IAWA Journal, Vol. 31 (1), 2010: 53–66

TORUS-BEARING PIT MEMBRANES IN CERCOCARPUS

Roland Dute1,*, Jaynesh Patel1 and Steven Jansen2,3

SUMMARY Intervascular pit membranes of Cercocarpus possess torus thickenings. The thickenings, or pads, consist of lignified, secondary wall material. Torus pad deposition occurs late in cell ontogeny and is not associated with a microtubule plexus. Half-bordered pit pairs between tracheary elements and parenchyma cells often have a torus pad on the membrane surface facing the conducting cell. In contrast, a thick protective layer fills the pit cavity on the side of the parenchyma cell. Ontogeny of the torus thickenings in Cercocarpus represents a third mode of torus de- velopment in eudicots when compared to that occurring in Osmanthus / Daphne and Ulmus /Celtis. Key words: Bordered pit, Cercocarpus, pit membrane, torus, tracheid, vessel element, xylem.

INTRODUCTION At one time, intervascular pit membranes in eudicotyledon woods were thought to be exclusively homogeneous. Then, in 1978, Ohtani and Ishida observed torus-bearing pit membranes in three species of Osmanthus (Oleaceae) and in three species of Daphne (Thymelaeaceae). Further work in several laboratories since that time has increased the number of species with torus-bearing pit membranes to 78 distributed within ten genera and five families (Table 1 and literature cited therein). Our laboratory has been concerned with the ontogeny of such pit membranes. These developmental studies were summarized by Coleman et al. (2004). Basically, two mechanisms of torus manufacture were recognized. One method, as illustrated by Osmanthus americanus and Daphne odora, involved a late deposition of torus pads in association with microtubule clusters. The second method, as found in Celtis occidentalis and Ulmus alata, involved early thickening of the pit membrane without benefit of a microtubule plexus. A recent study described intervascular pit membrane structure in the Rosales with an emphasis on species in the Rosaceae (Jansen et al. 2007). Included in this study were all four species of the genus Cercocarpus (according to Kartesz 1999): C. intricatus S.Wats., C. ledifolius Nutt., C. montanus Raf., and C. traskiae Eastw. Cercocarpus betu- loides Torrey & Gray, as used in this study, is considered a synonym of C. montanus (Kartesz 1999).

1) Department of Biological Sciences, Auburn University, Life Sciences Building, Auburn, Alabama 36849-5407, U. S. A. 2) Jodrell Laboratory, Royal Botanic Gardens, Kew, TW9 3DS, Richmond, Surrey, U. K. 3) Institute of Systematic Botany and Ecology, Ulm University, Albert-Einstein-Allee 11, D-89081 Ulm, Germany. *) To whom correspondence should be addressed [E-mail: [email protected]].

Downloaded from Brill.com09/25/2021 01:59:22PM via free access 54 IAWA Journal, Vol. 31 (1), 2010

Table 1. Eudicot species containing torus-bearing pit membranes in their woods.

Family Species Citation Oleaceae Chionanthus retusa Rabaey et al. 2008 Osmanthus americanus Dute & Rushing 1987 O. aurantiacus Ohtani 1983 O. fortunei Ohtani & Ishida 1978; Ohtani 1983 O. fragrans ” ” O. heterophyllus ” ” O. insularis Ohtani 1983 O. rigidus Ohtani 1983 O. serratulus Rabaey et al. 2006 O. suavis ” Picconia azorica Dute et al. 2008; Rabaey et al. 2008 P. excelsa ” ”

Thymelaeaceae Daphne acutiloba Dute et al. 1992 D. alpina ” D. altaica ” D. arbuscula ” D. arisanensis Dute et al. 1996 D. aurantiaca ” D. bholua ” D. blagayana Dute et al. 1992 D. × burkwoodii ” D. caucasica ” D. cneorum Dute et al. 1990 D. collina Dute et al. 1992 D. ericoides Dute et al. 1996 D. genkwa Dute et al. 1992 D. glomerata ” D. gnidioides Dute et al. 1996 D. gnidium Dute et al. 1992 D. jasminea Dute et al. 1996 D. kiusiana Ohtani & Ishida 1978; Ohtani 1983 D. laureola Dute et al. 1992 D. miyabeana Ohtani & Ishida 1978; Ohtani 1983 D. odora ” ” D. oleoides Dute et al. 1992 D. papyracea ” D. petraea Dute et al. 1996 D. retusa Dute et al. 1992 D. stapfii Dute et al. 1996 D. striata Dute et al. 1992 D. tangutica ” Wikstroemia albiflora Dute et al. 1996 W. kudoi ” W. pauciflora ” W. yakushimensis ” …………………………………………………………………………………………………

