TY - JOUR
AU - 
AB - Abstract Background and aims Deciduous angiosperm trees transport xylem sap through trunks and branches in vessels within annual growth rings. Utilizing previous growth rings for sap transport could increase vessel network size and redundancy but may expose new xylem to residual air embolisms in the network. Despite the important role of vessel networks in sap transport and drought resistance, our understanding of cross-ring connections within and between species is limited. Methods We studied cross-ring connections in four temperate deciduous trees using dye staining and X-ray microcomputed tomography (microCT) to detect xylem connectivity across growth rings and quantify their impact on hydraulic conductivity. Key results Acer rubrum and Fraxinus americana had cross-ring connections visible in microCT but only A. rubrum used previous growth rings for axial sap flow. Fagus grandifolia and Quercus rubra, however, did not have cross-ring connections. Accounting for the number of growth rings that function for axial transport improved hydraulic conductivity estimates. Conclusions These data suggest that the presence of cross-ring connections may help explain aspects of whole-tree xylem sap transport and should be considered for plant hydraulics measurements in these species and others with similar anatomy. Acer rubrum, connectivity, diffuse porous, earlywood, Fagus grandifolia, Fraxinus americana, growth rings, hydraulics, latewood, Quercus rubra, ring porous, xylem INTRODUCTION Deciduous angiosperm trees transport xylem sap through the trunk and branches in networks of conduits formed in annual growth rings. Within a growth ring, most axial xylem sap transport occurs in vessels because of their large diameter, and therefore lower resistance, relative to fibres and tracheids. Xylem within the trunk is often divided into two categories: the older, non-functional heartwood and the younger, functional sapwood. Sapwood is typically composed of several annual growth rings, with the majority of xylem sap flowing axially through the current-year xylem (Ellmore and Ewers, 1986; Melcher et al., 2003), but the contribution of previously formed growth rings is variable among species and wood types (Phillips et al., 1996; Umebayashi et al., 2008; Berdanier et al., 2016). Partitioning the total axial xylem sap flux into the relative contributions of each growth ring is difficult because of the spatial resolution of standard sap flow sensors compared with the width of annual rings and uncertainty about which rings are being measured (Granier et al., 1994; Phillips et al., 1996; Bush et al., 2010; Berdanier et al., 2016). Studies have addressed this issue with sap flow sensors installed at different depths (Phillips et al., 1996; Gebauer et al., 2008; Bush et al., 2010; Berdanier et al., 2016), dye tracing (Ellmore and Ewers 1986; Umebayashi et al., 2008, 2010) and hydraulics experiments (Melcher et al., 2003; Nolf et al., 2017), all suggesting that in some species xylem sap can flow across annual growth ring boundaries. Recruiting previous growth rings for axial xylem sap transport via cross-ring connections would increase the total number of functional xylem conduits, thereby increasing both network redundancy (Loepfe et al., 2007) and the ‘return’ on the initial investment in past xylem production. In all species, cross-ring connections between growth rings allow older xylem to serve as a large capacitive reservoir to buffer the xylem sap from reaching excessively negative water potentials (Tyree and Zimmermann, 2002; Borchert and Pockman, 2005; Fukuda et al., 2015; Matheny et al., 2015). Cross-ring connections could also expand the xylem vessel network and allow multiple growth rings to contribute to axial flow (Fig. 1A). However, because deciduous leaves are connected to the current-year xylem, the lowest-resistance pathway for axial xylem sap transport would usually be through the current annual ring. Therefore, the previous-year rings likely contribute to axial flow if there is high evaporative demand, if the current-year xylem becomes dysfunctional from embolism or damage (Fig. 1B), or if there are partial or ‘missing’ growth rings (Lorimer et al., 1999). If the current-year xylem becomes dysfunctional, xylem sap would need to move radially from the previous rings into the current-year xylem before it could enter the transpiration stream (Fig. 1B), either through lower-resistance direct intervessel connections or by diffusion through higher-resistance cell walls. Fig. 1. View largeDownload slide Diagrammatic representation of some potential radial pathways for xylem sap and embolism movement between annual rings. In (A), the boundary between latewood (LW) and earlywood (EW) of subsequent years with inter-conduit connections (vertical dashed lines) allows significant contributions of previous growth rings to the axial bulk flow. Radial flow encounters resistance (horizontal resistor symbols), which can be divided into two categories [transverse view of two conduits shown in the inset in (A)]: low-resistance flow via pits (solid arrows) between xylem conduits (circles) and high-resistance flow through the cell walls (dashed arrow). In (B), an embolism (vertical resistor symbol) arising in the current-year ring (2018) blocks axial flow, but radial connectivity allows alternative pathways around the blockage but with increased resistance (dashed arrow in B). In (C), residual embolism in previous year xylem (2017) can follow the same radial inter-conduit pathways via pit membrane connections. In (D), species without direct connections via pits across the growth ring boundary (vertical solid lines) can isolate the spread of residual embolism to current-year xylem but the lack of cross-ring pitting results in very high resistance pathways for axial and radial contributions from older xylem [narrower curved arrows than in (A–C)]. Fig. 1. View largeDownload slide Diagrammatic representation of some potential radial pathways for xylem sap and embolism movement between annual rings. In (A), the boundary between latewood (LW) and earlywood (EW) of subsequent years with inter-conduit connections (vertical dashed lines) allows significant contributions of previous growth rings to the axial bulk flow. Radial flow encounters resistance (horizontal resistor symbols), which can be divided into two categories [transverse view of two conduits shown in the inset in (A)]: low-resistance flow via pits (solid arrows) between xylem conduits (circles) and high-resistance flow through the cell walls (dashed arrow). In (B), an embolism (vertical resistor symbol) arising in the current-year ring (2018) blocks axial flow, but radial connectivity allows alternative pathways around the blockage but with increased resistance (dashed arrow in B). In (C), residual embolism in previous year xylem (2017) can follow the same radial inter-conduit pathways via pit membrane connections. In (D), species without direct connections via pits across the growth ring boundary (vertical solid lines) can isolate the spread of residual embolism to current-year xylem but the lack of cross-ring pitting results in very high resistance pathways for axial and radial contributions from older xylem [narrower curved arrows than in (A–C)]. Embolism formation resulting from exposure to drought or freeze–thaw events (Sperry and Sullivan, 1992) can lead to hydraulic dysfunction. A larger interconnected multi-year xylem network would increase the number of alternative pathways for xylem sap transport (Ellmore et al., 2006). However, an interconnected multi-year xylem network could also be a liability (Loepfe et al., 2007), as residual emboli, unless blocked by tyloses or gums (Cochard and Tyree, 1990), could spread to the current-year xylem through the same cross-ring connections (Fig. 1C). Alternatively, the absence of cross-ring connections could isolate residual embolisms to previous growth rings (Choat et al., 2015), thereby protecting the current-year growth from the legacy effects of past drought or frost (Fig. 1D). It is therefore important to determine the extent of cross-ring connections among species and their potential impact on xylem network organization and sap transport (Umebayashi et al., 2008). Identifying the specific pathway through which xylem sap travels across the latewood–earlywood boundary between two adjacent annual rings is critical for determining the resistance to radial xylem sap movement. Efficient radial flow of xylem sap across growth ring boundaries likely occurs through low-resistance pit membrane connections within pit fields between neighbouring cells (Tyree and Zimmermann, 2002). To our knowledge, direct cross-ring connections via inter-conduit pit fields have been identified and characterized in four species [Fraxinus excelsior (Oleaceae) (Burggraaf, 1972), Machilus thunbergii (Lauraceae) (Fujii et al., 2001), Fraxinus lanuginosa (Oleaceae) (Kitin et al., 2004) and Cryptomeria japonica (Cupressaceae) (Kitin et al., 2009)]. The functional implications of such connections and their prevalence among other woody angiosperms therefore represent a major gap in our knowledge about the relationships between the structure and function of the xylem. Isolating previous growth rings that contain embolisms could also be accomplished by limiting connections between earlywood and latewood within an annual ring rather than across growth ring boundaries (Burggraaf, 1972; Kitin et al., 2004). In this case, cross-ring connections might allow embolism-resistant latewood to hydrate the cambium and developing earlywood vessels of the subsequent year (Kitin and Funada, 2016). However, the higher resistance of the smaller-diameter latewood vessels should contribute little to the overall conductivity once the earlywood of the subsequent year is fully developed. Particularly in ring-porous species, with large-diameter earlywood vessels that are likely to have earlywood vessels in previous growth rings embolized by freeze–thaw cycles (Davis et al., 1999; Utsumi et al., 1999), infrequent connections between the earlywood and latewood within an annual growth ring would limit the risk of embolism propagation to the latewood and subsequent year’s earlywood. Cross-ring connections, and the multi-year functionality of previous growth rings, may also impact measurements of stem vulnerability to embolism spread (Fukuda et al., 2015; Nolf et al., 2017). Including previous growth rings during hydraulics measurements of deciduous species may impact conductivity measurements by including vessels that are not connected to the transpiration stream and are non-functional in planta. For example, if previous growth rings are embolized (as in many ring-porous species after a freeze–thaw cycle; Davis et al., 1999), maximum hydraulic conductivity measurements would be inflated following the standard technique of flushing embolisms. In contrast, in species that contain cross-ring connections it may be important to include the impact of previous growth rings that are still functional on measurements of whole-stem conductivity and resistance to embolism spread (Melcher et al., 2003). It is therefore important to report the number of growth rings included in stem samples and consider other techniques, such as dye or tracer perfusion, to identify functional conduits during hydraulics measurements (Pratt et al., 2018). We searched for cross-ring connections in four northern hardwood species: Acer rubrum, Fagus grandifolia, Fraxinus americana and Quercus rubra. To locate and visualize cross-ring connections, we used dye staining and X-ray microcomputed tomography (microCT) imaging, a technique useful for visualizing complex structures in the xylem (Wason et al., 2017). Finally, to determine how cross-ring connections and existing embolism in previous growth rings may impact vulnerability to embolism spread during drought in each species, we performed measurements to compare xylem vulnerability curves for branches containing only the current-year shoot with branches containing several years of growth. MATERIALS AND METHODS Vessel staining To quantify and characterize cross-ring connections we tested whether dye injected into current-year shoot xylem could pass into previous growth rings. In our first experiment (Supplementary Data Fig. S1), we collected three or four stems per species ~50–70 cm long and containing 2–7 years of growth, on the proximal end of the stem from saplings of two diffuse-porous species (Acer rubrum, Fagus grandifolia) and two ring-porous species (Fraxinus americana, Quercus rubra) at Harvard Forest in Petersham, MA, USA, during August and September 2017. Current-year xylem was exposed by removing 75 % of the length of the current year’s shoot from the distal end of the stem (Supplementary Data Fig. S1). The proximal and distal ends of the stem were both trimmed with a razor blade and fitted with tubing filled with deionized and 0.22-μm filtered water (Millipore Millipak, MilliporeSigma, USA). Native embolisms were removed by flushing stems (Sperry et al., 1988) at 100 kPa from the proximal end containing multiple growth rings for 30–60 min. Following removal of native embolisms, stems were perfused with ~1 % safranin dye solution (using deionized and filtered water, as above) at 100 kPa from the current year’s shoot until dye was visible in the solution exiting the multi-year end for 15 min (Supplementary Data Fig. S1). Perfusion of dye typically took 1–2 h. To minimize dye diffusion before the stems were sectioned, each sample was again flushed with water from the multi-year end and dye-stained stems were stored for 10 d at −80 °C before being sectioned. In the first experiment (Supplementary Data Fig. S1B–D), some cross-ring connections may have been missed because most dye travelled through the low-resistance flow path in the current-year ring rather than crossing growth ring boundaries into previous growth rings (Fig. 1A). Therefore, we verified the results of our first experiment with a related experiment (Supplementary Data Fig. S1E–G) with two stems from each species that would block flow through the current-year ring (as in Fig. 1B). Branches were collected that contained 2 years of growth on the multi-year end and both ends were trimmed, as above. The current-year ring on the multi-year end was sealed with glue (Loctite 409, Loctite, Düsseldorf, Germany) before stems were rehydrated overnight using vacuum infiltration. The experiment then proceeded like the previous dye experiment, by injecting dye from the current-year end at 100 kPa for 45 min followed by flushing with water for 45 min from the multi-year end. Sealing the current-year (outer) ring on the multi-year end and only injecting dye from the current-year end would therefore force dye across growth rings if cross-ring connections were present. If no cross-ring connections were present, almost no dye should flow through the stem. In both experiments, a sliding microtome (Gärtner et al., 2014) was used to collect cross-sections (7–9 μm thick) of each stem sample at ~1 cm below the dye insertion point in the current-year’s growth and at the midpoint of each of the previous years’ growth. Safranin is not soluble in Cytoseal 60 (Richard-Allan Scientific, MI, USA). Therefore, to avoid dye diffusion after sample preparation (Sano et al., 2005), each cross-section was dehydrated in ethanol and mounted in Cytoseal 60. Sections were viewed on an Olympus BX60 compound microscope (Olympus Optical, Tokyo, Japan). Each cross-section was analysed visually to determine which growth rings contained dye-stained vessels and representative images were taken