| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Physiology and Biochemistry |
2Department of Organismic and Evolutionary Biology, Harvard University. Cambridge, Massachusetts 02138 USA; 3Arnold Arboretum, Harvard University, Cambridge, Massachusetts 02138 USA
Received for publication January 13, 2006. Accepted for publication April 18, 2006.
ABSTRACT
The hydraulic resistance of pit membranes was measured directly in earlywood vessels of Fraxinus americana and Ulmus americana. The area-specific resistance of pit membranes (rmem) was higher than modeled or measured values obtained previously for hardwood species, with rmem of 5.24 x 103 MPa·s·m–1 for Fraxinus and 2.56 x 103 MPa·s·m–1 for Ulmus. The calculated resistance of pit canals was three orders of magnitude below total pit resistance indicating that pit membranes contributed the majority of resistance. Scanning electron microscopy indicated that pit membranes of Ulmus were thinner and more porous than those of Fraxinus, consistent with the difference in rmem between the species. Measurements of average vessel diameter and length and area of wall overlap with neighboring vessels were used to partition the vascular resistance between vessel lumen and pit membrane components. Pit membrane resistance accounted for 80% of the total resistance in Fraxinus and 87% in Ulmus in 2-yr-old branch sections. However, measurements of vessel dimensions in the trunk suggest that the division of resistance between pit membrane and lumen components would be closer to co-limiting in older regions of the tree. Thus, pit membrane resistance may be of greater relative importance in small branches than in older regions of mature trees.
Key Words: Fraxinus • resistance partitioning • Ulmus • xylem vessel
The transport of water to substantial heights against the force of gravity at rates sufficient to allow the net uptake of CO2 is a topic that has fascinated botanists for centuries. Plants achieve this feat by utilizing the hydrogen bonding between water molecules to maintain a column of liquid water under tension (Zimmermann, 1983
). However, water under tension is susceptible to spontaneous phase changes from liquid to vapor (cavitation), which results in embolism that can severely limit water transport. To counteract this potentially lethal problem, plants employ a highly compartmentalized transport system of xylem conduits that allows failures in the integrity of the water column to be isolated from functioning conduits. Conduits are separated by partially hydrolyzed primary cell walls known as pit membranes that permit the movement of water from one conduit to the next but limit the movement of gas and pathogens. For pit membranes to perform their protective function, pore sizes must be small; average pore diameters in the pit membranes of dicotyledonous tree species are estimated to be 5–20 nm (Choat et al., 2003
). However, such small pore sizes generate significant hydraulic resistance. Thus, water moving through the xylem encounters two principal resistances, the resistance due to the conduit walls as it moves along the conduit longitudinally (lumen resistance) and the resistance due to pitted walls and pit membranes as it moves laterally between conduits (Fig. 1).
|
), the resistance of pits and pit membranes is more difficult to measure or model because little is known about the fine structure of pit membranes. Indeed, the contribution of pits to vascular resistance has been largely ignored in models describing the movement of water through plants (West et al., 1999
). Currently, empirical data are available from studies comparing lumen and pit membrane resistances using a variety of indirect methods. For example, Schulte and Gibson (1988)
estimated pit membrane resistivity of six tracheid-bearing species by dissolving pit membranes with cellulase and reported that pit membrane resistivity accounted for 14–84% of the total resistivity. Using scaled-up physical models, Lancashire and Ennos (2002)
calculated the resistance of conifer bordered pits at 29% of total resistance for a "typical" tracheid. More recently, values for resistance of angiosperm pit membranes have been provided by models (Sperry and Hacke, 2004
) and measurements on progressively shortened stems (Sperry et al., 2005
; Wheeler et al., 2005
). Sperry et al. (2005)
found that end wall resistivity averaged 54% of total resistivity across a range of xylem structures, and this result was confirmed in a broader range of vessel-bearing species (Wheeler et al., 2005
; Hacke et al., 2006
).