Downloaded from Brill.com09/25/2021 01:59:22PM via free access Dute, Patel & Jansen — Torus-bearing pit membranes 55

(Table 1 continued)

Family Species Citation Rosaceae Cercocarpus intricatus Jansen et al. 2007 C. ledifolius ” C. montanus ” C. traskiae ”

Ulmaceae Planera aquatica Dute et al. 2004 Ulmus alata Wheeler 1983 U. americana Jansen et al. 2004 U. campestris Czaninski 1979 U. carpinifolia Jansen et al. 2004 U. coritana ” U. cornubiensis ” U. davidiana ” U. diversifolia ” U. effusa ” U. fulva ” U. glabra ” U. japonica ” U. laciniata ” U. macrocarpa ” U. montana ” U. parvifolia ” U. pedunculata ” U. plotii ” U. scabra ” U. thomasii Wheeler 1983 Zelkova acuminata Jansen et al. 2004 Z. crenata ” Z. cretica Jansen et al. 2007 Z. serrata ”

Cannabaceae Celtis australis Jansen et al. 2007 C. laevigata Wheeler 1983 C. occidentalis ” C. reticulata ”

Intervascular pit membranes of this genus appeared thickened under the light mi- croscope. These thickenings resolved themselves as tori when viewed with the electron microscope. This observation represented a new record for tori in angiosperms, and the first (and so far only) recorded instance of tori in the Rosaceae (Jansenet al. 2007). The aims of the present study were to investigate torus ontogeny in Cercocarpus, and to compare the developmental sequence with that of previously investigated species.

Downloaded from Brill.com09/25/2021 01:59:22PM via free access 56 IAWA Journal, Vol. 31 (1), 2010

Table 2. Sources of Cercocarpus wood specimens examined in this study.

Taxon Herbarium Date of Collection Collector(s) No.

C. betuloides Torrey & A.Gray AUA 29 June 1966 Crampton 7776 C. betuloides Torrey & A.Gray AUA 17 June 2007 Boyd s.n. C. betuloides Torrey & A.Gray K unknown Clokey & Templeton 4596 C. intricatus Wats. DAV 14 May 1977 Levin 1248 C. ledifolius Nutt. K 12 June 2005 S. Jansen, accession number 1980-6418 C. montanus Raf. K unknown Baker et al. 392 C. montanus Raf. var. argentus (Rydb.) F.L. Martin AUA 17 Aug 1980 Drost 72 var. paucidentatus (S.Wats.) F. L. Martin K unknown Rehder 397 var. paucidentatus (S.Wats.) F.L. Martin K unknown Rehder 56 C. traskiae Eastw. AHUC 6 April 1982 Crampton s.n.

MATERIALS AND METHODS

Sources of wood specimens examined in this study are listed in Table 2. Wood segments from herbarium specimens were prepared for light (LM), transmis- sion electron (TEM), and scanning electron (SEM) microscopy according to the methods of Dute et al. (2008). For light microscopy, wood specimens from Cercocarpus mon- tanus and C. betuloides were cut transversely with a razor blade and the resulting seg- ments soaked in two changes of absolute ethanol (30 minutes apiece). The material was then placed in absolute acetone overnight followed by gradual infiltration of the speci- mens with Spurr’s resin (Spurr 1969). Cross sections of 1.5 µm were cut on a Sorvall MT-2b ultramicrotome using a glass knife. Sections were affixed to glass slides and stained using benzoate-buffered, aqueous toluidine blue O (TBO). Images were captured using a Nikon D-70 digital camera attached to a Nikon Biophot microscope. For TEM, Spurr’s-embedded material was sectioned at about 80 nm using a dia- mond knife mounted in the MT-2b ultramicrotome. The resulting ultrathin sections were placed on copper grids, stained either with 1% KMnO4 in 1% sodium citrate (Donaldson 2002) or with uranyl acetate/lead citrate, and observed with a Zeiss EM 10 transmission electron microscope using an accelerating voltage of 60 kV. Unstained sections were viewed as controls for the KMnO4 treatment. For SEM, herbarium specimens from all four species were split to expose either radial or tangential longitudinal surfaces and attached to aluminum stubs using carbon- impregnated double stick tape. Exposed wood surfaces were coated with gold-palladium. The resulting preparations were viewed with a Zeiss DSM 940 operated at 15 kV or a Zeiss EVO 50 at 20 kV. Living branches from four individuals of C. betuloides were collected on 17 June 2007 and on 7 July 2008 in California. The collecting sites were a series of road cuts in a chaparral environment at 752 m altitude on the Los Angeles County/San Ber- nardino County border. The coordinates are 34° 10' 38.68" N; 117° 40' 32.59" W. The specimens were returned on ice in plastic bags to Auburn and preserved the next day.