for each species (Canon EOS 6D, Canon, USA). MicroCT imaging of cross-ring connections We used high-resolution microCT, which allows non-destructive 3D imaging of tissues that can be viewed in any orientation, to identify anatomical features of potential cross-ring connections (Brodersen, 2013; Wason et al., 2017). Although pit connections across growth rings may be visible in individual cross-sections using light microscopy, microCT allows detailed visualization of the extent of connections and examination of the 3-D nature of the cells involved. Xylem samples containing multiple growth rings were obtained by collecting one tree core (4.3 mm diameter) extracted at breast height with an increment borer from each of two canopy trees of A. rubrum, Fagus grandifolia, Fraxinus americana and Q. rubra at Harvard Forest. Tree cores were air-dried and dissected with a razor blade to a diameter of ~2.5 mm. Prepared tree core samples were transported to Beamline 8.3.2 at the Lawrence Berkeley National Laboratory Advanced Light Source, Berkeley, CA, USA, for imaging. The outer ~5 mm length of each core was imaged with two or three scans at different positions and 3-D reconstructions were completed following Wason et al. (2017). Final data sets were 3.2 × 3.2 × 2.7 mm with a voxel size of 1.25 μm3. For each species, four to 12 growth rings were captured at this resolution. MicroCT images were also collected at higher magnification, but with a smaller field of view, for more detailed investigations of the anatomy of growth ring boundaries and potential cross-ring connections. A transverse surface of a ~5-cm segment of the remaining sapwood from tree core samples was shaved flat with a microtome and examined under a dissecting microscope. For A. rubrum, three potential intervessel cross-ring connections were identified at low magnification with the dissecting microscope, marked, and trimmed to blocks of about 1.5 × 1.5 × 3 mm. For the remaining three species, ~10–15 growth ring boundaries were searched but no clear pairs of vessels with potential intervessel cross-ring connections were observed using the dissecting microscope. Therefore, for these three species one section of wood per species, containing two vessels near to each other across growth ring boundaries, was trimmed for scanning at high resolution. Each putative cross-ring connection was scanned at ×10 magnification yielding 3-D datasets that were 1.6 × 1.6 × 1.35 mm with a voxel size of 0.625 μm3. These higher-resolution scans served to validate the presence of cross-ring connections observed in the broader search at lower resolution across multiple rings and to provide higher detail of the xylem anatomy of connections. Each microCT dataset was examined for cross-ring connections in Avizo 9.4 software (FEI, Hillsboro, OR, USA) using a combination of 2-D and 3-D visualizations to better understand the nature of cross-ring connections. Xylem vulnerability curves Cross-ring connections and inclusion of multiple growth rings may impact estimates of xylem resistance to embolism spread. Therefore, we constructed percentage loss of conductivity (PLC) curves using the benchtop dry-down method (Choat et al., 2010). For each species, one curve was created using only the current-year shoot from branches (data presented in Wason et al., 2018) and a second curve was created using segments that contained more than one growth ring on both ends of the segment (5.1 ± 2.8 years; grand mean ± s.d.; n = 10–15 per curve; sample details in Supplementary Data Table S1). Branches used for the benchtop dry-down curves were 1–2 m long and were collected predawn in late August and September of 2017 at Harvard Forest. The benchtop dry-down method and a modified Sperry apparatus (Sperry et al., 1988) equipped with a flow meter (SLI-0430; Sensirion, Stäfa, Switzerland) was used to determine native conductivity (Knat) and then maximum conductivity (Kmax) after vacuum infiltration (full details are in Wason et al., 2018). The resistance of the flow meter (inner diameter = 430 μm) was small compared with stem resistance (mean 0.26 % of stem resistance) and was removed during calculations. Final stem samples used for conductivity measurements were 10 ± 3.9 cm long (grand mean ± s.d.; sample details in Supplementary Data Table S1). We calculated PLC as [1 − (Knat/Kmax)] × 100. The PLC curves were fitted for each species and stem segment type (current-year or multi-year) using the Weibull or sigmoidal curves in the fitplc package in R (Duursma and Choat, 2017). In addition, xylem-area-specific native conductivities were calculated for each species using only current-year xylem area or total xylem area based on that species’ ability to use previous-year growth rings, as indicated by microCT and dye-staining experiments, described above. In addition to the benchtop dry-down curves, described above, in late September and early October 2016 we also used the centrifuge method (Alder et al., 1997) to measure vulnerability curves for A. rubrum, Fagus grandifolia and Fraxinus americana (n = 5–6 branches per species). The centrifuge method requires 14-cm-long samples, which are not commonly available with current-year shoots for these species; therefore, centrifuge curves were measured on multi-year stems only. Briefly, multi-year stem segments (5.6 ± 3.8 years old; mean ± s.d.) were vacuum-infiltrated overnight to remove embolisms and Kmax was measured, as described above and in Wason et al. (2018). Next, stems were mounted in a custom centrifuge chamber with reservoirs and sponges on the side walls to prevent exposure to air (Tobin et al., 2013), and stems were spun to a desired water potential. Conductivity was measured on the Sperry apparatus (as described above) and stems were re-spun to successively more negative water potentials. This process was repeated until stems reached 100 % loss of conductivity (−4 MPa). Significant differences between vulnerability curves were determined using bootstrapped 95 % confidence intervals in the fitplc package in R (Duursma and Choat, 2017). RESULTS Dye movement across growth ring boundaries All A. rubrum stems had xylem vessel staining that indicated cross-ring connections where dye moved from the current-year xylem to previous growth rings (Fig. 2A). In some samples of A. rubrum, previous growth rings were almost completely stained and in others only some of the latewood vessels that were nearest to the growth ring boundary were stained. The other diffuse-porous species, Fagus grandifolia, had dye-stained vessels primarily in the earlywood of the current-year growth ring (Fig. 2B). However, some dye did move across the growth ring boundary, but this dye movement was diffuse and not obviously associated with any vessels (Fig. 2B). The two ring-porous species, Fraxinus americana and Q. rubra, had dye-stained earlywood and latewood vessels only within the current year’s growth ring (Fig. 2C, D). The second experiment (Supplementary Data Fig. S1), where the current-year growth ring was sealed on the multi-year end and dye was injected from the current year’s shoot, confirmed these results. In A. rubrum, dye moved across the growth ring boundary, and in the remaining three species dye moved very short distances and only in the current-year ring. Fig. 2. View largeDownload slide Representative cross-sections of dye staining in 2- to 3-year-old stem sections. Stems were perfused with stain solution basipetally from the current-year shoot to determine whether the dye could reach previous growth rings. In each image, the pith and previous growth rings are to the left and the newer rings and epidermis are to the right. Arrows indicate growth ring boundaries. All vessels near the dye injection point in the current-year shoot were stained, and flow appeared to preferentially move through the