The fact that pit membranes may account for a high proportion of vascular resistance is particularly important in the context of the ion mediated "hydrogel" response attributed to pit membrane polysaccharides. Zwieniecki et al. (2001b)
demonstrated that xylem hydraulic resistance could be significantly altered by changes in the ionic concentration of xylem sap. A high pit membrane resistance would facilitate hydrogel regulation of vascular resistance, allowing plants to respond to short-term environmental variation. This would be particularly advantageous in the complex branching canopies of mature trees in which hydraulic resistance and evaporative demand may vary greatly.
While a range of techniques has been employed to obtain estimates of pit membrane or end wall resistance, pit resistance has not been directly measured. We used the microcapillary technique of Zwieniecki et al. (2001a)
to measure the hydraulic resistance of intervessel pit membranes in two ring porous tree species, Fraxinus americana Marsh. and Ulmus americana L. These species were selected because the large diameters of their earlywood xylem vessels (60–120 µm) facilitate measurements on individual vessels using the microcapillary technique. In addition, the two species provide a contrast in pit structure, as Fraxinus has small pits with thick pit membranes (Sano, 2005
), while Ulmus has large pits with thinner membranes. To place our measurements of pit resistance in their hydraulic context, we compared them with lumen resistance for vessels of average dimensions to determine the relative contribution of each component to the total hydraulic resistance.
MATERIALS AND METHODS
Pit membrane hydraulic resistance
Branches of Fraxinus americana and Ulmus americana were collected from the Harvard University campus, Cambridge, Massachusetts, between June and August of 2004. Branches 1–2 m in length were cut from mature trees using extension shears and returned to the laboratory in sealed plastic bags. In the laboratory, smaller stem segments of 2–4 cm in length were cut from the branches in the 2-yr age zone. The ends of these segments were shaved with a fresh razor blade and flushed with a filtered 10 mmol/L KCl solution at 100 kPa to remove any emboli present in the stem. The transverse section of the cut stem was viewed using a stereomicroscope, and a glass microcapillary was inserted into an earlywood vessel in the outermost growth ring. The vessels selected were directly adjacent to a neighboring vessel in cross section, with which it shared a section of pitted wall. The capillary was glued in place using cyanoacrylic glue (Locite 409, Henkle Corp., Rocky Hill, Connecticut, USA) and a dilute safranin dye solution (0.01% w/v in deionized water) was perfused through the vessel at a low pressure (
50 kPa) to identify the vessel at the distal end of the segment. Measurements made during protocol development indicated that the safranin dye solution did not increase the hydraulic resistance of the pitted wall. The segment was then trimmed with a razor blade from the distal end until a direct connection between the two neighboring vessels could be seen. The vessel into which the microcapillary was glued was distinguished from its connected neighbor by injecting air into the vessel while observing the distal cut end under the stereomicroscope. A second microcapillary was then glued into the neighboring vessel at the distal end. Fluid could then be driven across the section of pitted wall between the neighboring vessels (Fig. 2).
|
P/F, where F is the flow rate of solution onto the balance in m3·s–1 and P is the pressure drop across the segment in MPa. Preliminary experiments indicated that the hydraulic resistance of the glass microcapillaries was negligible compared with the resistance of the pitted walls. The surface area of wall connecting the neighboring vessels was measured after cutting serial transverse sections (30 µm thick) every 120 µm along the stem segment with a sliding microtome. Images of the stained vessels were then collected at 200x magnification and analyzed using ImageJ software (National Institute of Health, Bethesda, Maryland, USA). The perimeter of the pitted wall that connected the two vessels was measured in each section, and the surface area of the pitted wall between each section was calculated as the area of a trapezoid. The total area of pitted wall was calculated as the sum of these sections. The average fraction of pit membrane area per pitted wall area (
) was measured from longitudinal sections through the earlywood xylem vessels of 2-yr-old branches from the same trees (Fig. 2). The surface area of pit membrane connecting the two vessels (Amem, in m2) was calculated as the product of pitted wall area and pit membrane fraction. The area-specific resistance of the pit membrane (rmem, in MPa·s·m–1) was then calculated as rmem = RhAmem. The resistance of an individual pit (Rind) was calculated as Rind = rmem/Aind, where Aind is the surface area of an individual pit membrane. The resistance due to pit canals (Rcan) was calculated from measurements of pit canal length and radius using the Hagan–Poiseuille equation. Four measurements of pit resistance were made for each species, with each replicate coming from a separate mature tree. Pit membrane structure was observed in each species using scanning electron microscopy. Samples branches were collected from the same trees on which pit resistance measurements were undertaken and sections were cut from 2-yr-old zone. Sections were split through the current year growth of earlywood vessels and air dried before being mounted on aluminum stubs with carbon tape. Samples were coated with platinum/palladium for 20 s at 30 mA using a sputter coater, and the sides of the wood samples were coated with colloidal carbon paste. This approach was designed to minimize the degree to which the coating could obscure the fine features of the pit membranes. Samples were observed with a field emission scanning electron microscope (FESEM, FEI, Hillsboro, Oregon, USA) at an accelerating voltage of 2.5 to 3.0 kV.