Downloaded from Brill.com09/25/2021 01:59:22PM via free access Dute, Patel & Jansen — Torus-bearing pit membranes 57

Routine procedures were used for TEM preparation of wood segments (Dute & Rushing 1988) that included glutaraldehyde primary fixation, osmium postfixation, dehydration and embedment in Spurr’s resin. Ultrathin sections were cut, placed on copper grids, and stained with uranyl acetate and lead citrate. Monitor sections of 1.5 µm were cut, mounted on glass slides, and stained with TBO for viewing with LM. Spurr’s embedded herbarium material of C. montanus was sectioned transversely at 3 µm, mounted on glass slides, and stained with 0.01% acriflavin hydrochloride according to the procedure of Donaldson (2002). Qualitative lignin distribution was determined using fluorescence microscopy (Donaldson 2002; Coleman et al. 2004).

Abbreviations used in the figures in this study: A = pit aperture; AN = annulus; M = margo of pit membrane; MI = microtubules; P = axial parenchyma cell; T = tracheid; TO = torus thicken- ing; V = vessel member. — Note: Figures 4, 5 & 7 are images of C. montanus; the remaining images are of C. betuloides.

Figures 1–3. Torus location and anatomy. – 1: Cross section of Cercocarpus wood show- ing a vessel member, tracheids, and an axial parenchyma cell. Tori are indicated by un- labeled arrows. – 2: Aspirated pit membrane between two tracheids in an herbarium speci- men. Note the comparative diameter of the torus and that of the pit apertures. – 3: Aspi- rated pit membrane with a distorted torus. — Scale bar = 2.5 µm for Fig. 1; 1 µm for Fig. 2 & 3.

Downloaded from Brill.com09/25/2021 01:59:22PM via free access 58 IAWA Journal, Vol. 31 (1), 2010

Measurements of pit dimensions were based on both SEM and TEM micrographs. In only a few cases, the horizontal pit diameter, pit aperture and torus diameter could be measured for the same pit on SEM images. Therefore, complementary faces of split radial surfaces were observed in order to measure pit dimensions for the same pit pair (Sano & Jansen 2006). When TEM images were used, the aperture size of a pit pair differed occasionally and the larger of the two pit apertures was measured. Transverse TEM-sections will only show the maximum diameter of the torus and aperture if cut through the pit centre. All measurements were conducted with ImageJ software.

RESULTS

Wood of Cercocarpus contains both vessel members and tracheids. (Rather than enter the debate over tracheid type, we prefer to use the generic term “tracheid” to indicate imperforate, water-conducting cells.) Vessel members can be distinguished from trac- heids in transverse section by their greater diameter (average tangential diameter of 27

Figures 4 & 5. SEM of torus-bear-ing pit membranes (RLS). – 4: Face view of a pit membrane. The distinction between the torus and fibrillar margo is evident. The arrow indicates a cluster of plasmodes- matal channels. – 5: Pit membrane (1) with a torus and an exposed pit border (2) with an aperture. Note the shape of both torus and aperture and their comparative diameters. — Scale bar = 1 µm for Fig. 4; 2 µm for Fig. 5.