largest-diameter vessels. Fig. 2. View largeDownload slide Representative cross-sections of dye staining in 2- to 3-year-old stem sections. Stems were perfused with stain solution basipetally from the current-year shoot to determine whether the dye could reach previous growth rings. In each image, the pith and previous growth rings are to the left and the newer rings and epidermis are to the right. Arrows indicate growth ring boundaries. All vessels near the dye injection point in the current-year shoot were stained, and flow appeared to preferentially move through the largest-diameter vessels. Anatomical observations of cross-ring connections MicroCT image analysis of growth ring boundaries supported the dye-staining results. In A. rubrum, 18 cross-ring connections were detected in a total of 13 analysed growth rings (Table 1) in both the higher (0.625 μm3)- and lower (1.25 μm3)-resolution scans. Cross-ring connections in A. rubrum had latewood vessels in the previous-year xylem that were 2.5 times smaller in diameter than the earlywood vessels to which they connect (Table 1, Fig. 3A, B). However, cross-ring connections in A. rubrum were not directly between two vessels, but instead formed by an imperforate tracheary element between the two vessels at the growth ring boundary. These tracheary elements had extensive pit fields (arrow in Fig. 3C) in both periclinal walls, thereby forming a vessel–‘imperforate tracheary element’–vessel connection spanning the growth ring boundary (Figs 3D–F and 4). In Fraxinus americana, 12 cross-ring intervessel connections were detected in eight growth rings analysed with microCT (Table 1). Intervessel cross-ring connections in Fraxinus americana had latewood vessels with a diameter 10 times smaller than the earlywood vessels they connected to across the growth ring boundary (Table 1; Fig. 5A, B) with dense intervessel pitting (Fig. 5C–F). No cross-ring connections were observed in microCT images of Fagus grandifolia (Supplementary Data Fig. S2) or Q. rubra (Table 1). No species had direct cross-ring connections between fibres or tracheids. Table 1. Frequency of cross-ring intervessel connections in xylem of four northern hardwood trees from the north-eastern USA observed using X-ray microCT. Total tangential area searched was the total area of all growth rings that were examined via microCT. Mean earlywood and latewood diameters are reported for vessels connected across growth rings Species Cross-ring intervessel connections Number of rings searched Total tangential area searched (mm2) Earlywood vessel diameter, μm (mean ± s.d.) Latewood vessel diameter μm (mean ± s.d.) Acer rubrum 18 13 40.8 52.8 ± 17.1 21.1 ± 6.7 Fagus grandifolia 0 5 25.2 – – Fraxinus americana 12 8 32.0 227 ± 22.8 21.3 ± 4.9 Quercus rubra 0 7 29.8 – – Species Cross-ring intervessel connections Number of rings searched Total tangential area searched (mm2) Earlywood vessel diameter, μm (mean ± s.d.) Latewood vessel diameter μm (mean ± s.d.) Acer rubrum 18 13 40.8 52.8 ± 17.1 21.1 ± 6.7 Fagus grandifolia 0 5 25.2 – – Fraxinus americana 12 8 32.0 227 ± 22.8 21.3 ± 4.9 Quercus rubra 0 7 29.8 – – View Large Table 1. Frequency of cross-ring intervessel connections in xylem of four northern hardwood trees from the north-eastern USA observed using X-ray microCT. Total tangential area searched was the total area of all growth rings that were examined via microCT. Mean earlywood and latewood diameters are reported for vessels connected across growth rings Species Cross-ring intervessel connections Number of rings searched Total tangential area searched (mm2) Earlywood vessel diameter, μm (mean ± s.d.) Latewood vessel diameter μm (mean ± s.d.) Acer rubrum 18 13 40.8 52.8 ± 17.1 21.1 ± 6.7 Fagus grandifolia 0 5 25.2 – – Fraxinus americana 12 8 32.0 227 ± 22.8 21.3 ± 4.9 Quercus rubra 0 7 29.8 – – Species Cross-ring intervessel connections Number of rings searched Total tangential area searched (mm2) Earlywood vessel diameter, μm (mean ± s.d.) Latewood vessel diameter μm (mean ± s.d.) Acer rubrum 18 13 40.8 52.8 ± 17.1 21.1 ± 6.7 Fagus grandifolia 0 5 25.2 – – Fraxinus americana 12 8 32.0 227 ± 22.8 21.3 ± 4.9 Quercus rubra 0 7 29.8 – – View Large Fig. 3. View largeDownload slide MicroCT scans (0.625 μm3 resolution) of xylem in Acer rubrum showing a cross-ring connection between vessels. (A) Volume rendering of a 0.25 × 0.3 × 0.75-mm block of xylem tissue showing a cross-ring connection (arrow) between the latewood (left) and earlywood vessel (right). The earlywood vessel also connects to a radial file of other earlywood vessels with extensive pitting. (B) Transverse slice from the top of the block showing vessel lumen in black and xylem tissue in white. The locations of the (C) tangential (i, ii) and (D) radial (iii, iv) slices are indicated in (B). (C) Tangential slice approximately parallel to the growth ring boundary showing the pitted vessel wall (arrow) at the growth ring boundary. (D) Radial view showing the extensive pitting on both walls of the imperforate tracheary element between the latewood (left) and earlywood (right) vessel (arrow). (E) Diagram of transverse view (as in B) showing connections across the growth ring boundary (dashed line) between large-diameter earlywood (EW) and narrow-diameter latewood (LW) vessels separated by an imperforate tracheary element (T). (F) Radial view (as in D) depicting potential cross-ring flow (arrows) between latewood (LW) and earlywood (EW) vessels through the tracheary element (T). Scale bars (B–D) = 100 μm. Scale bar in (B) can be used as a reference for the 3-D rendering in (A). Fig. 3. View largeDownload slide MicroCT scans (0.625 μm3 resolution) of xylem in Acer rubrum showing a cross-ring connection between vessels. (A) Volume rendering of a 0.25 × 0.3 × 0.75-mm block of xylem tissue showing a cross-ring connection (arrow) between the latewood (left) and earlywood vessel (right). The earlywood vessel also connects to a radial file of other earlywood vessels with extensive pitting. (B) Transverse slice from the top of the block showing vessel lumen in black and xylem tissue in white. The locations of the (C) tangential (i, ii) and (D) radial (iii, iv) slices are indicated in (B). (C) Tangential slice approximately parallel to the growth ring boundary showing the pitted vessel wall (arrow) at the growth ring boundary. (D) Radial view showing the extensive pitting on both walls of the imperforate tracheary element between the latewood (left) and earlywood (right) vessel (arrow). (E) Diagram of transverse view (as in B) showing connections across the growth ring boundary (dashed line) between large-diameter earlywood (EW) and narrow-diameter latewood (LW) vessels separated by an imperforate tracheary element (T). (F) Radial view (as in D) depicting potential cross-ring flow (arrows) between latewood (LW) and earlywood (EW) vessels through the tracheary element (T). Scale bars (B–D) = 100 μm. Scale bar in (B) can be used as a reference for the 3-D rendering in (A). Fig. 4. View largeDownload slide Example of connections across a growth ring boundary (GR) in Acer rubrum showing imperforate tracheary elements (T) between the latewood (LW) and earlywood (EW) vessels in a transverse section stained with safranin. Fig. 4. View largeDownload slide Example of connections across a growth ring boundary (GR) in Acer rubrum showing imperforate tracheary elements (T) between the latewood (LW) and earlywood (EW) vessels in a transverse section stained with safranin. Fig. 5. View largeDownload slide MicroCT scans (1.25 μm3 resolution) of xylem in Fraxinus americana showing a cross-ring connection between vessels. (A) Overhead and slightly angled view of a volume rendering of the tangential surface of the xylem tissue visible in (B). (B) Transverse slice through the xylem showing vessel lumen in black and