Taxodioxylon (Hartig) Gothan, 1905

IAMANDEI, STĂNILĂ, IAMANDEI, EUGENIA, VELITZELOS, DIMITRIOS & VELITZELOS, EVANGELOS, 2022, Palaeoxylotomical Studies In The Cenozoic Petrified Forests Of Greece. Part Two - Conifers, Acta Palaeontologica Romaniae 18 (1), pp. 65-111 : 76-79

Genus Taxodioxylon (Hartig) Gothan, 1905 Taxodioxylon gypsaceum (Göppert) Kräusel, 1949

Fig. 5 , a-i.

Material

From the studied samples, 7 specimens with cupressaceous xylostructure of "taxodiaceous" type closer to that of current Sequoia , were grouped under the genus Taxodioxylon . These samples were collected from Aegean area, either from dispersed fragments of petrified wood, or from in situ big silicified trunks found in the continental part of Greece, in the late Oligocene deposits of Evros: 76

sample Aet.315 from Aetohory and sample Lfk.313 from Lefkimi. Also, some samples were collected from the insular part, from early Miocene deposits of Limnos Island: Li.134, 135, and of Lesbos Island, the samples Lsv.335, 336 and 337. The studied material is registered under the above specified numbers as “Velitzelos Collection” and stored in the Collection of the Faculty of Geology and Geoenvironment, of NKUA.

Microscopic description

Growth rings – are relatively wide in cross section, of 11- 40(-65) cells or more, often with slightly gradual transition from earlywood to latewood, sometimes quite abrupt, giving a distinct aspect to the growth-rings boundaries, marked by some rows of smaller thick walled late-wood tracheids, as final wood. Normal resin canals absent, but sometimes tissue or even canals with traumatic character appear.

Tracheids – with polygonal shape (with 4-5-6 sides) in cross section, often variably sized, having the radial / tangential diameter of 35-65(-80) / 25-45(-70) μm and relatively thin walls - of 3-5 μm the double wall in the early-wood, usually gradually diminishing in size toward the late-wood, which is represented by 2-5(-13) rows of tracheids with smaller lumen, of 10-18 / 15-25 μm in diameters and thicker walls – of 6-10 μm double wall. The cells are regularly arranged in 1-9 radial rows between two successive rays, sometimes with intercellular spaces. Their density is 1500-1716 tracheids per mm 2. In longitudinal view, the tracheids show rounded or tapered endings, are of high type but their length is often difficult to measure. Tracheidal pitting on the radial walls are usually 1-3-seriate, oppositely arranged, as abietinean round bordered pits, with 18-22 μm in diameter, with large chambers and round to slightly elliptic apertures. Sometimes certain irregularities appear, as storeys of opposite pairs of pits in the uniseriate row, or even as small portions biseriate. Crassulae are usually present. On the tangential tracheidal walls pitting is absent but, few uniseriate smaller pits, of 8-10(15) μm can seldom appear. Also, sometimes, on the endings of tracheids, warty areas appear. Helical or callitroid thickenings absent. Organic deposits absent.

Axial parenchyma - appears rather abundant in cross section, in short tangential lines or in diffuse arrangement, dispersed throughout the entire growth ring as smaller cells among the tracheids, or in the terminal wood as slightly irregular short tangential lines, usually with dark content. In the longitudinal sections, the parenchyma shows thin and smooth transverse end-walls or, sometimes, with 1(-2) weak nodules. Inside the parenchyma cells, globular or granular resin content appear.