Downloaded from Brill.com09/25/2021 01:59:22PM via free access Dute, Patel & Jansen — Torus-bearing pit membranes 59

µm vs. 12 µm, n = 25) and narrower secondary walls (mean thickness of 2.93 vs. 1.27 µm, n = 15). Also, vessel member walls stain more of a blue-green with TBO then do those of tracheids. Tori are found in pit membranes connecting vessel member to vessel member, vessel member to tracheid, and tracheid to tracheid (Fig. 1). Torus-bearing pit membranes are present in both early- and latewood and also are found in conducting elements of metaxylem. Tori often are aspirated in wood of herbarium specimens (Fig. 2). The stresses associ- ated with air drying sometimes are reflected in the distorted structure of the torus (Fig. 3). Specimens of fresh, preserved wood in contrast are typically nonaspirated (Fig. 1). Observed with the SEM, the torus is circular and sits atop the pit membrane. Tiny openings (possibly remnants of plasmodesmata) were observed in two tori (Fig. 4). This occurrence was unusual, and the remainder of tori observed lacked such perfora- tions. The rare occurrence of such openings was first noted by one of us (Jansen) during preparation of a previous manuscript (Jansen et al. 2007). Only two instances of plas- modesmata traversing intervascular pit membranes were noted in TEM specimens. Torus diameter in tracheids ranges from 1.48–3.64 µm with an average of 2.35 ± 0.57 µm (± SE, n = 28), and in vessel elements from 1.46–3.2 µm, with a mean diam- eter of 2.17 ± 0.42 µm (± SE, n = 21). These numbers fall within the range given for the four species of Cercocarpus previously studied (Jansen et al. 2007) but exceptions do exist. Horizontal pit size is on average 4.88 ± 0.82 µm for both vessel elements and tracheids (n = 49), varying from 3.19 µm to 6.79 µm. Torus diameter of tracheids generally is greater than that of the apertures (Fig. 5; Jansen et al. 2007). There is a good correlation between torus diameter and pit size for both tracheids and vessel ele- ments (R2 = 0.42, P < 0.0001). A weak correlation (R2 = 0.26, P < 0.0001) between pit diameter and aperture diameter was only found for tracheids. Pit apertures in tracheids are somewhat smaller (on average 1.14 ± 0.35 µm, n = 28) than in vessel elements

Figures 6 & 7. Torus structure and composition. – 6: A torus with but one pad (thickening). The fibrillar nature of the margo is evident. – 7: Wood from an herbarium specimen stained with KMnO4. The torus pads stain heavily (arrows). The wall material trapped between the pads is unstained. – Scale bar = 1 µm for Fig. 6 & 7.

Downloaded from Brill.com09/25/2021 01:59:22PM via free access 60 IAWA Journal, Vol. 31 (1), 2010

(1.23 ± 0.33 µm, n = 28). In vessel elements the pit apertures leading to torus-bearing pit membranes vary in outline from circular to distinctly fusiform. Elongate apertures might explain instances where the aperture diameter is equal to (Fig. 1 & 6) or greater than the torus diameter. Variation in torus diameter relative to pit diameter and pit ap- erture can be described by the torus overlap, which corresponds to the fraction of the pit border width that is covered by the torus. Torus overlap values vary between 0.09 and 0.51, with slightly higher mean values for tracheids (0.30 ± 0.09) than for vessel elements (0.26 ± 0.08). The torus thickness is on average 342 ± 135 nm (n = 17) with values varying be- tween 171 and 617 nm, while the non-thickened part of intertracheary pit membranes varies from 101 nm to 251 nm, with an average thickness of 157 ± 35 nm (n = 27). No significant differences were observed in pit membrane thickness and torus thickness between intervessel and intertracheid pit pairs. The torus, when stained with uranyl acetate and lead citrate, consists of an electron- dense central layer positioned between two pads or thickenings that stain in an identical manner to the secondary cell wall (Fig. 2). However, some sectioned material shows tori with only one pad (Fig. 6). Potassium permanganate staining indicates that the torus pads of mature pit mem- branes are lignified, whereas the compound middle lamella of the pit membrane is not (Fig. 7). Control (unstained) pit membranes are either invisible or show a clear com- pound middle lamella with a denser torus pad on either side but with much less clarity than stained material. Acriflavin-stained tori give a green fluorescence indicative of lignification. Torus pad initiation occurs late in ontogeny after the pit border has been completed or nearly so (Fig. 8–10). Pads are deposited separately with the older cell of the pair depositing its pad first. Detailed investigation shows no microtubule plexus associated with pad deposition (Fig. 9 & 10). Microtubules are found associated with the secondary wall of the pit borders (Fig. 9 & 10) and occasionally with the primary wall of the pit membrane before deposition of the torus pad. Figure 11 shows a developing torus pad that appears somewhat disrupted by the process of fixation. The image clearly shows a fibrillar component to the torus thickening.