xylem tissue in white and the cross-ring connection (arrow) between the latewood (left) and earlywood vessel (right). Note the rough-textured cell wall between the two vessels indicating cross-ring pitting [arrow in (A)]. The locations of the (C) tangential (i, ii) and (D) radial (iii, iv) slices are indicated in (B). (C) Tangential slice parallel to the growth ring boundary showing the pitted vessel wall (arrow) between the latewood and earlywood vessels. (D) Radial view showing the pitting between the latewood and earlywood vessel (arrow). (E) Diagram of transverse view (as in B) showing connections across the growth ring boundary (dashed line) between large-diameter earlywood (EW) and narrow-diameter latewood (LW) vessels. (F) Radial view (as in D) depicting potential cross-ring flow (arrows) between latewood (LW) and earlywood (EW) vessels. Scale bar (B) = 200 μm; (C, D) = 100 μm. Scale bar in (B) can be used as a reference for the 3-D rendering in (A). Note that this scan was collected at lower resolution, and therefore has less detail, than the scan in Fig. 3. Fig. 5. View largeDownload slide MicroCT scans (1.25 μm3 resolution) of xylem in Fraxinus americana showing a cross-ring connection between vessels. (A) Overhead and slightly angled view of a volume rendering of the tangential surface of the xylem tissue visible in (B). (B) Transverse slice through the xylem showing vessel lumen in black and xylem tissue in white and the cross-ring connection (arrow) between the latewood (left) and earlywood vessel (right). Note the rough-textured cell wall between the two vessels indicating cross-ring pitting [arrow in (A)]. The locations of the (C) tangential (i, ii) and (D) radial (iii, iv) slices are indicated in (B). (C) Tangential slice parallel to the growth ring boundary showing the pitted vessel wall (arrow) between the latewood and earlywood vessels. (D) Radial view showing the pitting between the latewood and earlywood vessel (arrow). (E) Diagram of transverse view (as in B) showing connections across the growth ring boundary (dashed line) between large-diameter earlywood (EW) and narrow-diameter latewood (LW) vessels. (F) Radial view (as in D) depicting potential cross-ring flow (arrows) between latewood (LW) and earlywood (EW) vessels. Scale bar (B) = 200 μm; (C, D) = 100 μm. Scale bar in (B) can be used as a reference for the 3-D rendering in (A). Note that this scan was collected at lower resolution, and therefore has less detail, than the scan in Fig. 3. Xylem vulnerability curves We compared PLC curves generated for each species using two different methods. First, the benchtop method was used for current-year stems (reported in Wason et al., 2018) because of the limitations of minimum sample length required for the centrifuge. We then compared these data with benchtop-derived curves for multi-year stems as well as centrifuge PLC curves for multi-year stems. For both A. rubrum and F. grandifolia, benchtop and centrifuge curves generally showed good agreement. Current-year A. rubrum stems had a more negative water potential at 12 % loss of conductivity (P12) than multi-year stems from the centrifuge method, and less negative P88 than multi-year stems from the benchtop method, but otherwise the three curves were not significantly different based on overlapping 95 % confidence intervals (Fig. 6A; Table 2). Current-year Fagus grandifolia stems had similar resistance to embolism spread as multi-year stems using the benchtop method (Fig. 6B, Table 2); however, P50 for multi-year stems from the centrifuge method was 1 MPa more negative than from the benchtop method (Fig. 6B, Table 2). Table 2. Water potentials (MPa) at 12 % (P12), 50 % (P50) and 88 % (P88) loss of conductivity (PLC) for the four study species (with bootstrapped 95 % confidence intervals). Values are reported for benchtop curves conducted on current-year (C-yr) and multi-year (M-yr) stems as well as from centrifuge curves on multi-year stems. Estimates are not available (NA) when the mean lines or 95 % confidence intervals do not cross P12, P50 or P88 within the sampled data range (see Table 2, Fig. 6). Water potentials within a row with overlapping confidence intervals share a letter Species PLC C-yr benchtop M-yr benchtop M-yr centrifuge Acer rubrum P88 −2.9 (−3.0, −2.9)a −4.3 (NA – −3.1)b −3.6 (NA to −2.7)ab P50 −2.7 (−2.9, −2.7)a −3.0 (−3.8, −2.4)a −2.6 (−3.1, −2.3)a P12 −2.5 (−2.8, −2.3)a −1.8 (−3.1, −1.2)ab −1.6 (−2.1, −1.1)b Fagus grandifolia P88 −3.6 (NA, 2.4)a −3.6 (NA, −1.9)a −3.6 (NA, −2.8) A P50 −2.0 (−3.3, −1.5)ab −1.2 (−1.6, −0.6)a −2.2 (−2.6, −1.8)b P12 −0.8 (−1.2, −0.5)a −0.2 (−0.8, NA)a −1.0 (−1.7, −0.7)a Fraxinus americana P88 −3.5 (NA, −2.2) −4.0 (−2.4, NA) −2.1 (NA, −0.6) P50 −1.5 (−2.1, −0.9) NA NA P12 −0.4 (−1.0, NA) NA NA Quercus rubra P88 −5.0 (NA, −2.7) −2.5 (NA, −0.9) − P50 −2.8 (−3.4, −1.7) NA − P12 −1.1 (−2.2, −0.5) NA − Species PLC C-yr benchtop M-yr benchtop M-yr centrifuge Acer rubrum P88 −2.9 (−3.0, −2.9)a −4.3 (NA – −3.1)b −3.6 (NA to −2.7)ab P50 −2.7 (−2.9, −2.7)a −3.0 (−3.8, −2.4)a −2.6 (−3.1, −2.3)a P12 −2.5 (−2.8, −2.3)a −1.8 (−3.1, −1.2)ab −1.6 (−2.1, −1.1)b Fagus grandifolia P88 −3.6 (NA, 2.4)a −3.6 (NA, −1.9)a −3.6 (NA, −2.8) A P50 −2.0 (−3.3, −1.5)ab −1.2 (−1.6, −0.6)a −2.2 (−2.6, −1.8)b P12 −0.8 (−1.2, −0.5)a −0.2 (−0.8, NA)a −1.0 (−1.7, −0.7)a Fraxinus americana P88 −3.5 (NA, −2.2) −4.0 (−2.4, NA) −2.1 (NA, −0.6) P50 −1.5 (−2.1, −0.9) NA NA P12 −0.4 (−1.0, NA) NA NA Quercus rubra P88 −5.0 (NA, −2.7) −2.5 (NA, −0.9) − P50 −2.8 (−3.4, −1.7) NA − P12 −1.1 (−2.2, −0.5) NA − View Large Table 2. Water potentials (MPa) at 12 % (P12), 50 % (P50) and 88 % (P88) loss of conductivity (PLC) for the four study species (with bootstrapped 95 % confidence intervals). Values are reported for benchtop curves conducted on current-year (C-yr) and multi-year (M-yr) stems as well as from centrifuge curves on multi-year stems. Estimates are not available (NA) when the mean lines or 95 % confidence intervals do not cross P12, P50 or P88 within the sampled data range (see Table 2, Fig. 6). Water potentials within a row with overlapping confidence intervals share a letter Species PLC C-yr benchtop M-yr benchtop M-yr centrifuge Acer rubrum P88 −2.9 (−3.0, −2.9)a −4.3 (NA – −3.1)b −3.6 (NA to −2.7)ab P50 −2.7 (−2.9, −2.7)a −3.0 (−3.8, −2.4)a −2.6 (−3.1, −2.3)a P12 −2.5 (−2.8, −2.3)a −1.8 (−3.1, −1.2)ab −1.6 (−2.1, −1.1)b Fagus grandifolia P88 −3.6 (NA, 2.4)a −3.6 (NA, −1.9)a −3.6 (NA, −2.8) A P50 −2.0 (−3.3, −1.5)ab −1.2 (−1.6, −0.6)a −2.2 (−2.6, −1.8)b P12 −0.8 (−1.2, −0.5)a −0.2 (−0.8, NA)a −1.0 (−1.7, −0.7)a Fraxinus americana P88 −3.5 (NA, −2.2) −4.0 (−2.4, NA) −2.1 (NA, −0.6) P50 −1.5 (−2.1, −0.9) NA NA P12 −0.4 (−1.0, NA) NA NA Quercus rubra P88 −5.0 (NA, −2.7) −2.5 (NA, −0.9) − P50 −2.8 (−3.4, −1.7) NA − P12 −1.1 (−2.2, −0.5) NA − Species PLC C-yr benchtop M-yr benchtop M-yr centrifuge Acer rubrum P88 −2.9 (−3.0, −2.9)a −4.3 (NA – −3.1)b −3.6 (NA to −2.7)ab P50 −2.7 (−2.9, −2.7)a −3.0 (−3.8, −2.4)a −2.6 (−3.1, −2.3)a P12 −2.5 (−2.8, −2.3)a −1.8 (−3.1, −1.2)ab −1.6 (−2.1, −1.1)b Fagus grandifolia P88 −3.6 (NA, 2.4)a −3.6 (NA, −1.9)a −3.6 (NA, −2.8) A P50 −2.0 (−3.3, −1.5)ab −1.2 (−1.6, −0.6)a −2.2 (−2.6, −1.8)b P12 −0.8 (−1.2, −0.5)a −0.2 (−0.8, NA)a −1.0 (−1.7, −0.7)a Fraxinus americana P88 −3.5 (NA, −2.2) −4.0 (−2.4, NA) −2.1 (NA, −0.6) P50 −1.5 (−2.1, −0.9) NA NA P12 −0.4 (−1.0, NA) NA NA Quercus rubra P88 −5.0 (NA, −2.7) −2.5 (NA, −0.9) − P50 −2.8 (−3.4, −1.7) NA − P12 −1.1 (−2.2, −0.5) NA − View Large Fig. 6. View largeDownload slide Xylem vulnerability curves for current year’s growth only (C-yr; as reported in Wason et al., 2018) or stems containing multiple growth rings (M-yr). Curves were created using either the benchtop dry-down method (bench) or the centrifuge method (cent.). Centrifuge curves were generated from five or six stems