet, among extant conifers, the pycnoxylic form prevails and thus must have distinct evolutionary advantages. The most likely explanation is that the pycnoxylic plan offers greater adaptability in strongly seasonal environments, particularly where moisture is plentiful early in the growing season and scarce later. The conifers that have retained the manoxylic or weakly pycnoxylic structure are often restricted to relatively stable moisture regimes, that are more similar to earlier geologic periods. For example, of the extant genera of the Cupressaceae, many occupy wet–mesic sites in both North America and Asia (for example, Sequoia, Metasequoia, Taxodium, Thuja, Glyptostrobus) where they often attain great heights. However, as continuously moist sites have disappeared through time, most of these genera have become quite restricted or relicts in their natural ranges. As a consequence, genera in the strongly pycnoxylic Pinaceae have come to occupy much larger geographical ranges, reflecting their greater adaptability to moisture seasonality.
FOOTNOTES
The authors thank M. T. Tyree for critical reading of manuscript, J. S. Sperry for helpful advice, W. A. Halteman for statistical advice, D. Cirelli for technical assistance, and J. E. Kuser for donating a Metasequoia tree. Support for this project was provided by The Andrew W. Mellon Foundation and McIntire-Stennis funds, University of Maine. Maine Experiment Station report No. 2840. 
2Author for correspondence (e-mail: Richard.Jagels{at}maine.edu
)
LITERATURE CITED
Agarwal B. D Broutman L. J 1990 Analysis and performance of fiber composites, 2nd ed Wiley New York, New York, USA
Alden H. A 1997 Softwoods of North America. General Technical Report FPL-GTR-102 Forest Products Laboratory Madison, Wisconsin, USA
Anderson E. A 1951 Tracheid length variation in conifers as related to distance from the pith. Journal of Forestry 49: 38-42
Archer R. R 1986 Growth stresses and strains in trees Springer-Verlag Berlin, Germany
Arganbright D. C 1971 Influence of extractives on bending strength of redwood (Sequoia sempervirens). Wood and Fiber 2: 367-372
Bailey I. W 1953 Evolution of the tracheary tissue of land plants. American Journal of Botany 40: 4-8[CrossRef][Web of Science]
Bailey I. W 1958 The structure of tracheids in relation to movement of liquids, suspensions and undissolved gases. In K. V. Thimann {ed.}, The physiology of forest trees. Ronald Press, New York, New York, USA
Bannan M. W 1965 The length, tangential diameter, and length/width ratio of conifer tracheids. Canadian Journal of Botany 43: 967-984[CrossRef][Web of Science]
Barbour M. M Whitehead D 2003 A demonstration of the theoretical prediction that sap velocity is related to wood density in the conifer Dacrydium cupressinum. New Phytologist 158: 477-488[CrossRef][Web of Science]
Bendtson B. A 1978 Properties of wood from improved and intensively managed trees. Forest Products Journal 28: 61-72[Web of Science]
Boatwright S. W Garrett G. G 1983 The effect of microstructure and stress state on the fracture behavior of wood. Journal of Material Science 18: 2181-2199[CrossRef]
Bootle K. R 1983 Wood in Australia McGraw-Hill Sydney, Australia
Boyd J. D 1950 Tree growth stresses. II. The development of shakes and other visual failures in timber. Australian Journal of Applied Science 1: 296-312
Boyd J. D 1985 Biophysical control of microfibril orientation in plant cell walls Martinius Nijhoff/Junk Dordrecht, Netherlands
Browning B. L 1967 Methods of wood chemistry, vol II. Interscience New York, New York, USA
Butterfield B. G 1998 Microfibril angle in wood. Proceedings of International Association of Wood Anatomists/International Union of Forest Research Organization International Workshop University of Canterbury Christchurch, New Zealand
Carlquist S 1975 Ecological strategies of xylem evolution University of California Press Berkeley, California, USA
