Tyloses and Ecophysiology of the Early Carboniferous ...

Annals of Botany 91: 739±747, 2003 doi:10.1093/aob/mcg068, available online at www.aob.oupjournals.org

Tyloses and Ecophysiology of the Early Carboniferous Progymnosperm Tree Protopitys buchiana S T E P H E N E . S C H E C K L E R 1 , * and J EA N G A L T I E R 2 of Biology and Geological Sciences, and Museum of Natural History, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061-0406, USA and 2Botanique et Bioinformatique de l'Architecture des Plantes, CIRAD, TA40/PS2, Boulevard de la Lironde, F-34398 Montpellier Cedex 5, France

1Departments

Received: 16 October 2002 Returned for revision: 13 December 2002 Accepted: 16 January 2003 Published electronically: 19 March 2003

Trunk woods of Early Carboniferous Protopitys buchiana show the earliest example of tylose formation and the ®rst record for a progymnosperm. Protopitys tyloses are more densely located in inner trunk woods and near growth layer boundaries. We suggest, therefore, that an altered physiological state of living ray cells, during dormancy and/or following water stress, was necessary to make the woods vulnerable to tylose formation. Coupled with the distribution and proximity of abundant wood ray parenchyma to large xylem conducting cells, the positions of conduits ®lled with tyloses can be interpreted as ecophysiological responses of the plant to changes in local environment. In addition, some xylem conducting cells might have functioned as vessels. Fungal hyphae are present in some tracheary cells and in some areas with tyloses, but there is no evidence for wood trauma; we conclude, therefore, that these particular cases of tyloses are probably not induced by wound trauma. Protopitys buchiana wood thus shows structure/function similarities to modern woods with vessels, such as those of dicot angiosperms. This implies that ancient and modern plant ecophysiological responses correlate well with the physical parameters of their cellular construction. ã 2003 Annals of Botany Company Key words: Ecophysiology, Carboniferous, progymnosperm, lignophyte, Protopitys, tyloses, vessels, wood anatomy.

INTRODUCTION Cellular construction of tyloses, cell ingrowths of living wood parenchyma into the cavities of xylem conducting cells, is well understood (Esau, 1965; Fahn, 1982, 1997; Zimmermann, 1983; ZuÈrcher et al., 1985; Carlquist, 1988; Mauseth, 1988). However, much less is known of the ecophysiology and structure/function relationships of these cellular ingrowths, except that their presence correlates with vessel embolism (Kozlowski and Pallardy, 1996; Nilsen and Orcutt, 1996; Canny, 1997; Jaquish and Ewers, 2001). Descriptions of tyloses come almost entirely from modern dicot angiosperm woods with large vessel diameters and/or with ring-porous construction (Esau, 1965; Fahn, 1982, 1997; Zimmermann, 1983; ZuÈrcher et al., 1985; Carlquist, 1988; Mauseth, 1988). Dicots are crown group members of a phylogenetic group called `lignophytes' for those plants with a particular type of bifacial vascular cambium that makes secondary xylem inward and secondary phloem outward (Crane, 1985; Doyle and Donoghue, 1986; Nixon et al., 1994; Rothwell and Serbet, 1994; Doyle, 1996, 1998; Kenrick and Crane, 1997a, b; Bateman et al., 1998; Bowe et al., 2000; Chaw et al., 2000). Lignophytes include the spore-shedding Palaeozoic progymnosperms (Devonian to Carboniferous), such as Protopitys, which we describe here, as well as Palaeozoic to Recent gymnosperm and angiosperm seed plants. Monocots, which lack this type of cambial growth, are included because of their undoubted phylogenetic * For correspondence. E-mail [email protected]

derivation from ancestral woody dicot angiosperms (Kenrick, 1999; Mathews and Donoghue, 1999; Soltis et al., 1999; Graham and Olmstead, 2000; Williams and Friedman, 2002). Vascular cambia of all other woody non-seed-bearing plants, such as extinct lepidodendrids, calamites, sphenophylls, zygopterid and other Devonian fern-like plants, and modern Isoetes and Botrychium, are regarded as independently derived. These are, therefore, called `non-lignophytes', a term with no phylogenetic implication, since its members have only tenuous or no relationships with the others of this group (Kenrick and Crane, 1997a, b; Pryer et al., 2001). Although we mostly think of tyloses as the provenance of dicot angiosperms, they do in fact occur in some monocots, non-angiosperm lignophytes, and living and extinct nonlignophytes. Tyloses are known to occur in the primary xylem of a few Palaeozoic and living ferns and horsetails (e.g. Carboniferous Ankyropteris and Archaeocalamites; Williamson, 1877; McNichol, 1908; Bertrand, 1909; Holden, 1925, 1930; Walton, 1949; Bierhorst, 1960; Ogura, 1972; Carlquist and Schneider, 1999), as well as in some modern bamboos and gymnosperms such as Pinus and Gnetum (Chrysler, 1908; Coulter and Chamberlain, 1917; ZuÈrcher et al., 1985; Carlquist, 1996a; Liese and Weiner, 1996). Tyloses can even occur in such non-tracheary cell types as ®bre-tracheids or laticifers (Gottwald, 1972; Carlquist, 1996a). We report here the ®rst record (Figs 1±3) of a Palaeozoic lignophyte wood with tyloses, which is also the oldest

Annals of Botany 91/6, ã Annals of Botany Company 2003; all rights reserved

740

Scheckler and Galtier Ð Tyloses and Ecophysiology of Protopitys example of such a tissue. Excellent preservation of the three-dimensional distribution of these tyloses, which correlates with distinct phases of growth, allows us to reconstruct some of the ecophysiology of a Protopitys tree.