Vessel dimensions
The mean diameter of earlywood vessels was measured from transverse sections of 2-yr-old branches and trunks of the four trees used in pit resistance measurements. For 2-yr-old branches, sections were taken from branches used for pit resistance measurements, with one section taken from each branch and at least 100 vessels measured in each section. For trunks, cores were taken with an increment borer at breast height and 50–100 vessels were measured in sections taken from these cores.
The mean vessel length for each species was measured by perfusing stem segments with a silicon elastomer (Rhodorsil RTV 141, Rhodia, Cranbury, New Jersey, USA) colored red with a pigment (LSDR-11, Dow Corning, Kendallville, Indiana, USA) following methods outlined in Sperry et al. (2005)
. Branch samples 1–2 m in length were cut from trees and brought to the laboratory. Branches were cut in the 2-yr-old growth region and flushed with filtered deionized water for 15 min at 100 kPa to remove emboli. Branches were then placed into the pressure chamber and injected with the elastomer for 1 h at 0.5 MPa and cured in an oven for 1 h at 70°C (Wheeler et al., 2005
).
Mean vessel length was estimated from the number of vessel endings in a given length of stem. This approach, which is much less labor intensive than determining the entire distribution of vessel lengths, was selected because our goal was to estimate partitioning between end wall and lumen resistance based on measured values of rmem and mean vessel length. Stem segments were cut close to the 2–3-year terminal bud scar and injected in either the acropetal or basipetal direction. The number of vessel endings in a segment was calculated as the number of elastomer-filled vessels at the point of injection minus the number of filled vessels 10 cm from the injection point. We counted only the large earlywood vessels because these were the vessels used in pit conductivity measurements. The mean vessel length was calculated as the length of the segment divided by the fraction of vessel endings within the segment. For each species, three stem segments were injected in the acropetal direction and three were injected in the basipetal direction.
The average area of pitted wall overlap between vessels was measured using methods described by Wheeler et al. (2005)
. The fraction of the vessel length along which two adjacent vessels overlapped was calculated as the ratio of grouped vessels to the total of solitary and grouped vessels (with each cluster of vessels accounting for one group). The ratio L'/L was calculated as: L'/L = 1 – (length fraction)/2; the length fraction is divided by 2 because half of the total overlap per vessel should be in one end of the vessel (Fig. 1). The actual length of overlap between two adjacent vessels (l) was calculated from measured mean vessel lengths as l = L – L'. Measurements were made on the same transverse sections used for measurement of vessel diameter. For each species, four sections from 2-yr-old branches and four sections from trunks were analyzed.
Partitioning of vascular resistance
The hydraulic resistance of vessel lumen and pit membrane components was compared on the basis of average vessel dimensions and the values of rmem obtained experimentally.