Rays - appear numerous, linear and thin in cross section. In tangential section they appear exclusively uniseriate sometimes with some biseriate storeys, of middle to high type, having 1-16 cells or up to 30 cells high or, 77 sometimes, even more than 60 cells. The ray cells, in tangential view, appear usually polygonal-rounded or oval of 12-16-20 μm, rarely cribrated by small simple pits. Often lateral intercellular spaces are present. The ray-density is 6-10 rays per tangential mm. Regarding their composition, the rays are homogeneous, without ray-tracheids, having all procumbent parenchymal ray-cells with smooth tangential end-walls. Their horizontal walls are smooth and pitted. Indentures are not visible or are missing. The cross-field pitting is of taxodioid type, usually with 1-3 pits in one horizontal row in the normal cross fields, but up to 6 pits in the taller marginal cross fields, where they are arranged in two superposed rows. These pits appear as taxodioid oculipores of 5-8 / 8-10 μm, with large to very large inclined apertures. In cross fields with axial parenchyma small cupressoid pits appear.

Resin canals – are usually absent but, sometimes, axial traumatic canals appear.

Mineral inclusions – are absent.

Affinities and discussions

The microscopic description of the studied samples of fossil wood collected from Aegean area, from late Oligocene and early Miocene sediments, showed xylotomical details that are characteristic to Cupressaceae family of "taxodiaceous" type, having radial pits on tracheids up to triseriate, opposite and with crassulae, end-wall of ray-cells thin and smooth and 1-3 taxodioid pits in horizontal line in the cross-fields. These xylotomical details are very specific to Sequoia wood-type (Greguss, 1955; 1967).

Fossil wood with such anatomical details was described, in time, either as Taxodioxylon (Hartig) Gothan, 1905 or as Sequoioxylon Torrey, 1923 , both fossil genera being considered, in time, as valid fossil members of the former “ Taxodiaceae family”. Today, after advanced studies, all the “taxodiaceous genera" are incorporated in the family Cupressaceae sensu lato (Gadek et al., 2000). This family includes now 27–30 genera distributed in 7 subfamilies, some of them more exotic, like Taiwanioideae L.C. Li , Athrotaxidoideae L. C. Li and Callitroideae Saxton. The main subfamilies are: Cupressoideae Rich. ex Sweet , Sequoioideae Saxton , Taxodioideae Endl. ex Koch and Cunninghamioideae (Zucc. ex Endl.) Quinn , which comprise most the known cupressaceous taxa, living especially in the Northern hemisphere (Gadek et al., 2000; Xiang & Li, 2005; Farjon, 2005a, 2010; Christenhusz et al., 2011).

The fossil genus Taxodioxylon was created by Hartig (1848), taking as type-species the Taxodioxylon goeppertii Hartig. This species was considerd by Philippe (1993) as equivalent to the current Sequoia sempervirens (D. Don) Endl. , and this fact was admitted long before, by Gothan (1905, 1906), who emended first time the fossil genus Taxodioxylon . And, evaluating the structure of many previously described fossil "taxodiaceous” wood remains, it can be observed that many of them show simi- 78

larities, with both the genera, Sequoioxylon and Taxodioxylon , so that the separation of the two genera remained problematic for long time.

But, Kräusel (1949) has established Taxodioxylon as a comprehensive fossil genus that became a “pocket-

genus” for the fossil correspondents of many current members of the former “ Taxodiaceae ” family, making difficult to differentiate between several taxodiaceouswood types.

Later, Greguss (1967) considered that the name Taxodioxylon should remain reserved for structures characterized by the presence of up to 8 nodular thickenings on the horizontal end-walls of the axial parenchyma, which is typical to the current Taxodium . For fossil wood of Sequoia - type he used the name Sequoioxylon as a new genus since, most probably, he hasn't n the published paper of Torrey (1923) and he described some new species (Greguss, 1967), transferring the species name ‘ gypsaceum’ from Taxodioxylon to Sequoioxylon , as new combination.