MATERIALS AND METHODS Early Carboniferous (Visean) ¯uvial and volcaniclastic sediments from the Vosges region of France contain an abundant assemblage of short-distance transported, silicate permineralized, gymnospermous lignophyte woods (Galtier et al., 1998), which we have studied using microscopy. Specimens were etched using 40 % hydro¯uoric acid and peeled (Joy et al., 1956). However, petrologic ground thinsections proved to be superior to the peels for showing the ®ne wall details of xylem parenchyma and tracheary cells. Thin-sections were made in the transverse, radial, and tangential planes. Transmitted light photographs were taken using an Olympus BH2 microscope. A minimum of 100 measurements was taken for each parameter. Two primary specimens were used in this study and are kept in the collections of PaleÂobotanique, Universite Montpellier 2, France. They are identi®ed as Protopitys buchiana Goeppert by the characteristic wood anatomy of this species (Galtier et al., 1998, Plate I, Figs 1±4, 9) and by comparison with the original material of P. buchiana (Galtier et al., 1998, Plate I, Figs 10±11). One specimen (BOU 1500) was collected from the Bourbach-le-Bas quarry, while the other (BOU 1501) came from the Scheuermatt quarry. The two specimens are incomplete pieces of wood, each about 10 3 7 3 8 cm without the pith, but showing a minimum radial dimension of 7 cm, thus suggesting that the original trunk diameters were at least 15 cm. Both quarries contain abundant silicate permineralized plant axes in rhyolitic ®ne ash deposits that are dated to latest Visean to possibly earliest Namurian (336 + 3/ ± 5 Ma; Schaltegger et al., 1996; Galtier et al. 1998). We also re-examined three ground thin-sections (489, 490, 491; cross, radial and tangential sections, respectively) of P. buchiana, that were made for Solms-Laubach in 1897. We suspect that these are duplicate slides made from the specimens used for his 1893 study of this species for distribution to museums. They were borrowed by Galtier from the Kidston Collection of slides, housed at the Hunterian Museum, University of Glasgow, Scotland. The slides are labelled `Falkenburg (sic), Silesia, Culm, SolmsLaubach 1897'. Silesia is now part of Poland and the village of Falkenberg no longer appears on modern maps; it has apparently been renamed, but its modern equivalent is unknown to our Polish and German colleagues. Culm

F I G . 1. Transverse sections of Protopitys buchiana wood (BOU 1500 AT 01). Trunk exterior is up in A±C. A, Outer growth layer of a trunk that shows two closely spaced growth pauses (arrows). Ray parenchyma and tyloses are brown-coloured cells. B, Inner growth layer (arrows) of the same trunk. C, Enlarged area near that shown in A, showing tyloses (T) inside tracheary cells, and the physical relation between rays (R) and tracheary cells with tyloses. Bars = 200 mm (A and B) and 100 mm (C).

Scheckler and Galtier Ð Tyloses and Ecophysiology of Protopitys strata, however, are unequivocally Upper Visean (Early Carboniferous). The land source for the Vosges woods is interpreted from sedimentary features and geochemistry (Rex, 1986; Galtier et al., 1998) as a frequently disturbed, volcanogenic terrain of a palaeotropical to subtropical climate zone (approx. 10±20°S; Scotese and McKerrow, 1990; Scotese and Golonka, 1992; Van der Voo, 1993). Studies like these are able to successfully reconstruct ancient landscapes with remarkable precision, but it is rarely possible to interpret the ecophysiological status of individual community members.

R E SU L T S Several other gymnospermous lignophyte wood pieces from these Vosges localities also have tyloses (e.g. Paleoxylon bourbachensis BOU 1502), similar to the two trunk fragments of Protopitys buchiana (BOU 1500 and 1501) and Solms-Laubach's (1893) Polish material, so this is not unique to just one specimen or to a single population of P. buchiana. However, the magnitude and distribution of tyloses in the wood of one piece of P. buchiana (BOU 1500) is so exceptional that we are able to deduce many aspects of the ecophysiology of this particular P. buchiana tree. Transverse-sections of these Protopitys woods (Fig. 1) show that tyloses-®lled tracheary cells occur in prominent radial rows. Tracheary cells vary considerably in size, but only the larger ones have tyloses (Fig. 1C). Radial diameters are 22±95 mm (average 55 mm). Tangential diameters vary more, from as little as 15±20 mm, where a row of fusiform initials is lost, to over 100 mm in many other cell rows. Cross-sectional areas of the largest tracheary cells can equal or exceed 10 000 mm2, which compares well with the dimensions of vessels in many dicot angiosperms (Canny, 1997; Jaquish and Ewers, 2001). Rays of this plant are low (one to a few cells high), mostly uniseriate but with some local multiseriate areas, and are so numerous (Fig. 3A and B) that every tracheary cell has ray contacts on both radial sides. In these simple woods, ray distribution controls the overall arrangement of tyloses, as is shown by our transverse, radial and tangential views (Figs 1C, 2A±C and 3A). The density of tyloses-blocked tracheary cells, however, correlates with inner vs. outer position of a sample within a trunk and is affected by apparent tree dormancy. Inner trunk portions (compare Fig. 1A and B) have a much greater density of cells with tyloses. Maximum abundance of tyloses (Fig. 1A and B) is seen in earlywood tracheary cells just inside or outside latewood-marked growth boundaries. Growth boundaries are abrupt and subtle (Fig. 1A and B), consisting of just one to four latewood cells. Latewood tracheary cells differ from earlywood by their decreasing radial dimensions, but have no greater thickening of their secondary walls. Some growth boundaries can be followed from edge to edge, but others are discontinuous. Similar growth boundaries are typical of plants growing under wet/ dry climatic regimes where, however, there is often no correlation of growth pauses to the passing of calendar years (Chambers et al., 1998).