The Hagen–Poiseuille equation was used to calculate the resistance of the lumen:
|
| (1) |
is the dynamic viscosity of water at 20°C (1.002 x 10–9 MPa·s) and D is the diameter of the conduit. L' was used for the length of the conduit because, on average, water will travel only this distance as it moves through the vessel from one end wall to the next (Fig. 1): Hagen–Poiseuille flow involves no net movement of water in the transverse plane (Wilcox, 1997
) and a finite element analysis of flow through an oblique perforation plate by Schulte and Castle (1993)
showed deviations from this pattern only in the immediate neighborhood of the plate. The pit membrane resistance (Rmem) was calculated as the area-specific resistance (rmem) divided by the area of pit membrane between "typical" vessels (Amem). The area of membrane between two vessels was calculated as Amem = Dl. In this case, the diameter of the vessel is used because the vessels take on a more elliptical form while connected to each other, with the length of the straight connecting wall equaling the initial diameter of the vessels (Fig. 1). The resistance of the vessel was calculated using the Hagen–Poiseuille formula for a vessel with the same cross-sectional area. More elaborate corrections are possible (Lewis, 1992
) but of little effect when the deviations from a circular vessel cross-section are small, as here (Sperry et al., 2005
). Though a vessel tapers in the region of overlap, in this region an increasing proportion of the flux is carried in the connecting vessel as the resistivity increases, so it was assumed that this would contribute little to the overall resistance. The total resistance (Rtot) of the conduit was then calculated by adding the lumen and pit membrane resistances in series:
|
| (2) |
We also examined how the total resistivity (resistance per unit length Rtot/L) of the conduit varied with vessel length, assuming that the proportion of vessel overlap (lf = l/L) remained constant with the vessel length. In this case, the diameter D, the fraction of pit membrane area per wall area , and the overlap fraction lf were set at the average value for each species, while the vessel length varied. Substituting for Amem and replacing l with lfL then gives
|
| (3) |
|
| (4) |
|
| (5) |
RESULTS
Vascular anatomy
Earlywood vessels from 2-yr-old branches were both longer and wider in Fraxinus than in Ulmus (Table 1). Mean vessel diameter (±SE) was 79.9 (±0.9) µm in Fraxinus and 72.5 (±0.9) µm in Ulmus. In the trunk, mean vessel diameters were more than double that of 2-yr-old branches with 210.4 (±14.2) µm in Fraxinus and 170.8 (±14.0) µm in Ulmus. In this study, the narrow latewood vessels have been ignored for two reasons. First, all measurements of pit membrane resistance were made on earlywood vessels, and there is evidence that pit membrane structure may differ between earlywood and latewood vessels of ring porous trees (Jansen et al., 2004
). Second, up to 95% of flow takes place through earlywood vessels during the growing season; therefore, they are of much greater physiological relevance during this period (Ellmore and Ewers, 1986
).
|
) being greater in Ulmus than in Fraxinus (Table 1).
Pit membrane ultrastructure
Pit membranes of earlywood vessels from both species were homogeneous in structure as is typical for hardwood species (Figs. 3, 4). Although some Ulmus species are known to possess pit membranes with a torus-margo structure, this was not the case in earlywood vessels observed here (Fig. 4). Cellulose microfibrils were randomly arrayed in a tight meshwork in pit membranes of both species. Porosity varied between individual membranes; in some membranes openings of 10–50 nm were apparent, while in others no openings could be detected. In general, pit membranes of Fraxinus were less likely to have detectable openings than pit membranes of Ulmus, and some membranes were encrusted with material that covered the cellulose microfibrils (Fig. 3c, d). Most membranes of Ulmus contained detectable openings within the range of 10–50 nm in diameter indicating that cellulose microfibrils were less densely arrayed than in membranes of Fraxinus (Fig. 4b, c). In some cases, pores were obviously an artifact of preparation because the membrane had been deflected across the chamber and both tearing and enlarged openings were visible (Fig. 4d). Despite evidence of these artifactual increases in membrane pore size, images suggested that membranes of Ulmus were generally more porous and thinner than those of Fraxinus.