Figure 8. Pit membrane between two tra- cheids whose pit borders are complete or nearly so. Torus pads have yet to form. — Scale bar = 1 µm.

Downloaded from Brill.com09/25/2021 01:59:22PM via free access Dute, Patel & Jansen — Torus-bearing pit membranes 61

Figures 9–12. Development of torus pads. – 9 & 10: Two examples of pit membranes with devel- oping torus pads (unlabeled arrows). Microtubules are associated with the pit borders. – 11: An example of a developing torus pad showing its fibrillar nature. – 12: A pit membrane connecting mature and immature tracheary elements. In the former cell the surface of the margo is beginning to erode (unlabeled arrow). — Scale bar = 0.2 µm for Fig. 9–11; 0.5 µm for Fig. 12.

Downloaded from Brill.com09/25/2021 01:59:22PM via free access 62 IAWA Journal, Vol. 31 (1), 2010

Figures 13–15. Structure of half-bordered pit pairs. – 13: Axial parenchyma cell adjacent to both a vessel member and tracheids. The unla- beled arrows indicate the thickened regions of the protective layer where it is associated with simple pits. – 14: A detail of a half-bordered pit pair showing a torus thickening on the tracheid side (arrow) and a thickened, fibrillar protec- tive layer (double arrow) on the parenchyma cell side. – 15: A half-bordered pit pair whose membrane has a very thin, eccentric torus pad (arrow). — Scale bar = 2.5 µm for Fig. 13; 1 µm for Fig. 14; 0.5 µm for Fig. 15.

Torus pad deposition is followed by cell autolysis. Figure 12 shows a pit membrane between mature and immature tracheary elements. The former cell has lost its cy- toplasm, and incipient erosion of matrix material from the margo is evident. It appears as if vessel members mature (i.e. lose their cytoplasm) much earlier than do surrounding tracheids. The pit membranes also show a distinct “annulus,” which can be seen as an electron dense area near the rim of the pit border (Fig. 1 & 6). Vessel members and tracheids are as- sociated with parenchyma cells, both axial and ray (Fig. 1 & 13). Communication be- tween parenchyma cells is accomplished by simple pit pairs whose membranes are traversed by plasmodesmata, whereas parenchyma cells and mature conducting elements are connected by half-bordered pit pairs without cytoplasmic channels (Fig. 14). On the parenchyma cell side of the pit membrane, the pit cavity is filled with a fibrillar wall deposit known as the protective layer (Fig. 13–15). This layer extends around the cell’s circumference but is thickest in the pit cavity of the half-bordered pit pair (Fig. 14). A torus pad (Fig. 14) is sometimes observed on the pit membrane surface of the conducting element. Often the pad is poorly developed, being not only thin, but also incomplete and/or eccentric in deposition (Fig. 15). Deposition of the torus pad in a half-bordered pit pair does not involve a microtubule plexus and seems to occur before deposition of the “protective” layer (Fig. 16). In other instances a protective layer is present but a torus pad is not

Downloaded from Brill.com09/25/2021 01:59:22PM via free access Dute, Patel & Jansen — Torus-bearing pit membranes 63 visible. Whether a small torus pad would have been located by serial sectioning in such instances is uncertain. One instance was observed of a torus pad deposited in a blind pit (Fig. 17).

Figures 16 & 17. – 16: Developing half-bordered pit pair in which a torus pad is present (ar- row), but the protective layer has not yet formed. – 17: A blind pit with a torus pad. — Scale bar = 0.5 µm for both.