741

Heartwood- vs. sapwood-like variation in xylem conductance capacity is thus seen in these Protopitys trunks that otherwise lack visible accumulations of oils, tannins and other secondary metabolites characteristic of heartwood of modern trees (Zimmermann, 1983; Hillis, 1987). However, this vast increase of wood parenchyma volume that resulted from formation of tyloses (compare Fig. 1A and B) might well have been an important physiological adaptation for periodic dormancy and regrowth. Indeed, periods of apparent growth interruption, as evidenced by these faint growth layers, suggest that an altered physiological state of ray parenchyma cells somehow stimulated tyloses formation, just as in modern plants. But growth layers in Protopitys or other fossil or living woods from tropical to subtropical climates may re¯ect treespeci®c microclimatic or traumatic factors rather than any cyclicity of local or regional climate or rainfall distribution (Chambers et al., 1998). We are thus unable to estimate how much time is represented by the 6±7 mm distance from inner (Fig. 1B, arrows) to next outer (Fig. 1A, lower arrows) growth boundaries of this trunk. Indeed, the layer that we call the outer growth boundary can actually be seen to consist of two closely spaced pauses of growth (Fig. 1A, upper and lower sets of arrows). Tracheary cells of P. buchiana have mostly scalariform bordered inter-tracheal pits on their radial walls (Fig. 2A, B and D), unpitted tangential walls (Fig. 3B), and circular cross-®eld pitting at ray contact cell overlap areas (Fig. 2B± D at R). Most inter-tracheal pits have intact pit membranes (Fig. 2A, B and D, upper arrows), but a few localized areas appear to lack pit membranes (Fig. 2A, arrow; D, lower arrows), which resemble vessel perforations. The Protopitys woods, like the other Vosges woods, are penetrated by fungal hyphae which have traversed tracheary cell cavities longitudinally and pit pairs horizontally. There is no other evidence of trauma in these samples and, in the Protopitys woods studied, nearly all tyloses occur independently of any fungal activity. We do not believe, therefore, that these tyloses were wound-trauma induced. However, the presence of hyphae in a few cells with tyloses may indicate that some fungal infection occurred while the tree was alive. DISCUSSION Exceptionally rapid cellular preservation of quickly buried trunks gives us unusual insights into the life circumstances of an ancient ¯oodplain forest. Several wood pieces from trunks of the progymnosperm P. buchiana, and a few others of gymnospermous lignophyte-type, show abundant tyloses (Figs 1±3) that ®ll portions of long tracheary cells adjacent to ray contact areas (Figs 1C, 2B±D at R, 3A at arrow). Xylem cells with tyloses form prominent radial cell ®les that are especially numerous near faint growth layer boundaries (Fig. 1A and B). Similar faint growth layer boundaries are found in other contemporaneous gymnosperm lignophytes (Pitus, Eristophyton, Stanwoodia) from nearby volcanogenic terrains in Scotland (Galtier and Scott, 1991, 1994; Galtier et al., 1993), which were interpreted

742

Scheckler and Galtier Ð Tyloses and Ecophysiology of Protopitys

Scheckler and Galtier Ð Tyloses and Ecophysiology of Protopitys by Galtier and Scott (1994) as reactions of individual trees to stressful local environments rather than to ¯uctuating climates. We believe that this interpretation applies equally to our Protopitys specimens. De®nite climate-induced growth layers, however, are an ancient lignophyte cambium growth response that is well shown by Middle and Late Devonian aneurophyte progymnosperms (e.g. Matten and Banks, 1967; Scheckler and Banks, 1971; Scheckler, 1975; Dannenhoffer and Bonamo, 1989) that grew in an apparent savanna-like paleoclimate with pronounced wet/dry seasons (Scheckler et al., 2000). Studies of modern plant woods (Zimmermann, 1983; Carlquist, 1988; ZuÈrcher et al., 1985; Canny, 1997; Murakami et al., 1999) show that tyloses form only from living ray or axial wood parenchyma contact cells by expansion of their protoplasts to bulge through vessel pits. Mutual contacts of tyloses and divisions of their cells can ®ll xylem conducting cells of inner, older wood layers of live trees, or petiole bundles of smaller plants, with a parenchymatous cellular to pseudocellular tissue. Tyloses are said to form mostly during the dormant season when xylem conducting cells are under little or no transpirational tension, or have cavitated, and their water pressure is at or near equilibrium with the cell osmotic pressure of adjacent wood parenchyma (ZuÈrcher et al., 1985; Cochard and Tyree, 1990; Kozlowski and Pallardy, 1996; Liese and Weiner, 1996; Nilsen and Orcutt, 1996; Jaquish and Ewers, 2001). On the other hand, Canny (1997) describes tyloses forming in young, actively transpiring Helianthus annuus petioles in vessels that were vulnerable to cavitation or that were in a current state of embolism where their xylem sap had crossed over to the gas phase. Canny interpreted these observations to suggest that tyloses are an important mechanism for preserving tissue pressure, which is required to re®ll embolisms in the remaining vessels. Tyloses, from his point of view, provided support for his (Canny, 1995) compensating pressure theory of transpiration. The mechanisms of transpiration, especially the roles of embolism formation and re®lling, are still incompletely known (Canny, 1995, 1998, 2001a, b; Zimmermann et al., 1995; Jaquish and Ewers, 2001). Other workers suggest that embolism re®lling may involve recycling of MuÈnch water (Milburn, 1996) and/or capillary forces (Sherwin et al., 1998), in addition to, or instead of, cell turgor (Canny, 1995, 1998). Canny (1995) proposed that the compensating tissue pressures of adjacent living cells, which press against the vessels, help to protect the transpiration stream from cavitation and assist their re®lling. Regardless of the roles that tyloses might play in regulation of the transpiration stream, Canny's (1995) compensating pressure theory has