|
|
|
|
DISCUSSION
The area-specific hydraulic resistance of pit membranes was a significant component of vascular resistance in earlywood vessels of Fraxinus americana and Ulmus americana, with rmem being 5.32 x 103 (±0.39) MPa·s·m–1 for Fraxinus and 2.56 x 103 (±0.69) MPa·s·m–1 for Ulmus. The fact that pit membranes of Fraxinus had a significantly higher rmem than those of Ulmus indicates that there were intrinsic differences in the structure of the pit membranes between the two species in terms of thickness or porosity. Scanning electron micrographs revealed that pit membranes of Ulmus were generally more porous than those of Fraxinus. This is consistent with the observations of Sano (2004
, 2005
), who noted that pit membranes of Fraxinus mandshurica were thicker than those of three other dicot species he observed. It is also apparent from TEM observations that pit membranes of Fraxinus are thicker than those of Ulmus (B. Choat and S. Jansen [Royal Botanical Gardens, Kew, United Kingdom], unpublished data). If pit membranes of Ulmus are thinner and more porous than those of Fraxinus, this would explain the lower rmem of Ulmus measured in this study. However, it must also be noted that the structure and porosity of pit membranes in dried samples observed with scanning electron microscopy may differ from pit membranes in a hydrated state and thus caution must be used when interpreting data from scanning electron micrographs (Choat et al., 2003
).
Estimates of individual pit resistance were four to five orders of magnitude higher in Fraxinus and Ulmus than values given in previous studies from physical (Lancashire and Ennos, 2002
) and mathematical models (Sperry and Hacke, 2004
). Large differences in Rind could be expected between conifer margo-torus type pit membranes and homogeneous membranes of angiosperms because of the wider pore diameters found in the margo region and the larger size of conifer tracheid pit membranes (Hacke et al., 2004
). The area-specific resistance of pit membranes was also higher than modeled or measured values reported for homogeneous angiosperm pit membranes (Sperry and Hacke, 2004
, Sperry et al., 2005
, Wheeler et al., 2005
). Modeled values of area-specific resistance for homogeneous angiosperm pit membranes presented by Sperry and Hacke (2004)
were two to four orders of magnitude below values measured here in Fraxinus and Ulmus. The area-specific pit resistances obtained empirically by Wheeler et al. (2005)
were higher than modeled values, ranging from 31–626 MPa·s·m–1, but still lower than those measured in the present study. The higher values of rmem measured in Fraxinus and Ulmus may relate to differences in xylem anatomy. With the exception of Vitis vinifera, all of the angiosperm species examined by Wheeler et al. (2005)
were diffuse porous plants with short vessels. Both Fraxinus and Ulmus are ring porous species and have earlywood xylem vessels of much greater lengths than those in diffuse porous trees (Zimmermann and Jeje, 1981
). The higher rmem in these species is permissible because of the higher Rlum due to long vessels, i.e., higher rmem is balanced against higher Rlum. This is supported by data showing that mean area-specific resistance of pit membranes is higher for ring porous species than for diffuse porous species (Hacke et al., 2006
).
Based upon measurements of mean earlywood vessel diameter and length and the average overlap of adjacent xylem vessels, the resistance due to water moving through pit membranes was 80–87% of the total resistance for water flow through 2-yr-old branches. This is at the upper margin of previous estimates, with Schulte and Gibson (1988)
finding a range of 14–84% and Sperry et al. (2005)
finding between 31% and 76%. The higher ratio of Rmem/Rtot found in the present study was mainly attributable to the higher values of rmem measured for Fraxinus and Ulmus. Estimates of Rmem were also influenced by the average length of overlap calculated from solitary and grouped vessels in transverse sections. Given the convoluted nature of vascular networks in angiosperm trees, the representation of vessel overlap (Fig. 1) and the techniques used to estimate them may significantly underestimate the complexity of the actual flow path.