DISCUSSION

Water-conducting cells of Cercocarpus are programmed to deposit a torus pad on the pit membrane without regard to the identity of the neighboring cell (Jansen 2007 and this manuscript). In adjacent developing conducting elements, the older of the cells will show torus pad deposition before the younger. Also, an instance was encountered of a torus pad being deposited in a blind pit. Finally, torus pads often develop on the tracheary element side of the pit membrane of a half-bordered pit pair (first noted by Jansen et al. 2007 and confirmed in this manuscript). It is clear from these instances that tracheary elements are not in communication with neighboring cells as regards pit membrane development. Perhaps this lack of communication in Cercocarpus can be attributed to the paucity or lack of plasmodesmata in the developing pit membranes. Even the openings in the torus in Figure 4 have been associated with transport of molecules for torus synthesis rather than with communication per se (Barnett 1981). Blind pits with a torus thickening have been found in Osmanthus americanus (Dute & Rushing 1988) and Daphne odora (Dute et al. 1990). Half-bordered pit pairs in D. odora also had a torus pad develop on the side of the water-conducting cell (Dute et al. 1990; Coleman et al. 2004). We consider the torus pad in Cercocarpus to be secondary wall material due to the staining qualities of the pad (see for example Fig. 2) and to the time of deposition. In

Downloaded from Brill.com09/25/2021 01:59:22PM via free access 64 IAWA Journal, Vol. 31 (1), 2010 this respect, Cercocarpus is similar to Osmanthus and Daphne (Dute & Rushing 1988; Dute et al. 1990). In a recent article, Coleman et al. (2004) distinguished between spe- cies of Celtis and Ulmus that began torus thickening before the pit border was initiated and Osmanthus and Daphne that began torus thickening well after pit border initiation. Clearly Cercocarpus falls into the latter category, but unlike the latter two genera, torus pad development in Cercocarpus is not associated with a microtubule plexus. Thus, a third category of torus development is indicated. Cercocarpus species are shrubs or small trees with a wide distribution extending from the western Great Plains west to the Pacific coast and from Mexico north to Washington state (McGregor et al. 1986; Hickman 1993; Cronquist et al. 1997; USDA, NRCS 2009). Populations of the genus also exhibit a great distribution in altitude. In California, for example, specimens of Cercocarpus can be found from 100 to 3000 m above sea level. The plants favor dry environments such as rocky slopes, cliffs and chaparral (Bailey & Bailey 1976; McGregor et al. 1986; Hickman 1993; Cronquist et al. 1997). Carlquist (1988) considers individuals of Cercocarpus to be at a selective advantage under such environmental conditions because air embolisms should be confined by the tracheids. We would carry this hypothesis a step further. SEM observations of Cercocarpus wood show intervascular pit membranes with a tightly woven margo surrounding the torus (Jansen et al. 2007), a situation similar to that found in other torus-bearing angiosperms (Dute & Rushing 1987; Dute & Rushing 1990; Dute et al. 1990). We would postulate that the thickened torus prevents tearing of the central portion of the pit membrane during aspiration. Such a function was hypothesized previously by Wheeler (1983) and Dute and Rushing (1987). Thicker angiosperm pit membranes have been illustrated to be less porous than thinner pit membranes, which also results in higher air-seeding thresholds and therefore reduced vulnerability to drought-induced cavitation (Jansen et al. 2009). However, recent work by Wheeler et al. (2005) suggests another expla- nation. The authors published an experimental study relating inter-vessel pitting and cavitation in woody species of the Rosaceae. Among the species investigated were two species of Cercocarpus, C. ledifolius and C. montanus. The authors assumed that they had limited their study “to inter-vessel pits with homogeneous pit membranes”, not knowing that all species of Cercocarpus possess tori. Their experiments showed that both species exhibited high resistance to cavitation relative to most other rosaceous species investigated. This result is exemplified by relatively high P50 values, which represent the pressure that reduces hydraulic conductance by 50%: -7.5 MPa for C. betuloides, -4.9 MPa for C. ledifolius, and -5.8 MPa for C. montanus (Wheeler et al. 2005; Jacobsen et al. 2007). Further experiments indicated that the total pit area of a vessel was the critical factor in determining cavitation resistance (Wheeler et al. 2005; Hacke et al. 2006). Wheeler et al. (2005) assumed an absence of torus thickenings in their calculations, but the presence of a torus would decrease the functional (water- conducting) surface of pit membranes by a considerable amount. Presence of a torus thickening associated with a (more-or-less) uniformly microporous angiosperm-type of pit membrane has always been difficult to explain from a functional standpoint. Generally, it has been assumed that the torus thickening prevents membrane rupture during aspiration (see above), but the torus could simply be reducing the permeable