743

come under some criticism by Comstock (1999) and Stiller and Sperry (1999). There is, however, general consensus that physical blockage of conduits by tyloses precludes further water transport, redirects transport to younger, less vulnerable tracheary elements, and impedes fungal hyphae growth in damaged or traumatized wood (Zimmermann, 1983; ZuÈrcher et al., 1985; Carlquist, 1988; Schmitt and Liese, 1994; Liese and Weiner, 1996; Canny, 1997; Jaquish and Ewers, 2001). Interpretations of structure/functions of tyloses (Zimmermann, 1983; Carlquist, 1988; Schmitt and Liese, 1994; Kozlowski and Pallardy, 1996; Canny, 1997) focus mainly on `risk management' aspects of closing off conduits that are no longer needed for current growth or that might increase susceptibility to fungal invasion of long interconnected vessels. Conduits most likely to be ®lled by tyloses are wider vessels, which are more vulnerable to cavitation, with numerous intimate contacts with live wood parenchyma or vessels near wound areas (Zimmermann, 1983; ZuÈrcher et al., 1985; Carlquist, 1988; Schmitt and Liese, 1994). Similar structure/function correlations are unstudied in non-lignophytes and in gymnosperms with tyloses (Chrysler, 1908; Holden, 1925; Ogura, 1972; Carlquist, 1996a; Carlquist and Schneider, 1999). Schneider and Carlquist (2000), however, discuss the possibility that some suites of pteridophyte vessel characters `might be related to xeromorphy or mesomorphy'. Abundant and well-preserved tyloses of Protopitys and other early gymnospermous lignophyte woods offer a unique opportunity to test these explanations because of the structural simplicity of these Early Carboniferous woods compared with those of modern dicot angiosperms. It is thus easy to track the physical courses of wood parenchyma and tyloses-®lled conduits of these early woods, which have only tracheary cells and rays (Figs 1±3). By contrast, modern, tyloses-prone woods, such as those of some Quercus (oak) or Robinia (locust) species are notable for their seasonal zones of heterogeneous cells (Esau, 1965). Protopitys is an odd plant in many ways. The sole member of its progymnosperm group Protopityales, it may have been the last of the progymnosperms, which are otherwise mostly found in Mid- to Late-Devonian deposits. It possessed secondary xylem and secondary phloem (Walton, 1969), which were produced by a bifacial vascular cambium with ray and fusiform initials. Protopitys is thus a non-seed-plant lignophyte (Beck and Wight, 1988). Fertile organs are known only for Protopitys scotica (Walton, 1957) and they compare with those of some Late Devonian Aneurophytales (basal progymnosperms), such as Tetraxylopteris schmidtii (Bonamo and Banks, 1967; Scheckler, 1986), although the spores of each are rather different (Smith, 1962; Taylor and Scheckler, 1996).

F I G . 2. Radial longitudinal sections of Protopitys buchiana wood (BOU 1500 BLR 01). A±D, Views of xylem rays (R) and tracheary cells with tyloses (T). A, Tyloses present in overlap areas of some rays as they pass over mid-regions of larger tracheary cells. Possible vessel perforations at arrow. B, Well-preserved area showing cellular/pseudocellular tracheary cell-®lling by tyloses (T), scalariform-bordered pitting, and ray cross-®eld pitting (R). C, Enlarged area showing tyloses (T) inside tracheary cells and ray cross-®eld pitting (R). D, Higher magni®cation of A to show possible vessel perforations by loss of pit membranes (lower arrows). Most other pits have intact pit membranes (e.g. upper arrows). Ray cross-®eld pitting (R) visible at top. Empty cell lumens (L) contrast with those ®lled with tyloses (T). Bars = 100 mm.