The method used for measuring pit membrane resistance here provides a rough estimate of overlap between two vessels because the stem segment was cut back until two connected vessels could be seen at each end of the segment. The overlap of adjacent vessels was generally between 0.003 and 0.010 m in length, while the mean overlap lengths estimated from transverse sections were 0.073 m and 0.047 m for Fraxinus and Ulmus, respectively. The difference between the observed and estimated lengths of overlap suggests that many lateral connections exist along the length of earlywood vessels. This is supported by reconstructions of the vascular network in Fraxinus and Ulmus species (Zimmermann and Tomlinson, 1967
; Burggraaf, 1972
; Newbanks et al., 1983
; Kitin et al., 2004
).
The contribution of lateral overlaps to water movement between vessels depends on the relative positions of the overlapping vessels in the vertical file. If connections occur between vessels in the same vertical plane, then no pressure gradient will exist between them and therefore little exchange of water will take place. However, if there is enough vertical displacement between vessels with lateral connections, then exchange of water will occur there as well as across the end walls. Unfortunately, no vessel endings were recorded in previous three-dimensional (3-D) reconstructions of Fraxinus species, so lateral overlaps cannot be compared with vessel endings (Burggraaf, 1972
, Kitin et al., 2004
). In a reconstruction of Ulmus americana wood, many vessel endings can be seen as well as lateral connections; both vessel endings and lateral connections run between 1 and 6 mm (Newbanks et al., 1983
, Zimmermann, 1983
). Based on evidence from 3-D reconstructions, the overlap area estimated from transverse sections is likely to more closely represent the total overlap between one vessel along its length and all surrounding vessels rather than simply the end wall connections between two vessels directly above and below one another in a vertical file. Thus, the vessel configuration illustrated in Fig. 1 is a simplified form of the vascular network that does not represent the complexity of the actual network. For the purposes of the analysis in this study, we assumed that all connections between vessels, whether lateral or true end walls, were interfaces at which water is transferred from one vessel to another in the vertical series and that half of these connections would be below the vessel in question and half above. However, it is clear that further detailed analysis of plant vascular networks is required to test this assumption.
As noted in recent studies, when pit membrane and lumen resistances are considered together, resistivity (or conductivity) asymptotes at a certain vessel length and further decreases in resistivity are minimal (Sperry and Hacke, 2004
; Sperry et al., 2005
). Lancashire and Ennos (2002)
demonstrated that when conduit length is constrained, the optimal diameter at which resistance is minimized is reached when pit and lumen resistance are exactly co-limiting. This poses the question as to why vessels lengths are constrained below the point at which resistivity approaches an asymptote. From the perspective of reducing resistance along the xylem pathway, the optimal vessel would be long, wide and have maximum pit membrane overlap area with adjacent vessels. However, lowering resistance of the conduit must be balanced against the increasing vulnerability to embolism. The longer and wider an individual vessel, the greater the proportion of transport capacity that will be lost if the vessel becomes embolized. In addition, increasing surface area of pit membranes has been shown to increase the probability of air seeding occurring between vessels (Choat et al., 2005
; Wheeler et al., 2005
). Wheeler et al. (2005)
reported a significant inverse relationship between pit area per vessel and the tension at which a 50% loss of hydraulic conductivity occurred in 15 angiosperm species. This relationship occurs because the large pit membrane pores responsible for air seeding are apparently rare and the chance of having a pit membrane with a large pore increases with increasing surface area of interface between two vessels (Choat et al., 2004
; Wheeler et al., 2005
). Therefore, increases in conduit dimensions worsens the impact of an embolism on conductivity, while increases in pit membrane interface area increase the probability of water-stress-induced embolism occurring, as well as the probability of embolism propagating through the system.
