Dispersed plant mesofossils from the Middle Mississippian of eastern Germany: Bryophytes, pteridophytes and gymnosperms

Review of Palaeobotany and Palynology 193 (2013) 38–56

Contents lists available at SciVerse ScienceDirect

Review of Palaeobotany and Palynology journal homepage: www.elsevier.com/locate/revpalbo

Research paper

Dispersed plant mesofossils from the Middle Mississippian of eastern Germany: Bryophytes, pteridophytes and gymnosperms Maren Hübers a,⁎, Hans Kerp a, Jörg W. Schneider b, Birgit Gaitzsch b a b

Forschungsstelle für Paläobotanik am Institut für Geologie und Paläontologie, Westfälische Wilhelms-Universität Münster, Schlossplatz 9, 48143 Münster, Germany Geologisches Institut, TU Bergakademie Freiberg, Bernhard-von-Cotta-Strasse 2, 09599 Freiberg, Germany

a r t i c l e

i n f o

Article history: Received 24 July 2012 Received in revised form 2 December 2012 Accepted 19 January 2013 Available online 8 February 2013 Keywords: plant mesofossils bulk samples bryophytes sphenophytes pteridosperms Visean

a b s t r a c t Plant remains with organic-matter preservation are uncommon in the Mississippian due to the general rareness of Mississippian terrestrial deposits. Moreover, the fossils are often strongly affected by post-sedimentary processes. However, bulk-macerated samples of the upper Visean of Chemnitz-Glösa (Germany) contain several well-preserved dispersed organic remains including cuticle fragments. Even though it is difficult to assign these remains to specific taxa and they probably represent only part of the vegetation, they provide important information on Mississippian floras. The cuticles are assigned to sphenopsids and gymnosperms. In addition, several bryophyte remains were found; nine morphologically different types are recognised, which represent at least five different taxa. This is remarkable because only a few bryophytes have been recorded from the Carboniferous to date. © 2013 Elsevier B.V. All rights reserved.

1. Introduction Mississippian floras are rather poorly known in comparison to Pennsylvanian floras, due to the relatively limited number of plant-bearing localities and the preservation of the material. Plants are mostly preserved as petrifactions and as adpressions without organic matter. Reliable reconstructions of Mississippian plants are rare and natural affinities often remain unclear because they are based on limited data. Preservation of organic material, notably cuticles, is generally rare due to post-sedimentary processes such as high thermal alteration. This contribution provides descriptions of several types of dispersed organic plant remains from a temporary outcrop in the upper Visean of Chemnitz-Glösa, Saxony, Germany. The mesofossils were obtained by bulk maceration of terrestrial shales containing recognisable remains of Lyginopteris, Lepidodendron and macroscopically unidentifiable plant debris. Although at present precise taxonomic assignments of these plant remains are not possible, the diversity and preservation are noteworthy. Apart from lycopsid cuticles (Hübers et al., 2011a), and sphenophyte and pteridosperm cuticles described here, the latter with leaf-colonizing fungi (Hübers et al., 2011b), several types of bryophyte remains, sporangia and megaspores have been found. Lycopsid and sphenopsid cuticles are generally not very common, except for sphenophylls. The occurrence of bryophyte remains, of which several types can be distinguished within the Glösa material, is remarkable, because only a single putative bryophyte taxon (Thomas, 1972) has ⁎ Corresponding author. E-mail address: [email protected] (M. Hübers). 0034-6667/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.revpalbo.2013.01.006

previously been reported from Mississippian rocks. Embryophytes evolved before the tracheophytes, probably in the Ordovician–Silurian (Kenrick and Crane, 1997), or maybe even earlier judging from the cryptospore record (Taylor and Strother, 2008, 2009). The different types of Mississippian bryophytes reported here may shed new light on this problem. Although most attention is given to bryophytes, a number of other mesofossils should not remain unnoticed. Several sphenophyte and pteridosperm cuticles are described as well as some remains that cannot be attributed yet. The bryophytes in particular contribute important information for reconstructing wetland ecosystems during the Mississippian, a time interval in which terrestrial biotas are still comparatively poorly known (Scott et al., 1984). 2. Locality and material Eleven samples were collected in 1999 from a temporary roadcut exposure along the A4 motorway near the Chemnitz-Glösa exit, Saxony, Germany (50°52′34.67″ N; 12°54′38.73″ E). The rocks that represent the so-called “Early Molasses” facies of the Variscan Orogen belong to the Hainichen Subgroup of the Hainichen Basin (Fig. 1). The Hainichen Basin is dated as late Visean (Schneider et al., 2005; Gaitzsch et al., 2010). The relics of the Hainichen Basin are allochthonous, situated between Variscan high-grade metamorphic complexes of the Erzgebirge and the Saxonian Granulite Massif (Schneider et al., 2005). The Hainichen Basin has yielded a comparatively rich macroflora that was described by Hartung (1938); for an updated overview of the so-called Borna-Hainichen flora we refer to Kerp et al. (2006). Glösa is located c. 1 km NW of the classical localities at Borna-Heinersdorf, in