Downloaded from Brill.com09/25/2021 01:59:22PM via free access Dute, Patel & Jansen — Torus-bearing pit membranes 65 area of the membrane. However, this hypothesis does not explain why the torus is centrally located on the pit membranes or why the diameter of the torus generally is greater than that of the aperture. The torus in Cercocarpus represented c. 46% of the pit diameter, which is similar to values reported for gymnosperms. Hacke and Jansen (2009), for instance, reported 48% for three Pinaceae species, and 45% was found in earlywood tracheids of Douglas-fir (Domec et al. 2006). Values for torus overlap in Cercocarpus also correspond with data on torus overlap in gymnosperms (Hacke et al. 2004; Hacke & Jansen 2009) and agree with theoretical calculations by Hacke et al. (2004), who suggested that a maximum pit conductivity occurs for a torus overlap between 0.24 and 0.30. Therefore, tight scalings between tori and pit dimensions in angiosperms and gymnosperms seem to suggest similar functional implications and highlight the importance of intertracheary pitting for hydraulic efficiency and cavitation resistance.

ACKNOWLEDGEMENTS

We wish to thank Curtis Hansen, curator of the John D. Freeman Herbarium at Auburn University (AUA), for his assistance with the taxonomy of Cercocarpus as well as Robert Boyd (also of Auburn University) for his field collections ofC. betuloides. Thanks are also due to the DAV and AHUC herbaria and to Jean Shepard at the University of California, Davis, for providing herbarium specimens.

REFERENCES Bailey, L.H. & E.Z. Bailey. 1976. Hortus third, a concise dictionary of plants cultivated in the United States and Canada. Macmillan Publishing Co., Inc., New York. Barnett, J.R. 1981. Secondary xylem cell development. In: J.R. Barnett (ed.), Xylem cell devel- opment: 47–95. Castle House Publications Ltd., Tunbridge Wells, Kent. Carlquist, S. 1988. Comparative wood anatomy. Springer-Verlag, Berlin. Coleman, C.M., B.L. Prather, M.J. Valente, R.R. Dute & M.E. Miller. 2004. Torus lignification in hardwoods. IAWA J. 25: 435–447. Cronquist, A., N.H. Holmgren & P.K. Holmgren. 1997. Intermountain flora: vascular plants of the Intermountain West, U.S.A. The New York Botanical Garden, Bronx, New York. Czaninski, Y. 1979. Cytochimie ultrastructurelle des parois du xylème secondaire. Biol. Cellu- laire 35: 97–102. Domec, J.-C., B. Lachenbruch & F.C. Meinzer. 2006. Bordered pit structure and function determine spatial patterns of air-seeding thresholds in xylem of Douglas-fir (Pseudotsuga menziesii; Pinaceae). Amer. J. Bot. 93: 1588–1600. Donaldson, L.A. 2002. Abnormal lignin distribution in wood from severely drought stressed Pinus radiata trees. IAWA J. 23: 161–178. Dute, R.R., J.D. Freeman, F. Henning & L.D. Barnard. 1996. Intervascular pit membrane structure in Daphne and Wikstroemia – Systematic implications. IAWA J. 17: 161–181. Dute, R., L. Hagler & A. Black. 2008. Comparative development of intertracheary pit membranes in Abies firma and Metasequoia glyptostroboides. IAWA J. 29: 277–289. Dute, R.R., A.L. Martin & S. Jansen. 2004. Intervascular pit membranes with tori in wood of Planera aquatica J.F. Gmel. J. Alab. Acad. Sci. 75: 7–21. Dute, R.R. & A.E. Rushing. 1987. Pit pairs with tori in the wood of Osmanthus americanus (Oleaceae). IAWA Bull. n.s. 8: 237–244.