744

Scheckler and Galtier Ð Tyloses and Ecophysiology of Protopitys The wood samples studied show that while the overall pattern of tyloses-®lled tracheary elements is controlled by ray distribution (Figs 1C, 2A, B and 3A), it also correlates with the cross-sectional area of a tracheary cell. These cells vary in radial dimensions from 22±95 mm and in tangential dimensions from 15±100 mm (Fig. 1), but are quite uniform in any particular radial row of cells (Fig. 1A and B). In areas with multiple tyloses, it is mostly the larger tracheary cells that are blocked, whereas domains of smaller cells have fewer tyloses (Fig. 1C). Within a single cell, the larger midregion is far more likely to be ®lled than are the smaller ends, regardless of ray distribution (Figs 2A, B and 3A). In modern plants, vessel dimension is an important predictor of location of tyloses formation (Zimmermann, 1983; ZuÈrcher et al., 1985; Carlquist, 1988; Cochard and Tyree, 1990; Jaquish and Ewers, 2001). In ring-porous woods, for example, large earlywood vessels are usually blocked during the ®rst season of dormancy, while smaller latewood vessels might not be blocked until many years later, if at all. Factors other than wood construction, however, are important for tyloses presence/absence and distribution, perhaps including adaptations to particular environments. For example, the many New World species of the gymnosperm Gnetum have abundant tyloses in their vasicentric, paratracheal woods (Carlquist, 1996a), whereas the lianoid Indomalesian and Asiatic species, which also have vasicentric, paratracheal woods, seem to lack tyloses (Carlquist, 1996b). We are thus able to con®rm similar structure/function relationships as are seen in modern dicot angiosperms, where vulnerability to tyloses formation is correlated with conducting cell dimensions, proximity to wood parenchyma (ray or axial) and with changes of whole-tree physiological status during times of growth pause (Zimmermann, 1983; ZuÈrcher et al., 1985; Carlquist, 1988; Kozlowski and Pallardy, 1996; Jaquish and Ewers, 2001), or with the cavitation susceptibility of individual or groups of conducting cells (Canny, 1997, 1998). Furthermore, increases of heartwood-like xylem parenchyma (Hillis, 1987) of Protopitys are potential adaptations to stressful environments. The vast increase of wood parenchyma volume that resulted from formation of tyloses (compare Fig. 1A and B) might well have been an important physiological adaptation for periodic dormancy. Another intriguing possibility that we wish to consider is that P. buchiana might have had vessels in addition to large diameter tracheids. Pit membranes are visibly intact over most of the xylem cells' radial surfaces (Fig. 2B and D, upper arrows) except for some small circular zones where they are missing from between the scalariform-reticulate bars (Fig. 2A, arrow; D, lower arrows). Carlquist and Schneider (1999) document vessels in Marattiacae (and in

F I G . 3. Tangential longitudinal sections of P. buchiana wood. A, View of xylem rays (R) and a group of four cellular/pseudocellular tyloses (T and four cells beneath it) generated by protrusion of ray parenchyma of one ray (arrow) into an adjacent tracheary cell (to left of ray cell at arrow). Sample 491, Solms-Laubach 1897. B, View of undulate tracheary cells and density of ray contacts. BOU 1501 BLT 01. Bars = 100 mm.

Scheckler and Galtier Ð Tyloses and Ecophysiology of Protopitys many other fern families, e.g. Carlquist and Schneider, 1998, 1999, 2000, 2001; Schneider and Carlquist, 1998) and discuss their distribution in eusporangiate ferns. They cite McNichol's (1908) observation of tyloses in protoxylem and metaxylem of three genera of Marattiaceae as additional supporting evidence for their conclusion of the presence of vessels in this fern family. According to Carlquist and Schneider (1999), `tyloses . . . are virtually unknown in cells other than vessel elements'. However, it should be noted that tyloses are reported in tracheids of some species of Pinus (Chrysler, 1908; Coulter and Chamberlain, 1917; ZuÈrcher et al., 1985) and even in such cell types as ®bre-tracheids or laticifers (Gottwald, 1972; Carlquist, 1996a). Furthermore, the vessels of eusporangiate and most leptosporangiate ferns, or other fern-like plants, lack de®nitive perforation plates (Carlquist and Schneider, 1998, 1999, 2000, 2001; Schneider and Carlquist, 1998, 2000; Li et al., 1999). Instead, clusters of pits located almost anywhere along the vessel have perforated or porose pit membranes. These perforations are dif®cult or impossible to see by light microscopy of longitudinal sections of the living species, but were successfully visualized by Carlquist and Schneider and Li et al. by SEM of xylem macerates. Schneider and Carlquist (2000) note that, `the concept of ``vessel element'' is changing as SEM . . .' studies reveal a multiplicity of types . . .', whereby a gradient from typical tracheid to vessel exists depending on the degree and type of increase of porosity of pit membranes in both ferns and angiosperms (Carlquist and Schneider, 2001, 2002). Recent studies also show that pit membrane porosity and hydraulic resistance can vary signi®cantly with the physiological status of a plant and the types of ions carried in its xylem stream (Zwieniecki et al., 2001). Whether pteridophyte vessels are produced by pectin modi®cation, dissolution, or rupture of pit membranes is unknown. Growing evidence for the widespread occurrence of vessels in modern non-seed plants, documented by Carlquist and Schneider, and Li et al. (1999), suggests that similar perforation of pit membranes may have produced functional vessels in many extinct plant groups, including the progymnosperm P. buchiana. Regardless of whether Protopitys proves to have had vessels, the presence of tyloses in its larger xylem conducting cells apparently demonstrates that similar structure/function parameters operate in plants, where appropriate models are compared, regardless of geological time or plant phylogenetical grouping (Niklas, 1997; Carlquist and Schneider, 1999). The present study of Early Carboniferous woods con®rms that signi®cant ecophysiological information for a tree resides in the details of its cellular construction (Carlquist, 1975; Zimmermann and Brown, 1980). However, additional information (Chapman, 1994), such as position of a sample within a plant, may be needed before other local or regional palaeo-environmental conclusions can be made. The larger tracheary cell dimensions of P. buchiana compared with those of its gymnospermous lignophyte contemporaries (Galtier et al., 1998), and its abundance of ray parenchyma, provided the physical capacity to form tyloses. But its growth in an erratic, unstable volcanogenic environment