M. Hübers et al. / Review of Palaeobotany and Palynology 193 (2013) 38–56

39

Fig. 1. Geological map of southern Saxony showing the positions of Carboniferous–Permian basins: A = Hainichen Basin, B = Flöha Basin, C = Oelsnitz Basin, D = Zwickau Basin (after Schneider et al., 2005).

the southwestern part of the basin. Hartung (1938) recorded over 30 taxa from Borna, including sphenopsids (Archaeocalamites, Sphenophyllum, Bowmanites, Pothocites, Mesocalamites), lycopsids (Lepidodendron, Lepidophloios, Lepidostrobus) and many pteridophylls (Adiantites, Alloiopteris, Aneimites, Anisopteris, Cardiopteridium, Diplothmema, Fryopsis, Lyginopteris, Neuropteris, Pecopteris, Spathulopteris, Sphenopteris and Zeilleria). This flora shows a clear differentiation into hygrophilous (coal-forming) associations and those growing on better drained soils. The most common macroscopically recognisable plant remains at Chemnitz-Glösa are Lyginopteris bermudensiformis (Schlotheim) Patteisky

and Lepidodendron lossenii Weiss; the latter can form monotypic stands. Archaeocalamites grew in monospecific stands along rivers and lake shores. A more detailed account on the Borna-Hainichen flora and a comparison with coeval floras from western and central Europe are currently in preparation. The Hainichen Subgroup consists of two formations, the Ortelsdorf Formation (70–700 m) and the Berthelsdorf Formation (20–>200 m). Both formations contain plant remains (Fig. 2). Rößler and Schneider (1997) recognised four plant communities in the Berthelsdorf Formation that was exposed in the sandpits of Borna and in the so-called

Fig. 2. Stratigraphy, lithology, facies and fossil content of the Hainichen Subgroup (after Gaitzsch et al., 2010). The arrow indicates the position of the samples.

40

M. Hübers et al. / Review of Palaeobotany and Palynology 193 (2013) 38–56

Lingke´sche clay pit in Glösa, based on data of Máyas (1920), Hartung (1938) and Nindel (1955): 1. hydro- to hygrophilous plant associations along streams and rivers, partially with monospecific Archaeocalamites stands, sometimes associated with arborescent lycopsids (mostly Lepidodendron veltheimii Sternberg, Lepidodendron volkmannianum Sternberg and Lepidophloios laricinus Sternberg) and various ferns, 2. hygrophilous backswamp associations with arborescent lycopsids (mostly Lepidodendron lossenii), Archaeocalamites and various pteridophylls, 3. monospecific stands with Sphenophyllum stimulosum Hartung and Bowmanites sphenasterophylloides Hartung, 4. hygro- to mesophilous, monospecific fern- and pteridosperm stands with Alloiopteris quercifolia (Goeppert) Potonié, Fryopsis polymorpha (Goeppert) Wolfe emend. Hübers et Kerp, neuropterids and numerous sphenopterids as pioneer plants on ruderal soils. The Ortelsdorf Formation and the overlying Berthelsdorf Formation are both fining-upward sequences, from coarse-grained fan delta respectively alluvial fan to alluvial plain deposits with thin coal seams. The boundary between the two formations is marked by a change from alluvial plain deposits with sand- and siltstones and coal seams to conglomerates (Fig. 2). The Berthelsdorf Formation that was exposed in the Borna sandpits was deposited in an alluvial plain environment, with thick sandstone beds representing the main channels.

The sediments in the abandoned Lingke´sche clay pit probably represent a coeval floodplain facies with small, short-lived swamps. The sediments of the upper part of the Ortelsdorf Formation were deposited under similar conditions in similar environments. As the setting and flora are comparable, it is assumed that the same plant associations existed. The entire, up to 800 m thick fining-upward sequence of the Ortelsdorf Formation is dominated by compositionally immature terrigenous clastics. The lower part of the formation shows the vertical and lateral transition from a subaqueous nearshore fan delta environment into terrestrial distal alluvial fan deposits. The upper part consists of floodplain and channel deposits, locally with swamp deposits characterised by decimetre-thick coal seams and carbonaceous shales. The uppermost 60 m of the Ortelsdorf Formation, exposed at Chemnitz-Glösa, show interbeddings of decimetre to metre thick greyish siltstone beds, sandy siltstones and mostly poorly sorted channel conglomerates. The very shallow but wide channel geometry and bed-load transport indicate braided to anastomosing river systems. The overbank siltstones are often finely rooted; in places stigmarian root horizons were observed. Layers with current-oriented, in places chaotic accumulated trunks and twigs of Archaeocalamites, lycopsids and other plants indicate flooding events. The sediment samples described here are from an interval between 40 and 60 m below the top of the formation, where centimetre to decimetre thick coaly siltstones and coal seams occur together with several root horizons. The environment can be characterised as unstable with fast sedimentation and common reworking hampering the

Plate I. Bryophyta.