Carlquist S 1988 Comparative wood anatomy: systematic, ecological, and evolutionary aspects of dicotyledon wood Springer Berlin, Germany
Cave I. D Walker J. C. F 1994 Stiffness of wood in fast-grown plantation softwoods: the influence of microfibril angle. Forest Products Journal 44: 43-48
Côté W. A. Jr. 1965 Cellular ultrastructure of woody plants Syracuse University Press Syracuse, New York, USA
Côté W. A. Jr. Day A. C Simpson B. W E Timell T 1966 Studies of larch arabinogalactan. I. The distribution of arabinogalactan in larch wood. Holzforschung 20: 178-192[Web of Science]
Dadswell H. E 1958 Wood structure variations occurring during tree growth and their influence on properties. Journal of the Institute of Wood Science 1: 11-32
DeZeeuw C 1965 Variability in wood. In W. A. Côté Jr Cellular ultrastructure of woody plants 457-471 Syracuse University Press Syracuse, New York, USA
Domec J. C Gartner B. L 2002 Age- and position-related changes in hydraulic versus mechanical dysfunction of xylem: inferring the design criteria for Douglas-fir wood structure. Tree Physiology 22: 91-104[Abstract]
Easterling K. E Harryson R Gibson L. J Ashby M. F 1982 On the mechanics of balsa and other woods. Proceedings of the Royal Society of London, A 383: 31-41
Elias T. S 1987 Trees of North America Gramercy New York, New York, USA
Gartner B. L 1995 Patterns of xylem variation within a tree and their hydraulic and mechanical consequences. In B. L. Gartner [ed.], Plant stems, physiology and functional morphology, 125–149. Academic Press, New York, New York, USA
Gillis P. P Hsu C. H 1979 An elastic, plastic theory of longitudinal growth stresses. Wood Science and Technology 13: 97-115
Grabner M Müller U Gierlinger N Wimmer R 2005 Effects of heartwood extractives on mechanical properties of larch. International Association of Wood Anatomists Journal 26: 211-220
Greenhill A. G 1881 Determination of the greatest height consistent with stability that a vertical pole or mast can be made, and of the greatest height to which a tree of given proportions can grow. Proceedings of the Cambridge Philosophical Society 4: 65-73
Greguss P 1955 Indentification of living gymnosperms on the basis of xylotomy Akadémiai Kiado Budapest, Hungary
Groom L Mott L Shaler S 2002 Mechanical properties of individual southern pine fibers. I. Determination and variability of stress-strain curves with respect to tree height and juvenility. Wood and Fiber Science 34: 14-27[Web of Science]
Hacke U. G Sperry J. S Pittermann J 2004 Analysis of circular bordered pit function. II. Gymnosperm tracheids with torus-margo pit membranes. American Journal of Botany 91: 386-400
Hacke U. G Sperry J. S Pockman W. T Davis S. D McCulloh K. A 2001 Trends in wood density and structure are linked to prevention of xylem implosion by negative pressure. Oecologia 126: 457-461[CrossRef][Web of Science]
Harada H 1965 Ultrastructure and organization of gymnosperm cell walls. In W. A. Côté Jr Cellular ultrastructure of woody plants 215-233 Syracuse University Press Syracuse, New York, USA
Hiller C. H Brown R. S 1967 Comparison of dimensions and fibril angles of loblolly pine tracheids formed in wet or dry growing seasons. American Journal of Botany 54: 453-460[CrossRef][Web of Science]
Jagels R Dyer M. V 1983 Morphometric analysis applied to wood structure. I. Cross-sectional cell shape and area change in red spruce. Wood and Fiber Science 15: 376-386[Web of Science]
Jagels R Gardner D. J Brann T. B 1982 Improved techniques for handling and staining wood fibers for digitizer-assisted measurement. Wood Science 14: 165-167[Web of Science]
Jagels R Hornbeck J Marden S 1994 Drought and cold stress-induced morphometric changes in tree rings of red spruce. Maine Agricultural and Forest Experiment Station Technical Bulletin 159 Orono Maine, USA