Downloaded from Brill.com09/25/2021 01:59:22PM via free access 66 IAWA Journal, Vol. 31 (1), 2010

Dute, R.R. & A.E. Rushing. 1988. Notes on torus development in the wood of Osmanthus americanus (L.) Benth. & Hook. ex Gray (Oleaceae). IAWA Bull. n.s. 9: 41–51. Dute, R.R. & A.E. Rushing. 1990. Torus structure and development in the woods of Ulmus alata Michx., Celtis laevigata Wild., and Celtis occidentalis L. IAWA Bull. n.s. 11: 71–83. Dute, R.R., A.E. Rushing & J.D. Freeman. 1992. Survey of intervessel pit membrane structure in Daphne species. IAWA Bull. n.s. 13: 113–123. Dute, R.R., A.E. Rushing & J.W. Perry. 1990. Torus structure and development in species of Daphne. IAWA Bull. n.s. 11: 401–412. Hacke, U.G. & S. Jansen. 2009. Embolism resistance of three boreal conifer species varies with pit structure. New Phytol. 182: 675–686. Hacke, U.G., J.S. Sperry & J. Pittermann. 2004. Analysis of circular bordered pit function. II. Gymnosperm tracheids with torus-margo pit membranes. Amer. J. Bot. 91: 386–400. Hacke, U.G., J.S. Sperry, J.K. Wheeler & L. Castro. 2006. Scaling of angiosperm xylem struc- ture with safety and efficiency. Tree Physiol. 26: 689–701. Hickman, J.C. 1993. The Jepson manual: higher plants of California. Univ. of California Press. Jacobsen, A.L., R.B. Pratt, F.W. Ewers & S.D. Davis. 2007. Cavitation resistance among 26 chaparral species of southern California. Ecol. Monogr. 77: 99–115. Jansen, S., B. Choat & A. Pletsers. 2009. Morphological variation of intervessel pit membranes and implications to xylem function in angiosperms. Amer. J. Bot. 96: 1–12. Jansen, S., B. Choat, S. Vinckier, F. Lens, P. Schols & E. Smets. 2004. Intervascular pit mem- branes with a torus in the wood of Ulmus (Ulmaceae) and related genera. New Phytol. 163: 51–59. Jansen, S., Y. Sano, B. Choat, D. Rabaey, F. Lens & R.R. Dute. 2007. Pit membranes in trache- ary elements of Rosaceae and related families: new records of tori and pseudotori. Amer. J. Bot. 94: 503–514. Kartesz, J.T. 1999. A synonymized checklist and atlas with biological attributes for the vascular flora of the United States, Canada, and Greenland. First Edition. In: J.T. Kartesz & C.A. Mechan (eds.), Synthesis of the North American Flora, Version 1.0. North Carolina Botani- cal Garden, Chapel Hill. McGregor, R.L., T.M. Barkley, R.E. Brooks & E.K. Schofield. 1986. Flora of the Great Plains. University Press of Kansas, Lawrence, Kansas. Ohtani, J. 1983. SEM investigation on the micromorphology of vessel wall sculptures. Res. Bull. Coll. Exp. For. Hokkaido Univ. 40: 323–386. Ohtani, J. & S. Ishida. 1978. Pit membrane with torus in dicotyledonous woods. J. Jap. Wood Res. Soc. 24: 673–675. Rabaey, D., S. Huysmans, F. Lens, E. Smets & S. Jansen. 2008. Micromorphology and system- atic distribution of pit membrane thickenings in Oleaceae: tori and pseudo-tori. IAWA J. 29: 409–424. Rabaey, D., F. Lens, E. Smets & S. Jansen. 2006. The micromorphology of pit membranes in tracheary elements of Ericales: New records of tori and pseudo-tori. Ann. Bot. 98: 943–951. Sano, Y. & S. Jansen. 2006. Perforated pit membranes in imperforate tracheary elements of some angiosperms. Ann. Bot. 97: 1045–1053. Spurr, A.R. 1969. A low viscosity epoxy resin embedding medium for electron microscopy. J. Ultrastruct. Res. 26: 31–45. USDA, NRCS. 2009. The PLANTS database, accessed: 8 April 2009 (http://plants.usda.gov), Baton Rouge, Louisiana: National Plant Data Center. Wheeler, E.A. 1983. Intervascular pit membranes in Ulmus and Celtis native to the United States. IAWA Bull. n.s. 4: 79–88. Wheeler, J.K., J.S. Sperry, U.G. Hacke & N. Hoang. 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 Environ. 28: 800–812.

Downloaded from Brill.com09/25/2021 01:59:22PM via free access