1 2 3 4 5 6

Type B-Ia, overview of a leaf fragment consisting of two parts. The central costa is marked by more elongate cells. Accession no. Glösa-Profil KS27-074; scale bar = 100 μm. Type B-Ib, overview of a ?laminar fragment. Accession no. Glösa-Profil KS11-031; scale bar = 100 μm. Type B-Ia, detail of Plate I, 1 showing the pattern of striae and grooves leading to the pores; scale bar = 10 μm. Type B-Ib, detail of Plate I, 2 showing cell pattern with high anticlinal walls, arrows indicating two periclinal walls; scale bar = 50 μm. Type B-Ib, detail of Plate I, 2 showing porous cells with otherwise smooth surface, pores showing thickened rim; scale bar = 10 μm. Type B-Ia, detail of Plate I, 1 showing pores of the periclinal walls and high anticlinal walls. Note the difference between the “height” of the longitudinal and transversal anticlinal walls; scale bar = 30 μm.

Plate II. Bryophyta. (see on page 42) 1 2 3 4 5 6 7 8

Type B-IIc, overview showing two types of cells, the elongate cells forming fenestrae enclosing the hyaline cells. Accession no. Glösa-Profil KS11-090; scale bar = 100 μm. Type B-IIc, detail of Plate II, 1, arrows on three different walls indicating the bistratose nature; scale bar = 50 μm. Type B-IIc, detail of Plate II, 1, arrow indicates the two superimposed layers of elongate cells demonstrating the bistratose nature; scale bar = 50 μm. Type B-IIa, overview showing the regular network of elongate cells forming fenestrae enclosing isodiametric hyaline cells (see also Fig. 4). Accession no. Glösa-Profil KSgrPr-068; scale bar = 100 μm. Type B-IIa, detail of Plate II, 4 showing the individual elongate cells and hyaline cells; scale bar = 50 μm. Type B-IIa, detail of Plate II, 4 showing the outlines of the elongate cells and the isodiametric hyaline cells. Note the endings of the elongate cells in the corners; scale bar = 50 μm. Type B-IIb, showing an overview with two different types of fenestrae; polygonal hyaline cells bearing papillae. Accession no. Glösa-Profil KSgrPr-008; scale bar = 100 μm. Type B-IIb, accession no. Glösa-Profil KSgrPr-054 showing papillae; scale bar = 50 μm.

Plate III. Bryophyta. (see on page 43)

1 2 3 4 5

Type Type Type Type Type

B-IIIa, overview showing elongate cells with pores and fibrils. Accession no. Glösa-Profil KS11-029; scale bar = 50 μm. B-IIIa, detail of Plate III, 1 showing fibrils; scale bar = 10 μm. B-IIIb, overview showing short cells and pores. Accession no. Glösa-Profil KS27-048; scale bar = 100 μm. B-IIIb, detail of Plate III, 3 showing elevated pore with narrow rim; scale bar = 10 μm. B-IIIa, overview showing cells with oblique end walls. Accession no. Glösa-Profil KS11-002; scale bar = 50 μm.

Plate IV. ?Bryophyta. (see on page 44)

1 2 3 4 5 6

Type B-IV, overview showing different cell types. Accession no. Glösa-Profil KS27-143; scale bar = 100 μm. Type B-IV, detail of Plate IV, 1 showing double walls (grey arrow) and septae (black arrows); scale bar = 100 μm. Type B-V, showing a differentiation into several cell types: normal cells, larger lighter-coloured cells and slightly darker cells with thicker walls. Accession no. Glösa-Profil KS27-147; scale bar = 100 μm. Type B-V, detail of Plate IV, 3 showing normal cells with mode of division, larger lighter cells and darker cells; scale bar = 100 μm. Type B-V, fragment showing a differentiation into normal cells and larger lighter cells. Accession no. Glösa-Profil KS27-145; scale bar = 100 μm. Type B-V, detail of Plate IV, 4, showing the large cells with intact periclinal walls; scale bar = 100 μm.

M. Hübers et al. / Review of Palaeobotany and Palynology 193 (2013) 38–56

Plate I.

41

42

M. Hübers et al. / Review of Palaeobotany and Palynology 193 (2013) 38–56

Plate II (caption on page 40).