745

completed the requirements by promoting short periods of dormancy following, perhaps, episodes of cavitationproducing water stress. From this perspective, we predict that continued examination of fossil plant structure will provide additional tests of the universality of many of our current explanations and assumptions of the interplay between ecology, physiology and plant structure that are otherwise based almost entirely on studies of living plant systems. CONCLUSIONS Trunk woods of P. buchiana (Progymnospermopsida) from Early Carboniferous volcaniclastic sediments of the Vosges region of France, and similar strata in Poland, have abundant tyloses that are distributed according to ray geometry, ray overlap with large tracheary cells, and growth layer boundaries. Tyloses thus have a radial as well as a tangential component of distribution that may be related to erratic or, perhaps, recurring seasonal episodes of water stress and dormancy. This pattern of tyloses distribution, along with the depositional setting in a volcanogenic terrain of subtropical paleolatitudes, suggests that this is a paleoecophysiological response of this early lignophyte tree to life in a seasonal wet/dry climate zone with many locally disturbed habitats. ACKNOWLEDGEMENTS We thank J. Guiraud (Montpellier) for preparation of thinsections of the Vosges material and Jeff Liston, Hunterian Museum, University of Glasgow, for the loan of SolmsLaubach sections 489, 490 and 491; Drs G. W. Rothwell and E. L. Schneider for their helpful comments and critical evaluations of the manuscript; and VPI&SU (1998 sabbatical research leave), the National Science Foundation (9728719 and 9982787), and Harvard University (1999 Bullard Fellowship) for their support to S.E.S. L I T E RA TU R E C I TE D Bateman RM, Crane PR, DiMichele WA, Kenrick P, Rowe NP, Speck T, Stein WE. 1998. Early evolution of land plants: phylogeny, physiology and ecology of the primary terrestrial radiation. Annual Review of Ecology and Systematics 29: 263±292. Beck CB, Wight DC. 1988. Progymnosperms. In: Beck CB, ed. Origin and evolution of gymnosperms. New York: Columbia University Press, 1±84. Bertrand P. 1909. EÂtudes sur la fronde des ZygopteÂrideÂes. Lille: Imprimerie L. Danel. Bierhorst DW. 1960. Observations on tracheary elements. Phytomorphology 10: 249±305. Bonamo PM, Banks HP. 1967. Tetraxylopteris schmidtii: its fertile parts and its relationships within the Aneurophytales. American Journal of Botany 54: 755±768. Bowe LM, Coat G, dePamphilis CW. 2000. Phylogeny of seed plants based on all three genomic compartments: Extant gymnosperms are monophyletic and Gnetales' closest relatives are conifers. Proceedings of the National Academy of Sciences of the USA 97: 4092±4097. Canny MJ. 1995. A new theory for the ascent of sap ± cohesion supported by tissue pressure. Annals of Botany 75: 343±357. Canny MJ. 1997. Tyloses and the maintenance of transpiration. Annals of Botany 80: 565±570.

746

Scheckler and Galtier Ð Tyloses and Ecophysiology of Protopitys

Canny MJ. 1998. Applications of the compensating pressure theory of water transport. American Journal of Botany 85: 907±909. Canny MJ. 2001a. Contributions to the debate on water transport. American Journal of Botany 88: 43±46. Canny MJ. 2001b. Embolisms and re®lling in the maize leaf lamina, and the role of the protoxylem lacuna. American Journal of Botany 88: 47±51. Carlquist S. 1975. Ecological strategies of xylem evolution. Berkeley: University of California Press. Carlquist S. 1988. Comparative wood anatomy. Berlin: Springer-Verlag. Carlquist S. 1996a. Wood, bark, and stem anatomy of New World species of Gnetum. Botanical Journal of the Linnean Society 120: 1±19. Carlquist S. 1996b. Wood and bark anatomy of lianoid Indomalesian and Asiatic species of Gnetum. Botanical Journal of the Linnean Society 121: 1±24. Carlquist S, Schneider EL. 1998. SEM studies on vessels in ferns. 10. Selected Osmundaceae and Schizaeaceae. International Journal of Plant Sciences 159: 788±797. Carlquist S, Schneider EL. 1999. SEM studies on vessels in ferns. 12. Marattiaceae, with comments on vessel patterns in eusporangiate ferns. American Journal of Botany 86: 457±464. Carlquist S, Schneider EL. 2000. SEM studies on vessels in ferns. 14. Ceratopteris, and the signi®cance of widespread vessel occurrence in ferns. Aquatic Botany 66: 1±8. Carlquist S, Schneider EL. 2001. Vessels in ferns: structural, ecological, and evolutionary signi®cance. American Journal of Botany 88: 1±13. Carlquist S, Schneider EL. 2002. The tracheid-vessel element transition in angiosperms involves multiple independent features: cladistic consequences. American Journal of Botany 89: 185±195. Chambers JQ, Higuchi N, Schimel JP. 1998. Ancient trees in Amazonia. Nature 391: 135±136. Chapman JL. 1994. Distinguishing internal developmental characteristics from external palaeoenvironmental effects in fossil wood. Review of Palaeobotany and Palynology 81: 19±32. Chaw SM, Parkinson CL, Cheng Y, Vincent TM, Palmer JD. 2000. Seed plant phylogeny inferred from all three plant genomes: Monophyly of extant gymnosperms and origin of Gnetales from conifers. Proceedings of the National Academy of Sciences of the USA 97: 4086±4091. Chrysler MA. 1908. Tyloses in tracheids of conifers. New Phytologist 7: 198±204. Cochard H, Tyree MT. 1990. Xylem dysfunction in Quercus: vessel sizes, tyloses, cavitation and seasonal changes in embolism. Tree Physiology 6: 393±407. Comstock JP. 1999. Why Canny's theory doesn't hold water. American Journal of Botany 86: 1077±1081. Coulter JM, Chamberlain CJ. 1917. Morphology of gymnosperms. Revised edition. Chicago: University of Chicago Press. Crane PR. 1985. Phylogenetic analysis of seed plants and the origin of angiosperms. Annals of the Missouri Botanical Garden 72: 716±793. Dannenhoffer JM, Bonamo PM. 1989. Rellimia thomsonii from the Givetian of New York: secondary growth in three orders of branching. American Journal of Botany 76: 1312±1325. Doyle JA. 1996. Seed plant phylogeny and the relationships of Gnetales. International Journal of Plant Sciences 157 (Suppl.): S3±S39. Doyle JA. 1998. Phylogeny of vascular plants. Annual Review of Ecology and Systematics 29: 567±599. Doyle JA, Donoghue MJ. 1986. Seed plant phylogeny and the origin of angiosperms: an experimental cladistic approach. Botanical Review 52: 321±431. Esau K. 1965. Plant anatomy. 2nd edn. New York: John Wiley & Sons. Fahn A. 1982. Plant anatomy. 3rd edn. Oxford: Pergamon Press. Fahn A. 1997. Plant anatomy. 5th edn. Oxford: Butterworth/Heinemann Publishers. Galtier J, Scott AC. 1991. Stanwoodia, a new genus of probable early gymnosperms from the Dinantian of East Kirkton, Scotland. Transactions of the Royal Society of Edinburgh, Earth Sciences 82: 113±123. Galtier J, Scott AC. 1994. Arborescent gymnosperms from the ViseÂan of East Kirkton, West Lothian, Scotland. Transactions of the Royal Society of Edinburgh, Earth Sciences 84: 261±266. Galtier J, Schneider J-L, Grauvogel-Stamm L. 1998. Arborescent gymnosperms and the occurrence of Protopitys from the Lower