per species and each point represents the mean ± one standard error. All curves are modelled with Weibull functions except the multi-year curves in (C) and (D), which would not converge and are therefore modelled with sigmoidal functions. Fig. 6. View largeDownload slide Xylem vulnerability curves for current year’s growth only (C-yr; as reported in Wason et al., 2018) or stems containing multiple growth rings (M-yr). Curves were created using either the benchtop dry-down method (bench) or the centrifuge method (cent.). Centrifuge curves were generated from five or six stems per species and each point represents the mean ± one standard error. All curves are modelled with Weibull functions except the multi-year curves in (C) and (D), which would not converge and are therefore modelled with sigmoidal functions. In the long-vesselled species Q. rubra and F. Americana, multi-year PLC curves did not match current-year PLC curves. For both species, the centrifuge method suggested that stems were highly vulnerable compared with the current-year benchtop curves. Current-year Fraxinus americana stems were more resistant to embolism spread than the apparently very vulnerable multi-year stems, but all three curves converged at >90 % loss of conductivity at water potentials below −1.8 MPa (Fig. 6C, Table 2). Current-year Q. rubra stems were more resistant to embolism spread than the apparently very vulnerable multi-year stems at water potentials down to −2.5 MPa (Fig. 6D, Table 2). Dye-staining and microCT data suggest that A. rubrum can use more than one growth ring for axial transport, whereas Fagus grandifolia, Fraxinus americana and Q. rubra use only the current-year growth ring. Therefore, we standardized native conductivity by the xylem area used for axial transport in each species. Generally, xylem-specific conductivity of multi-year stems aligned with current-year stems (Fig. 7A–D) except for Fagus grandifolia and Q. rubra, where conductivities were slightly higher than expected from current-year stems (Fig. 7B, D). Fig. 7. View largeDownload slide Native xylem-specific conductivity of stems containing only current-year (C-yr) xylem or multiple years (M-yr) of xylem. Conductivities were normalized by the xylem area of functional growth rings based on dye staining and microCT analysis: total xylem cross-sectional area for (A) Acer rubrum, and current-year xylem cross-sectional area for (B) Fagus grandifolia, (C) Fraxinus americana, and (D) Quercus rubra. Note: y-axis scaling differs between top and bottom panels. Fig. 7. View largeDownload slide Native xylem-specific conductivity of stems containing only current-year (C-yr) xylem or multiple years (M-yr) of xylem. Conductivities were normalized by the xylem area of functional growth rings based on dye staining and microCT analysis: total xylem cross-sectional area for (A) Acer rubrum, and current-year xylem cross-sectional area for (B) Fagus grandifolia, (C) Fraxinus americana, and (D) Quercus rubra. Note: y-axis scaling differs between top and bottom panels. DISCUSSION Presence of cross-ring connections Cross-ring connections observed in A. rubrum with dye staining and microCT are consistent with dye-staining, sap flow and hydraulics experiments on other Acer spp. (Braun, 1970; Chaney and Kozlowski, 1977; Melcher et al., 2003; Umebayashi et al., 2008) and other diffuse-porous species (Umebayashi et al., 2008; Berdanier et al., 2016). Rather than direct intervessel connections, as observed in the only other diffuse-porous species studied in this detail [Machilus thunbergii (Lauraceae); Fujii et al., 2001], cross-ring connections in A. rubrum were characterized by imperforate tracheary elements at the growth ring boundary with extensive pitting on both periclinal walls (Figs 3 and 4). Latewood vessels were narrower than earlywood vessels across the growth ring boundary and dye did not stain every vessel in previous growth rings, suggesting that cross-ring sap flow has a higher resistance than axial sap flow. This finding supports previous data showing that 30–50 % of the axial flow in Acer sp. occurs in previous growth rings (Melcher et al., 2003; Spicer and Holbrook, 2005). We found no evidence of cross-ring connections in the other diffuse-porous species, Fagus grandifolia. The growth ring boundary was composed of parenchyma cells and narrow fibres with sporadic pitting on radial (anticlinal) walls only. We also found little to no staining in the latewood of the current-year ring, similar to Fagus japonica (Umebayashi et al., 2008), suggesting that Fagus spp. latewood may be a high-resistance pathway with a limited role in axial sap transport. The lack of cross-ring connections but apparent diffusion across the growth ring boundary suggests that previous growth rings may be accessible via higher-resistance pathways, such as flow through cell walls or extracellular spaces. Although most deciduous ring-porous species primarily use current-year growth rings for axial xylem sap transport (Chaney and Kozlowski, 1977; Ellmore and Ewers, 1986; Berdanier et al., 2016), connections have been observed in Fraxinus species (Braun, 1970; Burggraaf, 1972; Kitin et al., 2004). We found 12 cross-ring intervessel connections in the ring-porous species Fraxinus americana (Fig. 5), as observed in the similar species, Fraxinus lanuginosa (Kitin et al., 2004). Despite these connections, dye did not flow across growth ring boundaries (Fig. 2C), as also found by Chaney and Kozlowski (1977), likely because the much higher resistance in narrow latewood vessels (and occasional latewood vascular tracheids) compared with the wide earlywood vessels to which they connect. Therefore, latewood conduits in previous rings of Fraxinus spp. are unlikely to serve as a redundant axial flow pathway, but instead may keep the cambium and developing leaves hydrated during earlywood development (Kudo et al., 2015; Kitin and Funada, 2016). In the other ring-porous species, Q. rubra, we found no latewood vessels that paralleled earlywood vessels across a growth ring boundary and no evidence of any cross-ring connections. Although other Quercus spp. may have cross-ring connections via vasicentric tracheids (Sano et al., 2011), we found no evidence of dye stain crossing growth ring boundaries in Q. rubra. This suggests that Q. rubra may rely almost completely on the current-year ring for axial transport. Our results support previous findings (Spicer and Holbrook, 2005; Umebayashi et al., 2008), indicating that in temperate deciduous ring-porous species such as Q. rubra most sap flow occurs in outermost regions of the sapwood. Significance of cross-ring connections Cross-ring connections in A. rubrum and other Acer spp., as well as the ability to use previous growth rings for xylem sap transport, may explain missing growth rings in Acer spp. (Lorimer et al., 1999) and why Acer spp. show evidence of vulnerability segmentation (Choat et al., 2005; Lens et al., 2011; Brodersen, 2016; Gleason et al., 2016; Wason et al., 2018). Acer rubrum has more resistant pit membranes in multi-year branches than the other four species (Wason et al., 2018), perhaps to limit embolism spread in xylem that will be used for multiple years. For A. rubrum, vulnerability curves were nearly identical for stems with current-year xylem only and stems with multiple rings (Figs 6 and 7). This contrasts with studies suggesting that resistance to air seeding declined in older growth rings (Melcher et al., 2003) or in the innermost ring containing primary xylem (Fukuda et al., 2015), but supports other research showing no decline in resistance to air seeding for A. rubrum across three growth rings (Wason et al., 2018). As in A. rubrum, most vessels in Fagus grandifolia are below the 44-μm diameter threshold (Wason et al., 2018) above which embolism induced by freeze–thaw