The analysis on which Fig. 5 is based assumes a constant diameter for xylem vessels as vessel length increases. Average vessel diameters increase in the basipetal direction for most tree species (Zimmermann, 1983
) and vessel diameters in the trunks of Fraxinus and Ulmus were more than double those in 2-yr-old branches. The idea that lumen and end wall resistances may be equal in hardwood species has been discussed in the literature for some time (Zimmermann and Brown, 1971
); Hacke et al. (2006)
reported that, on average, pit and lumen resistances were co-limiting in diffuse and ring porous tree species although there was significant spread around the 1:1 line. Equal partitioning of resistance was not observed in 2-yr-old branches of Fraxinus and Ulmus although it is possible that the resistance is partitioned differently in other regions of the tree. Mean vessel lengths calculated for the trunk, assuming that Rmem/Rlum remained constant, were 2.5 m for Fraxinus and 0.67 m for Ulmus; for lumen and pit membrane resistances to be equal, the vessel lengths calculated from mean diameters and wall overlaps in the trunk were 4.4 m and 1.4 m for Fraxinus and Ulmus, respectively.
Zimmermann and Jeje (1981)
measured vessel lengths in Fraxinus americana trees and found that in the trunk, vessels were commonly 5–10 m in length. Unfortunately, similar vessel length data does not exist for Ulmus, but Newbanks et al. (1983)
, referring to unpublished data, stated that the distribution of vessel lengths in mature elm trees was similar to that of ash and oak. These results indicate that mean vessel lengths in the trunk are likely to be of equal or greater length than those calculated for balanced Rmem/Rlum. Therefore, Rlum is likely to dominate in the trunk, while Rmem will dominate in terminal branches of these species. If Rmem is of greater relative importance in small branches, this suggests that ion-mediated flow enhancement caused by pit membrane hydrogels would have greater impact in small branches than in the trunk of these species (Zwieniecki et al., 2001b
, 2004
). The ability to manipulate hydraulic resistance in the network of small branches would greatly assist plants in responding to highly variable canopy microenvironments as well as optimizing exploitive growth.
FOOTNOTES
1 The authors thank J. Wilson for assistance with measurements and R. Schalek for help with electron microscopy. This research was supported by grants from the Andrew W. Mellon Foundation, the USDA (NRICGP 2001-35100-10615) and the National Science Foundation (IBN 0078155). 
4 Author for correspondence (bchoat{at}ucdavis.edu
); present address: Department of Viticulture and Enology, University of California, Davis, Davis, California 95616 USA; phone: (530) 752-0380; fax: (530) 752-0382
LITERATURE CITED
Burggraaf P. D.. 1972. Some observations on course of vessels in wood of Fraxinus excelsior L. Acta Botanica Neerlandica 21: 32-47.
Choat B. Ball M. Luly J. Holtum J.. 2003. Pit membrane porosity and water stress-induced cavitation in four co-existing dry rainforest tree species. Plant Physiology 131: 41-48.
Choat B. Jansen S. Zwieniecki M. A. Smets E. Holbrook N. M.. 2004. Changes in pit membrane porosity due to deflection and stretching: the role of vestured pits. Journal of Experimental Botany 55: 1569-1575.
Choat B. Lahr E. C. Melcher P. J. Zwieniecki M. A. Holbrook N. M.. 2005. The spatial pattern of air seeding thresholds in mature sugar maple trees. Plant, Cell & Environment 28: 1082-1089.[CrossRef]
Ellmore G. S. Ewers F. W.. 1986. Fluid flow in the outermost xylem increment of a ring-porous tree, Ulmus americana. American Journal of Botany 73: 1771-1774.[CrossRef][ISI]
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. Wheeler J. K. Castro L.. 2006. Scaling of angiosperm xylem structure with safety and efficiency. Tree Physiology 26: 619-701.
Jansen S. Choat B. Vinckier S. Lens F. Schols P. Smets E.. 2004. Intervascular pit membranes with a torus in the wood of Ulmus (Ulmaceae) and related genera. New Phytologist 163: 51-59.[CrossRef][ISI]
Kitin P. B. Fujii T. Abe H. Funada R.. 2004. Anatomy of the vessel network within and between tree rings of Fraxinus lanuginosa (Oleaceae). American Journal of Botany 91: 779-788.
Lancashire J. R. Ennos A. R.. 2002. Modelling the hydrodynamic resistance of bordered pits. Journal of Experimental Botany 53: 1485-1493.