Jagels R Seifert B Shottafer J. E Wolfhagen J. L Carlisle J. D 1988 Analysis of wet-site archeological wood samples. Forest Products Journal 38: 33-38
Jagels R Visscher G. E Lucas J Goodell B 2003 Palaeo-adaptive properties of the xylem of Metasequoia: mechanical/hydraulic compromises. Annals of Botany 92: 79-88
Jane F. W Wilson K White D. J. B 1970 The structure of wood Adam and Charles Black London, UK
Kaya F Smith I 1993 Variation in crushing strength and some related properties of red pine. Wood Science and Technology 27: 229-239
Keating W. G Bolza E 1982 Characteristic properties and uses of timbers, vol. I. Southeast Asia, Northern Australia and the Pacific Inkata Press Melbourne, Australia
Kolb K. J Sperry J. S 1999 Differences in drought adaptation between subspecies of sagebrush (Artemesia tridentata). Ecology 80: 2373-2384[CrossRef][Web of Science]
Kretschmann D. E Alden H. A Verrill S 1998 Variations of microfibril angle in loblolly pine: comparison of iodine crystallization and x-ray diffraction techniques. In B. G. Butterfield [ed.], Microfibril angle in wood, 157–176. Proceedings of International Association of Wood Anatomists/International Union of Forest Research Organization International Workshop, University of Canterbury, Christchurch, New Zealand
Larson P. R 1966 Changes in chemical composition of wood cell walls associated with age in Pinus resinosa. Forest Products Journal 16: 37-45
Larson P. R 1969 Wood formation and the concept of wood quality. Yale University, School of Forestry, Bulletin No 74 New Haven, Connecticut, USA
Larson P. R 1994 The vascular cambium: development and structure Springer-Verlag Berlin, Germany
Lewis A. M Tyree M. T 1985 The relative susceptibility to embolism of larger vs. smaller tracheids in Thuja occidentalis. International Association of Wood Anatomists Bulletin, new series 6: 93
Long J. N Smith F. W Scott D. R. M 1981 The role of Douglas-fir stem, sapwood and heartwood in the mechanical and physiological support of crowns and development of stem form. Canadian Journal of Forest Research 11: 459-464[CrossRef]
Mark R 1965 Tensile stress analysis of the cell walls of coniferous tracheids. In W. A. Côté Jr Cellular ultrastructure of woody plants 493-533 Syracuse University Press Syracuse, New York, USA
Mark R. E 1967 Cell wall mechanics of tracheids Yale University Press New Haven, Connecticut, USA
Markwardt L. J Wilson T. R. C 1935 Strength and related properties of woods grown in the United States. USDA Technical Bulletin No. 479 Washington D.C., USA
Marshall A. C 1998 Sandwich construction. In S. T. Peters [ed.], Handbook of composites, 254–290. Chapman & Hall, New York, New York, USA
Matsumura J Butterfield B. G 2001 Microfibril angle in the root wood of Pinus radiata and Pinus nigra. International Association of Wood Anatomists Journal 22: 57-62
McCulloh K. A Sperry J. S 2005 Patterns in hydraulic architecture and their implications for transport efficiency. Tree Physiology 25: 257-267[Abstract]
McMahon T. A Kronauer R. E 1976 Tree structure; deducing the principle of mechanical design. Journal of Theoretical Biology 59: 443-466[CrossRef][Web of Science][Medline]
McMillin C. W 1973 Fibril angles of loblolly pine wood as related to specific gravity, growth rate and distance from pith. Wood Science and Technology 7: 251-255[CrossRef][Web of Science]
Mork E 1928 Die Qualität des Fichtenholzes unter besonder Rücksihtnahme auf Schleif- und Papierholz. Papier Fabrikant 26: 741-747
Niklas K. J 1992 Plant biomechanics: an engineering approach to plant form and function University of Chicago Press Chicago, Illinois, USA
Panshin A. J deZeeuw C 1980 Textbook of wood technology, 4th ed McGraw-Hill New York, New York, USA
Polman J. E Michon S. G. L Militz H Helmink A. Th. F 1999 The wood of Metasequoia glyptostroboides (Hu et Cheng) of Dutch origin. Holz als Roh- und Werkstoff 57: 215-221[CrossRef][Web of Science]