events can be problematic (Davis et al., 1999). Therefore, previous growth rings in these diffuse-porous species may stay hydrated for many years, as found for the related species Acer palmatum and Fagus japonica (Umebayashi et al., 2008). Hydrated previous growth rings in Fagus grandifolia may serve as a capacitor and buffer the effects of drought. Despite the lack of cross-ring connections in Fagus grandifolia, vulnerability curves of current-year and multi-year stems were similar, suggesting that the resistance to embolism spread in these rings had not declined. Furthermore, normalizing native conductivity by current-year xylem area overestimated xylem specific conductivity, suggesting that previous growth rings were still hydrated and transported water when exposed on both ends of the segment, although they would not directly connect to the transpiration stream in planta. In Fraxinus americana and Q. rubra, earlywood vessels are large enough to experience freeze–thaw embolisms (Wason et al., 2018), while most latewood vessels are not (Davis et al., 1999; Umebayashi et al., 2008). Furthermore, many earlywood vessels in previous-year growth rings of the ring-porous species Q. rubra and F. americana can be blocked by tyloses (Meyer and Côté, 1968; Cochard and Tyree, 1990; Micco et al., 2016). Indeed, we found that normalizing native conductivity by current-year xylem area better estimated xylem-specific conductivity for Fraxinus americana (Fig. 7C). Normalizing conductivity of multi-year stems to current-year xylem area for Q. rubra led to slight overestimation of conductivity (Fig. 7D), perhaps related to vessel size or density differences between the current-year ring and previous growth rings that contain primary growth with fewer and smaller vessels per area (Fig. 2D; Wason et al., 2018). With residual embolisms in earlywood of previous growth rings, cross-ring connections could provide a pathway for embolisms to spread into current-year xylem. Despite having cross-ring intervessel connections, Fraxinus spp. have a reduced risk of embolism spread by limiting earlywood-to-latewood vessel connections within a ring (Burggraaf, 1972; Kitin et al., 2004). Indeed, in one microCT sample we found that across four growth rings and 44 earlywood vessels, only two connected to latewood vessels within the same ring whereas four connected to latewood vessels across a ring boundary. In Q. rubra, on the other hand, the absence of any cross-ring connections further reduces the risk of residual embolism spreading into the current-year growth ring. Importantly, the lack of dye spreading into previous growth rings of Fraxinus americana, Q. rubra and Fagus grandifolia further supports the ephemeral nature of each growth ring for axial xylem sap transport and may explain the lack of vulnerability segmentation in these species (Wason et al., 2018) and strong segmentation of Acer spp. (Choat et al., 2005; Johnson et al., 2016; Wason et al., 2018). SUPPLEMENTARY DATA Supplementary data are available online at https://academic.oup.com/aob and consist of the following. Figure S1: schematic diagram of the two dye-staining experiments. Figure S2: microCT images of a growth ring boundary in Fagus grandifolia. Table S1: stem characteristics of samples used in hydraulics measurements. FUNDING This work was supported by the National Science Foundation (IOS-1557917). J.W.W. was partially supported by the USDA National Institute of Food and Agriculture, McIntire Stennis Project Number ME0-42001 through the Maine Agricultural and Forest Experiment Station. ACKNOWLEDGEMENTS We thank Harvard Forest and D. Parkinson for facilitating this research, as well as K. Anstreicher and A. Rubenstein for help with data collection. This research used resources of the Advanced Light Source, which is a DOE Office of Science User Facility under contract number DE-AC02-05CH11231. LITERATURE CITED Alder NN , Pockman WT , Sperry JS , Nuismer S . 1997 . Use of centrifugal force in the study of xylem cavitation . Journal of Experimental Botany 48 : 665 – 674 . Google Scholar Crossref Search ADS WorldCat Berdanier AB , Miniat CF , Clark JS . 2016 . Predictive models for radial sap flux variation in coniferous, diffuse-porous and ring-porous temperate trees . Tree Physiology 36 : 932 – 941 . Google Scholar Crossref Search ADS PubMed WorldCat Borchert R , Pockman WT . 2005 . Water storage capacitance and xylem tension in isolated branches of temperate and tropical trees . Tree Physiology 25 : 457 – 466 . Google Scholar Crossref Search ADS PubMed WorldCat Braun HJ . 1970 . Funktionelle Histologie der sekundären Sprossachse . Berlin : Gebrüder Borntraeger . Google Preview WorldCat COPAC Brodersen CR . 2013 . Visualizing wood anatomy in three dimensions with high-resolution X-ray micro-tomography (μCT) – a review . IAWA Journal 34 : 408 – 424 . Google Scholar Crossref Search ADS WorldCat Brodersen CR . 2016 . Finding support for theoretical tradeoffs in xylem structure and function . New Phytologist 209 : 8 – 10 . Google Scholar Crossref Search ADS PubMed WorldCat Burggraaf PD . 1972 . Some observations on the course of the vessels in the wood of Fraxinus excelsior L . Acta Botanica Neerlandica 21 : 32 – 47 . Google Scholar Crossref Search ADS WorldCat Bush SE , Hultine KR , Sperry JS , Ehleringer JR . 2010 . Calibration of thermal dissipation sap flow probes for ring- and diffuse-porous trees . Tree Physiology 30 : 1545 – 1554 . Google Scholar Crossref Search ADS PubMed WorldCat Chaney WR , Kozlowski TT . 1977 . Patterns of water movement in intact and excised stems of Fraxinus americana and Acer saccharum seedlings . Annals of Botany 41 : 1093 – 1100 . Google Scholar Crossref Search ADS WorldCat Choat B , Lahr EC , Melcher PJ , Zwieniecki MA , Holbrook NM . 2005 . The spatial pattern of air seeding thresholds in mature sugar maple trees . Plant, Cell & Environment 28 : 1082 – 1089 . Google Scholar Crossref Search ADS WorldCat Choat B , Drayton WM , Brodersen C , et al. 2010 . Measurement of vulnerability to water stress-induced cavitation in grapevine: a comparison of four techniques applied to a long-vesseled species . Plant, Cell & Environment 33 : 1502 – 1512 . Google Scholar PubMed WorldCat Choat B , Brodersen CR , McElrone AJ . 2015 . Synchrotron X-ray microtomography of xylem embolism in Sequoia sempervirens saplings during cycles of drought and recovery . New Phytologist 205 : 1095 – 1105 . Google Scholar Crossref Search ADS PubMed WorldCat Cochard H , Tyree MT . 1990 . Xylem dysfunction in Quercus: vessel sizes, tyloses, cavitation and seasonal changes in embolism . Tree Physiology 6 : 393 – 407 . Google Scholar Crossref Search ADS PubMed WorldCat Davis SD , Sperry JS , Hacke UG . 1999 . The relationship between xylem conduit diameter and cavitation caused by freezing . American Journal of Botany 86 : 1367 – 1372 . Google Scholar Crossref Search ADS PubMed WorldCat Duursma R , Choat B . 2017 . fitplc – an R package to fit hydraulic vulnerability curves . Journal of Plant Hydraulics 4 : e-002 . Google Scholar Crossref Search ADS WorldCat Ellmore GS , Ewers FW . 1986 . Fluid flow in the outermost xylem increment of a ring-porous tree, Ulmus americana . American Journal of Botany 73 : 1771 – 1774 . Google Scholar Crossref Search ADS WorldCat Ellmore GS , Zanne AE , Orians CM . 2006 . Comparative sectoriality in temperate hardwoods: hydraulics and xylem anatomy . Botanical Journal of the Linnean Society 150 : 61 – 71 . Google Scholar Crossref Search ADS WorldCat Fujii T , Lee S-J , Kuroda N , Suzuki Y . 2001 . Conductive function of intervessel pits through a growth ring boundary of Machilus thunbergii . IAWA Journal 22 : 1 – 14 . Google Scholar Crossref Search ADS WorldCat Fukuda K , Kawaguchi D , Aihara T , et al. 2015 . Vulnerability to cavitation differs between current-year and older xylem: non-destructive observation with a compact magnetic resonance imaging system of two deciduous diffuse-porous species . Plant, Cell & Environment 38 : 2508 – 2518 . Google Scholar Crossref Search ADS PubMed WorldCat Gärtner H , Lucchinetti S , Schweingruber FH . 