Lewis A. M.. 1992. Measuring the hydraulic diameter of a pore or conduit. American Journal of Botany 79: 1158-1161.[CrossRef][ISI]
Newbanks D. Bosch A. Zimmermann M. H.. 1983. Evidence for xylem dysfunction by embolization in dutch elm disease. Phytopathology 73: 1060-1063.[ISI]
Sano Y.. 2005. Inter- and intraspecific structural variations among intervascular pit membranes as revealed by field-emission scanning electron microscopy. American Journal of Botany 92: 1077-1084.
Sano Y. Z.. 2004. Intervascular pitting across the annual ring boundary in Betula platyphylla var. japonica and Fraxinus mandshurica var. japonica. International Association of Wood Anatomists Journal 25: 129-140.
Schulte P. J. Gibson A. C.. 1988. Hydraulic conductance and tracheid anatomy in 6 species of extant seed plants. Canadian Journal of Botany 66: 1073-1079.[ISI]
Schulte P. J. Castle A. L.. 1993. Water-flow through vessel perforation plates—a fluid mechanical approach. Journal of Experimental Botany 44: 1135-1142.
Sperry J. S. Hacke U. G.. 2004. Analysis of circular bordered pit function. I. Angiosperm vessels with homogenous pit membranes. American Journal of Botany 91: 369-385.
Sperry J. S. Hacke U. G. Wheeler J. K.. 2005. Comparative analysis of end wall resistivity in xylem conduits. Plant, Cell & Environment 28: 456-465.[CrossRef]
West G. B. Brown J. H. Enquist B. J.. 1999. A general model for the structure and allometry of plant vascular systems. Nature 400: 664-667.[CrossRef][ISI]
Wheeler J. K. Sperry J. S. Hacke U. G. Hoang N.. 2005. Inter-vessel pitting and cavitation in woody Rosaceae and other vesselled plants: a basis for a safety versus efficiency trade-off in xylem transport. Plant, Cell & Environment 28: 800-812.[CrossRef]
Wilcox D. C.. 1997. Basic fluid mechanics DCW Industries, La Canada, California, USA.
Zimmermann M. H.. 1983. Xylem structure and the ascent of sap Springer-Verlag, New York, New York, USA.
Zimmermann M. H. Brown C. L.. 1971. Trees: structure and function Springer-Verlag, New York, New York, USA.
Zimmermann M. H. Jeje A. A.. 1981. Vessel length distribution in stems of some American woody plants. Canadian Journal of Botany 59: 1882-1892.[ISI]
Zimmermann M. H. Tomlinson P. B.. 1967. A method for the analysis of the course of vessels. International Association of Wood Anatomists Bulletin 1: 2-7.
Zwieniecki M. A. Melcher P. J. Holbrook N. M.. 2001a. Hydraulic properties of individual xylem vessels of Fraxinus americana. Journal of Experimental Botany 52: 257-264.
Zwieniecki M. A. Melcher P. J. Holbrook N. M.. 2001b. Hydrogel control of xylem hydraulic resistance in plants. Science 291: 1059-1062.
Zwieniecki M. A. Melcher P. J. Feild T. S. Holbrook N. M.. 2004. A potential role for xylem-phloem interactions in the hydraulic architecture of trees: effects of phloem girdling on xylem hydraulic conductance. Tree Physiology 24: 911-917.[ISI][Medline]
This article has been cited by other articles:
|
|
 |
|
  N. Schmitz, S. Jansen, A. Verheyden, J. G. Kairo, H. Beeckman, and N. Koedam Comparative Anatomy of Intervessel Pits in Two Mangrove Species Growing Along a Natural Salinity Gradient in Gazi Bay, Kenya Ann. Bot., August 1, 2007; 100(2): 271 - 281. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Jansen, Y. Sano, B. Choat, D. Rabaey, F. Lens, and R. R. Dute Pit membranes in tracheary elements of Rosaceae and related families: new records of tori and pseudotori Am. J. Botany, April 1, 2007; 94(4): 503 - 514. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