Roderick M. L Berry S. L 2001 Linking wood density with tree growth and environment: a theoretical analysis based on the motion of water. New Phytologist 149: 473-485[CrossRef][Web of Science]
Rundel P. W Stecker R. E 1977 Morphological adaptations of tracheid structure to water stress gradients in the crown of Sequoiadendron giganteum. Oecologia 27: 135-139[CrossRef][Web of Science]
Schimleck L Evans R Ilic J 2003 Application of near infrared spectroscopy to the extracted wood of a diverse range of species. IAWA Journal 24: 429-438[Web of Science]
Schniewind A. P 1972 Elastic behavior of the wood fiber. In B. A. Jayne [ed.], Theory and design of wood and fiber composite materials. Syracuse University Press, Syracuse, New York, USA
Singleton R DeBell D. S Gartner B. L 2003 Effect of extraction on wood density of western hemlock (Tsuga heterophylla (Raf.) Sarg). Wood and Fiber Science 35: 363-369[Web of Science]
Sjöström E 1981 Wood chemistry: fundamentals and applications Academic Press New York, New York, USA
Sperry J. S Hacke U. G 2004 Analysis of circular bordered pit function. I. Angiosperm vessels with homogeneous pit membranes. American Journal of Botany 91: 369-385
Taneda H Tateno M 2004 The criteria for biomass partitioning of the current shoot: water transport versus mechanical support. American Journal of Botany 91: 1949-1959
Tyree M. T Zimmermann M. H 2002 Xylem structure and the ascent of sap, 2nd ed Springer Berlin, Germany
U. S. Forest Products Laboratory. 1999 Wood handbook: wood as an engineering material General Technical Report FPL-GTR-113. U.S. Forest Products Laboratory, Madison, Wisconsin, USA
Uggla C Magel E Moritz T Sundberg B 2001 Function and dynamics of auxin and carbohydrates during earlywood/latewood transition in scots pine. Plant Physiology 125: 2029-2039
Uprichard J. M 1971 Cellulose and lignin content in Pinus radiata D. Don: within-tree variations in chemical composition, density and tracheid length. Holzforschung 25: 97-105[Web of Science]
Visscher G. E 2002 Wood anatomy of Metasequoia: separation from Glyptostrobus and function/structure considerations Master's thesis University of Maine, Orono, Maine, USA
Visscher G. E Jagels R 2003 Separation of Metasequoia and Glyptostrobus (Cupressaceae) based on wood anatomy. International Association of Wood Anatomists Journal 24: 439-450
Walker J. C. F Woolons R. C 1998 Cell wall organization and the properties of xylem: a speculative review. In B. G. Butterfield [ed.] Proceedings of International Association of Wood Anatomists/International Union of Forest Research Organization International Workshop, Microfibril angle in wood 13-26 University of Canterbury Christchurch, New Zealand
Wardrop A. B Addo-Ashong F. W 1963 The anatomy and fine structure of wood in relation to its mechanical failure. CSIRO D.F.P. reprint no. 560 169-200
Wardrop A. B Dadswell H. E 1950 The nature of reaction wood. II. The cell wall organization of compression wood tracheids. Australian Journal of Scientific Research, B 3: 1-12
Weibel E. R Bolander R. P 1973 Stereological techniques for electron microscopic morphometry. In M. A. Hayat [ed.] Principles and techniques of electron microscopy, vol. 3, Biological applications Van Nostrand Reinhold New York, New York, USA
Wellwood R. W 1962 Tensile testing of small wood samples. Pulp and Paper Magazine of Canada 63: T61-T67
Woodrum C. L Ewers F. W Telewski F. W 2003 Hydraulic, biomechanical, and anatomical interactions of xylem from five species of Acer (Aceraceae). American Journal of Botany 90: 693-699
Zobel B. J McElwee R. L 1958 Natural variation in wood specific gravity of loblolly pine and an analysis of contributing factors. TAPPI Journal 41: 158-161
Zobel B. J Sprague J. R 1998 Juvenile wood in forest trees Springer-Verlag Berlin, Germany
CiteULike Complore Connotea Del.icio.us Digg Facebook Reddit Technorati Twitter What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