carboniferous of the Vosges, France. Review of Palaeobotany and Palynology 99: 203±215. Galtier J, Brown RE, Scott AC, Rex GM, Rowe NP. 1993. A late Dinantian ¯ora from Weaklaw, East Lothian, Scotland. Special Papers in Palaeontology 49: 57±74. Gottwald H. 1972. Tyloses in ®bre tracheids. Wood Science Technology 6: 121±127. Graham SW, Olmstead RG. 2000. Utility of 17 chloroplast genes for inferring the phylogeny of the basal angiosperms. American Journal of Botany 87: 1712±1730. Hillis WE. 1987. Heartwood and tree exudates. Berlin: Springer-Verlag. Holden HS. 1925. On the occurrence of cavity parenchyma and tyloses in ferns. Journal of the Linnean Society. Botany 47: 141±153. Holden HS. 1930. On the structure and af®nities of Ankyropteris corrugata. Philosophical Transactions of the Royal Society London B 218: 79±113. Jaquish LL, Ewers FW. 2001. Seasonal conductivity and embolism in the roots and stems of two clonal ring-porous trees, Sassafras albidum (Lauraceae) and Rhus typhina (Anacardiaceae). American Journal of Botany 88: 206±212. Joy KW, Willis AJ, Lacey WS. 1956. A rapid cellulose peel technique in paleobotany. Annals of Botany 20: 635±637. Kenrick P. 1999. The family tree ¯owers. Nature 402: 358±359. Kenrick P, Crane P. 1997a. The origin and early evolution of plants on land. Nature 389: 33±39. Kenrick P, Crane PR. 1997b. The origin and early diversi®cation of land plants: a cladistic study. Washington and London: Smithsonian Institution Press. Kozlowski TT, Pallardy SG. 1996. Physiology of woody plants. 2nd edn. San Diego: Academic Press. Li R, Zhang D, Zhang H. 1999. Scanning electron microscope observations on the vessels of ferns: Adiantum, Matteuccia, and Osmunda from Heilongjiang Province, China. International Journal of Plant Sciences 160: 595±602. Liese W, Weiner G. 1996. Ageing of bamboo culms. A review. Wood Science and Technology 30: 77±89. McNichol M. 1908. On cavity parenchyma and tyloses in ferns. Annals of Botany 22: 401±413. Mathews S, Donoghue MJ. 1999. The root of angiosperm phylogeny inferred from duplicate phytochrome genes. Science 286: 947±950. Matten LC, Banks HP. 1967. Relationship between the Devonian progymnosperm genera Sphenoxylon and Tetraxylopteris. Bulletin of the Torrey Botanical Club 94: 321±333. Mauseth JD. 1988. Plant anatomy. Menlo Park: Benjamin Cummings Publishing. Milburn JA. 1996. Sap ascent in vascular plants: challengers to the Cohesion Theory ignore the signi®cance of immature xylem and the recycling of MuÈnch water. Annals of Botany 78: 399±407. Murakami Y, Funada R, Sano Y, Ohtani J. 1999. The differentiation of contact cells and isolation cells in the xylem ray parenchyma of Populus maximowiczii. Annals of Botany 84: 429±435. Niklas KJ. 1997. The evolutionary biology of plants. Chicago: University of Chicago Press. Nilsen ET, Orcutt DM. 1996. Plant stress physiology. Berlin: SpringerVerlag. Nixon KC, Crepet WL, Stevenson D, Friis EM. 1994. A re-evaluation of seed plant phylogeny. Annals of the Missouri Botanical Garden 81: 484±533. Ogura Y. 1972. Comparative anatomy of vegetative organs of the pteridophytes. Encyclopedia of Plant Anatomy/Handbuch der P¯anzenanatomie. Berlin: GebruÈder Borntraeger. Pryer KM, Schneider H, Smith AR, Cran®ll R, Wolf PG, Hunt JS, Sipes SD. 2001. Horsetails and ferns are a monophyletic group and the closest living relatives to seed plants. Nature 409: 618±622. Rex GM. 1986. The preservation and palaeoecology of the Lower Carboniferous silici®ed plant deposits at Esnost, near Autun, France. Geobios 19: 773±800. Rothwell GW, Serbet R. 1994. Lignophyte phylogeny and the evolution of spermatophytes: a numerical cladistic analysis. Systematic Botany 19: 443±482. Schaltegger U, Schneider JL, Maurin JC, Corfu F. 1996. Precise U-Pb chronometry of 345±340 Ma old magmatism related to