2014 . New perspectives for wood anatomical analysis in dendrosciences: the GSL1-microtome . Dendrochronologia 32 : 47 – 51 . Google Scholar Crossref Search ADS WorldCat Gebauer T , Horna V , Leuschner C . 2008 . Variability in radial sap flux density patterns and sapwood area among seven co-occurring temperate broad-leaved tree species . Tree Physiology 28 : 1821 – 1830 . Google Scholar Crossref Search ADS PubMed WorldCat Gleason SM , Westoby M , Jansen S , et al. 2016 . Weak tradeoff between xylem safety and xylem-specific hydraulic efficiency across the world’s woody plant species . New Phytologist 209 : 123 – 136 . Google Scholar Crossref Search ADS PubMed WorldCat Granier A , Anfodillo T , Sabatti M , et al. 1994 . Axial and radial water flow in the trunks of oak trees: a quantitative and qualitative analysis . Tree Physiology 14 : 1383 – 1396 . Google Scholar Crossref Search ADS PubMed WorldCat Johnson DM , Wortemann R , McCulloh KA , et al. 2016 . A test of the hydraulic vulnerability segmentation hypothesis in angiosperm and conifer tree species . Tree Physiology 36 : 983 – 993 . Google Scholar Crossref Search ADS PubMed WorldCat Kitin P , Funada R . 2016 . Earlywood vessels in ring-porous trees become functional for water transport after bud burst and before the maturation of the current-year leaves . IAWA Journal 37 : 315 – 331 . Google Scholar Crossref Search ADS WorldCat Kitin PB , 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 . Google Scholar Crossref Search ADS PubMed WorldCat Kitin P , Fujii T , Abe H , Takata K . 2009 . Anatomical features that facilitate radial flow across growth rings and from xylem to cambium in Cryptomeria japonica . Annals of Botany 103 : 1145 – 1157 . Google Scholar Crossref Search ADS PubMed WorldCat Kudo K , Yasue K , Hosoo Y , Funada R . 2015 . Relationship between formation of earlywood vessels and leaf phenology in two ring-porous hardwoods, Quercus serrata and Robinia pseudoacacia, in early spring . Journal of Wood Science 61 : 455 – 464 . Google Scholar Crossref Search ADS WorldCat Lens F , Sperry JS , Christman MA , Choat B , Rabaey D , Jansen S . 2011 . Testing hypotheses that link wood anatomy to cavitation resistance and hydraulic conductivity in the genus Acer . New Phytologist 190 : 709 – 723 . Google Scholar Crossref Search ADS PubMed WorldCat Loepfe L , Martinez-Vilalta J , Pinol J , Mencuccini M . 2007 . The relevance of xylem network structure for plant hydraulic efficiency and safety . Journal of Theoretical Biology 247 : 788 – 803 . Google Scholar Crossref Search ADS PubMed WorldCat Lorimer CG , Dahir SE , Singer MT . 1999 . Frequency of partial and missing rings in Acer saccharum in relation to canopy position and growth rate . Plant Ecology 143 : 189 – 202 . Google Scholar Crossref Search ADS WorldCat Matheny AM , Bohrer G , Garrity SR , Morin TH , Howard CJ , Vogel CS . 2015 . Observations of stem water storage in trees of opposing hydraulic strategies . Ecosphere 6 : art165 . Google Scholar Crossref Search ADS WorldCat Melcher PJ , Zwieniecki MA , Holbrook NM . 2003 . Vulnerability of xylem vessels to cavitation in sugar maple. Scaling from individual vessels to whole branches . Plant Physiology 131 : 1775 – 1780 . Google Scholar Crossref Search ADS PubMed WorldCat Meyer RW , Côté WA . 1968 . Formation of the protective layer and its role in tylosis development . Wood Science and Technology 2 : 84 – 94 . Google Scholar Crossref Search ADS WorldCat Micco VD , Balzano A , Wheeler EA , Baas P . 2016 . Tyloses and gums: a review of structure, function and occurrence of vessel occlusions . IAWA Journal 37 : 186 – 205 . Google Scholar Crossref Search ADS WorldCat Nolf M , Lopez R , Peters JMR , et al. 2017 . Visualization of xylem embolism by X-ray microtomography: a direct test against hydraulic measurements . New Phytologist 214 : 890 – 898 . Google Scholar Crossref Search ADS PubMed WorldCat Phillips N , Oren R , Zimmermann R . 1996 . Radial patterns of xylem sap flow in non-, diffuse- and ring-porous tree species . Plant, Cell & Environment 19 : 983 – 990 . Google Scholar Crossref Search ADS WorldCat Pratt RB , Jacobsen AL , Meinzer F . 2018 . Identifying which conduits are moving water in woody plants: a new HRCT-based method . Tree Physiology 38 : 1200 – 1212 . Google Scholar Crossref Search ADS PubMed WorldCat Sano Y , Okamura Y , Utsumi Y . 2005 . Visualizing water-conduction pathways of living trees: selection of dyes and tissue preparation methods . Tree Physiology 25 : 269 – 275 . Google Scholar Crossref Search ADS PubMed WorldCat Sano Y , Morris H , Shimada H , Ronse De Craene LP , Jansen S . 2011 . Anatomical features associated with water transport in imperforate tracheary elements of vessel-bearing angiosperms . Annals of Botany 107 : 953 – 964 . Google Scholar Crossref Search ADS PubMed WorldCat Sperry JS , Sullivan JEM . 1992 . Xylem embolism in response to freeze-thaw cycles and water stress in ring-porous, diffuse-porous, and conifer species . Plant Physiology 100 : 605 – 613 . Google Scholar Crossref Search ADS PubMed WorldCat Sperry JS , Donnelly JR , Tyree MT . 1988 . A method for measuring hydraulic conductivity and embolism in xylem . Plant, Cell & Environment 11 : 35 – 40 . Google Scholar Crossref Search ADS WorldCat Spicer R , Holbrook NM . 2005 . Within-stem oxygen concentration and sap flow in four temperate tree species: does long-lived xylem parenchyma experience hypoxia? Plant, Cell & Environment 28 : 192 – 201 . Google Scholar Crossref Search ADS WorldCat Tobin MF , Pratt RB , Jacobsen AL , De Guzman ME . 2013 . Xylem vulnerability to cavitation can be accurately characterised in species with long vessels using a centrifuge method . Plant Biology 15 : 496 – 504 . Google Scholar Crossref Search ADS PubMed WorldCat Tyree MT , Zimmermann MH . 2002 . Xylem structure and the ascent of sap . Berlin : Springer . Google Preview WorldCat COPAC Umebayashi T , Utsumi Y , Koga S , et al. 2008 . Conducting pathways in north temperate deciduous broadleaved trees . IAWA Journal 29 : 247 – 263 . Google Scholar Crossref Search ADS WorldCat Umebayashi T , Utsumi Y , Koga S , et al. 2010 . Xylem water-conducting patterns of 34 broadleaved evergreen trees in southern Japan . Trees 24 : 571 – 583 . Google Scholar Crossref Search ADS WorldCat Utsumi Y , Sano Y , Funada R , Fujikawa S , Ohtani J . 1999 . The progression of cavitation in earlywood vessels of Fraxinus mandshurica var japonica during freezing and thawing . Plant Physiology 121 : 897 – 904 . Google Scholar Crossref Search ADS PubMed WorldCat Wason JW , Huggett BA , Brodersen CR . 2017 . MicroCT imaging as a tool to study vessel endings in situ . American Journal of Botany 104 : 1424 – 1430 . Google Scholar Crossref Search ADS PubMed WorldCat Wason JW , Anstreicher KS , Stephansky N , Huggett BA , Brodersen CR . 2018 . Hydraulic safety margins and air-seeding thresholds in roots, trunks, branches and petioles of four northern hardwood trees . New Phytologist 219 : 77 – 88 . Google Scholar Crossref Search ADS PubMed WorldCat © The Author(s) 2019. Published by Oxford University Press on behalf of the Annals of Botany Company. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - The functional implications of tracheary connections across growth rings in four northern hardwood trees
JO - Annals of Botany
DO - 10.1093/aob/mcz076
DA - 2019-09-24
UR - https://www.deepdyve.com/lp/oxford-university-press/the-functional-implications-of-tracheary-connections-across-growth-5K67cEdSoQ
SP - 1
VL - Advance Article
IS - 
DP - DeepDyve
ER -