However, Privé-Gill (1977) considered that the diagnosis of the Taxodioxylon -genus is sufficiently comprehensive and that establishing new areas of competence for other genera is likely to complicate the determination of fossil wood, since a great intraspecific variability can be observed even in the current genus Sequoia , especially regarding the aspect of the axial parenchyma. A consistent discussion on the nomenclatural and taxonomical treatment of the taxodiaceous wood genera is conducted by Dolezych (2011), who suggested a revision of the morphogenus Taxodioxylon , and till there she proposed to continue to use the name Taxodioxylon , with a more precise description regardind the distinct growth rings, the ray cell walls predominately smooth, the radial pitting on tracheids mostly biseriate, the presence of wood parenchyma, and the cross-field pits which are predominantely taxodioid.

• Using the table of Kräusel (1949, p. 168) in or- der to identify our studied specimens, in which we observed that the transverse wall of the axial parenchyma is usually thin and smooth and the cross-field pits are taxodioid or even cupressoid in the cross-fields with axial parenchyma, details which are typical also to the current Sequoia sempervirens . Such fossil wood-structure is usually identified as Taxodioxylon gypsaceum (Göpp.) Kräusel. The next presented identifications of T. gypsaceum are very similar to our studied specimens.

• van der Burgh (1973) described a T. gypsaceum and a T. germanicum (Greguss) van der Burgh (initially described by Greguss, in 1967 as Sequoioxylon germanicum ), both considered as extinct forms corresponding to the extant Sequoia sempervirens , noting that the thin end-walls in the ray-cells could be an important differentiator to other Taxodioxylon species.

• Later, by the study of some wood specimens from the quarry Schleenhainn, Germany, near Leipzig, collected from an upper Eocene brown coal seam, a Taxodioxylon gypsaceum (in van der Burgh & Meijer, 1996), which is very similar to our specimens. Here is made an interesting discussion on this species considered as a highly variable polyphyletic species, which was probably widespread in the European Cenozoic, and had a wide range of variation. However, since too few fossils wood samples have been studied to describe a fossil species, not sufficient to justify a biological variability.

• Gottwald (1992) described a Taxodioxylon sp. from the Eocene of Helmstedt, Germany, as having xylotomic details of Sequoia - type.

• Süss & Velitzelos (1997) have described some species of Taxodioxylon as: T. gypsaceum , T. albertense (Penhallow) Shimakura, 1940 and T. pseudoalbertense Nishida et H. Nishida, 1985 - as stem-wood, and a new species - T. megalonissum Süss & Velitzelos - as root-wood, which mostly resembles that of the current Sequoia sempervirens . Also, compiling about 48 Mesozoic and Cenozoic species of Taxodioxylon , they made a revision of this genus, the first one after Kräusel (1949), and they gave a new genus diagnosis, in order to include also details characteristic for the root-wood (Süss & Velitzelos, 1997), but they did it without a corect emendation, as Dolezych (2011) observed, and it was not followed by all the palaeoxylotomists.

• Biondi, E., Brugiapaglia, E. (1991, 2000) identified some fossil woods from Dunarobba ( Italy) as Taxodioxylon gypsaceum but, some xylotomical arguments point to Glyptostrobus wood-type ( Martinetto, 1994; Vassio et al., 2008).

• Selmeier (2001), describing a fossil wood from Czech Rep., identified it as Taxodioxylon gypsaceum and said that most of the Cenozoic wood-remains (not from upper Cretaceous) assigned to this species confirm a highly variable polyphyletic species in which the pits of cross field vary from taxodioid to glyptostroboid or cupressoid.

• More recently, Teodoridis & Sakala (2008), Koutecký & Sakala (2015) and Koutecký et al. (2019) describing a Taxodioxylon gypsaceum (stem and root-wood) from Czech Rep., confirm the equivalence with the xylotomy of the modern Sequoia , citing Privé-Gill (1977), Dolezych & van der Burgh (2004), Dolezych & Schneider (2006b). In fact, they send to Sequoia wood-type or to an extinct form, Quasisequoia couttsiae (Heer) Kunzmann 1999 , a species also described by van der Burgh & Meijer (1996) from the Late Eocene of Schleenhain, Germany, in a coal-layer with the wood of Taxodioxylon gypsaceum and leaves and cones belonging exclusively to Quasisequoia couttsiae . More than this, is proposed for wood of Quasisequoia Srinivasan et Friis a genus name - Quasisequoioxylon Dolezych et Van der Burgh (in Dolezych, 2005), which would be intermediate between Cupressinoxylon Goeppert , characterized by cupressoid cross field pits and Taxodioxylon (Hartig) Gothan , with 2-3 taxodioid cross field pits on radial tracheidal walls (Dolezych 2005: 256; Dolezych & Schnei- der, 2006b).