Scheckler and Galtier Ð Tyloses and Ecophysiology of Protopitys syn-convergence extension in the Southern Vosges (Central Variscan Belt). Earth and Planetary Science Letters 144: 403±419. Scheckler SE. 1975. A fertile axis of Triloboxylon ashlandicum, a progymnosperm from the Upper Devonian of New York. American Journal of Botany 62: 923±934. Scheckler SE. 1986. Evidence for the relationship between Protopityales and Aneurophytales (Progymnospermopsida). American Journal of Botany 73: 705±706. Scheckler SE, Banks HP. 1971. Anatomy and relationships of some Devonian progymnosperms from New York. American Journal of Botany 58: 737±751. Scheckler SE, Lindsay-Postnikoff DL, Chameroy EJ. 2000. Late Devonian forests with the ®rst large trees (Archaeopteris). American Journal of Botany 87: 76. Schmitt U, Liese W. 1994. Wound tyloses in Robinia pseudoacacia L. IAWA Journal 15: 157±160. Schneider EL, Carlquist S. 1998. SEM studies on vessels in ferns. 9. Dicranopteris (Gleicheniaceae) and vessel patterns in leptosporangiate ferns. American Journal of Botany 85: 1028±1032. Schneider EL, Carlquist S. 2000. SEM studies on vessels in ferns. 17. Psilotaceae. American Journal of Botany 87: 176±181. Scotese CR, Golonka J. 1992. PALEOMAP paleogeographic atlas. PALEOMAP progress report no. 20. Arlington: Department of Geology, University of Texas. Scotese CR, McKerrow WS. 1990. Revised world maps and introduction. In: McKerrow WS, Scotese CR, eds. Paleozoic palaeogeography and biogeography. Geological Society Memoir 12: 1±21. Sherwin HW, Pammenter NW, February E, Vander Willigen C, Farrant JM. 1998. Xylem hydraulic characteristics, water relations and wood anatomy of the resurrection plant Myrothamnus ¯abellifolius Welw. Annals of Botany 81: 567±575. Smith DL. 1962. Three fructi®cations from the Scottish Lower Carboniferous. Palaeontology 5: 225±237. È ber die in den Kalksteinen des Kulm Solms-Laubach H Graf zu. 1893. U von GlaÈtzisch-Falkenberg in Schlesien enthaltenen structurbietenden P¯anzenreste. II. Botanische Zeitung 51: 197±210. Soltis PS, Soltis DE, Chase M. 1999. Angiosperm phylogeny inferred from multiple genes as a research tool for comparative biology. Nature 402: 402±404.

747

Stiller V, Sperry JS. 1999. Canny's compensating pressure theory fails a test. American Journal of Botany 86: 1082±1086. Taylor TN, Scheckler SE. 1996. Devonian spore ultrastructure: Rhabdosporites. Maurice Streel Commemorative Volume. Review of Palaeobotany and Palynology 93: 147±158. Van der Voo R. 1993. Paleomagnetism of the Atlantic Tethys and Iapetus oceans. Cambridge: Cambridge University Press. Walton J. 1949. On some Lower Carboniferous Equisetineae from the Clyde area. I Protocalamostachys arranensis gen. et sp. nov. ± a hitherto undescribed type of strobilus. II The nodal structure of Asterocalamites GoÈpperti Solms sp. Transactions of the Royal Society of Edinburgh 61: 729±736. Walton J. 1957. On Protopitys (GoÈppert): with a description of a fertile specimen `Protopitys scotica' sp. nov, from the Calciferous Sandstone Series of Dunbartonshire. Transactions of the Royal Society of Edinburgh 63: 333±340. Walton J. 1969. On the structure of a silici®ed stem of Protopitys and roots associated with it from the Carboniferous Limestone, Lower Carboniferous (Mississippian) of Yorkshire, England. American Journal of Botany 56: 808±813. Williams JH, Friedman WE. 2002. Identi®cation of diploid endosperm in an early angiosperm lineage. Nature 415: 522±526. Williamson WC. 1877. On the organization of the fossil plants of the Coal-measures. VIII. Philosophical Transactions of the Royal Society of London 167: 213±270, pl. 5±16. Zimmermann MH. 1983. Xylem structure and the ascent of sap. Berlin: Springer-Verlag. Zimmermann MH, Brown CL. 1980. Trees: structure and function. Berlin: Springer-Verlag. Zimmermann U, Meinzer F, Bentrup F-W. 1995. How does water ascend in tall trees and other vascular plants? Annals of Botany 76: 545±551. ZuÈrcher E, Kucera LJ, Bosshard HH. 1985. Bildung und morphologie der thyllen. Eine LiteraturuÈbersicht. Vierteljahrschrift Naturforschung Gesellschaft ZuÈrich 130: 311±333. Zwieniecki MA, Melcher PJ, Holbrook NM. 2001. Hydrogel control of xylem hydraulic resistance in plants. Science 291: 1059±1062.