In the same time, there were many palaeoxylologists in the world ( U.S., Canada, Russia, Hungary, Romania, Ukraine, China, Turkey) who have admitted the validity of the genus Sequoioxylon Torrey 1923 , and have described species under this genus name: Andrews (1936); Greguss (1967); Nagy (1969); Mathiensen (1970); Roy & Stewart (1971); Petrescu & Popa (1971); Petrescu & Dragastan (1971); Petrescu & Popa (1971); Petrescu (1978); Özgüven (1971); Blokhina (1997- 2004); Blokhina et al. (2000); Sakınç et al. (2007); Iamandei & Iamandei (2000, 2016, 2017); Iamandei et al. (2004; 2008a; 2012a, 2020); YiTiemei et al. (2013); Afonin (2013); Uhl et al. (2014); Tian et al. (2018); Li et al. (2018); Akkemik, (2019a); Akkemik et al., (2005, 2009, 2013, 2019acd, 2020); Acarca-Bayam et al. (2018); Özgenc et al. (2018); Güngör et al. (2019); Polat et al. (2019).

Bu, as a conclusion, the genus name Sequoioxylon Torrey, 1923 should not be more usable, since is no more considered a valid genus ( IFPNI, 2014), being in contradiction with ICN rules (Turland et al., 2018 - Shenzen Code) and all the described species under this genus must be reconsidered as species of Taxodioxylon .

Thus, our studied specimens show xylotomical characters which are very similar to those of the extant Sequoia sempervirens , having relatively wide growth rings, abrupt transition, distinct boundaries, no resin canals, (but sometimes traumatic canals can appear); tracheids polygonal in cross section, with radial pitting 1-3-seriate, as round pits of 18-22 μm, sometimes slightly irregularly arranged, and with crassulae; tangential pitting usually absent; axial parenchyma in short tangential lines or in diffuse arrangement, with transverse end walls thin and smooth, rarely weakly nodular; rays exclusively uniseriate, rarely with biseriate storeys, of middle to high type, with lateral intercellular spaces, homogeneous, with smooth end-wall, no indentures, 1-3 taxodioid type pits in cross fields, or 4- 6 in 2(-3) superposed rows. These xylotomical details are similar or almost identical to those described as on fossil wood from the European sites and identified as Taxodioxylon gypsaceum .

Based upon all these discussions which show great similarity in xylotomical features between the studied material and the species diagnosis and other published descriptions of Taxodioxylon gypsaceum , we assign our studied specimens to Taxodioxylon gypsaceum (Göpp.) Kräusel, 1949 , taking also into account the paper of Süss & Velitzelos (1997) who described the same species, also from the Aegean area.

Figures

Fig. 5 a-i, (graphic scale). Taxodioxylon gypsaceum (Göpp.) Kräusel, 1949. (Specimen Lfk.313). a-c: Cross section - quite gradual transition, distinct growth rings; parenchyma few, diffuse, or in small lines; d-f: Tangential section - unpitted tracheids; parenchyma with end-wall smooth or weakly nodular; g–i: Radial section - tracheidal pitting 1-3-seriate, opposite; rays homocellular, cells all procumbent; poorly preserved cross fields with 1-3 taxodioid pits (arrow), more numerous in taller marginal cross fields, in 2 superposed rows or slightly irregular.