M. Hübers et al. / Review of Palaeobotany and Palynology 193 (2013) 38–56

Plate III (caption on page 40).

43

44

M. Hübers et al. / Review of Palaeobotany and Palynology 193 (2013) 38–56

Plate IV (caption on page 40).

M. Hübers et al. / Review of Palaeobotany and Palynology 193 (2013) 38–56

development of forest vegetation. The vegetation type rather corresponds to the above-mentioned plant community 4, i.e., pioneering, ruderal plants on episodically disturbed soils. However, local occurrences of hygrophilous plant associations of backswamps with arborescent lycophytes (mostly Lepidodendron lossenii), Archaeocalamites and different pteridophylls (plant community 2) were present. All plant groups typical for this environment are represented in the Glösa material. The upper Ortelsdorf Formation at Glösa has been dated palynologically (Jäger and Wierich, 2006), and comprises the NM (Raistrickia nigra–Triquitrites marginatus) and VF (Tripartites vetustus–Rotaspora fracta) biozones of Clayton et al. (1977). The moss-bearing samples are from the uppermost part that is late Visean (early Brigantian) in age (VF-Zone) (Fig. 2). Although the single-zircon Pb/Pb evaporation age of 330 ± 4 Ma of a rhyolitic tuff at the base of the coal-bearing fossiliferous upper part of the formation (Gehmlich et al., 1998; Gaitzsch et al., 2010) is not very precise, it is consistent with the palynological dating.

3. Methods Sediment samples were bulk-macerated by treating them with 48% hydrofluoric acid to isolate the organic remains. Of one sample, no. 2324/2004, small pieces of cuticle-bearing sediment were carefully removed from the surface and treated with 48% hydrofluoric acid. Sample preparation was carried out as described by Kerp (1990) and Kerp and Krings (1999). The isolated plant remains were bleached using Schulze's reagent (20% HNO3 +KClO3), cleaned and finally treated with a 4% potassium hydroxide solution (KOHaq). Plant remains that were still too dark were bleached again with Schulze's reagent at a higher concentration of HNO3 (35%). Bleached cuticles were dehydrated in glycerol for a few days and finally mounted on microscope slides using glycerine-jelly. A total number of about 650 preparations were made. They are stored in the collection of the Forschungsstelle für Paläobotanik am Institut für Geologie und Paläontologie, Westfälische Wilhelms-Universität Münster, Germany. Significant specimens were photographed with a Nikon DS-5M camera mounted on a Leitz Diaplan microscope with Nomarski interference contrast. For details on methods see Kerp and Bomfleur (2011). SEM-pictures were taken using a SEM JEOL 840 and analySIS software.

4. Results The samples from Chemnitz-Glösa yielded a large number of organic remains. Some cuticles have already been described (Hübers et al., 2011a, 2011b). The following descriptions refer to the most important specimens that have not been described so far. The number of specimens listed under “Material” only refers to those specimens that have been prepared after maceration. However, it is no indication of the abundance or rareness of particular types in the bulk macerations. It is not possible to assign the various remains described here to particular genera or species, nor is it realistic to establish new taxa based on these fragmentary remains. Hence, they are informally described as types, which are classified as follows: type numbers of putative bryophyte remains are numbered with the prefix B, those of putative sphenophytes with the prefix S, those of putative pteridosperms with the prefix P, and those of forms that cannot be classified yet by the prefix U, followed by roman numbers. Within the bryophytes one type may comprise up to three different forms (a to c) that show some general similarity, but not enough to classify them as a single type. These types are characterised below, with brief discussions on the taxonomy and occurrences. Other aspects are considered in the general discussion.

45

4.1. Bryophyte remains Several organic remains assignable to bryophytes have been found. They are characterised and classified on the basis of overall morphological similarities to other fossil and living bryophytes. Three basic types assigned to the mosses have recently briefly been described and discussed by Hübers and Kerp (2012) and commented by Taylor and Taylor (2012). They are described, illustrated and discussed here in much more detail and seven forms are recognised within the three main types. In addition, two types provisionally assigned to the liverworts are documented. Due to the fragmentary nature of the material only informal types are described and it cannot be excluded that two or more types belonged to the same plant. Even though it is unclear how many natural species are involved, we anticipate that at least three species of mosses are represented in the Glösa material. 4.1.1. Type B-Ia (Plate I, 1, 3, 6) Material: 1 slide with two specimens, apparently belonging together: Glösa-Profil KS27-074. Description: Narrow double-layered, unistratose leaf-like structure. Cells mostly elongate, more or less rectangular to diamond-shaped, with shorter polygonal cells; cells 62–165 μm (generally ~90 μm) long and 17–42 μm (generally ~25–30 μm) wide, aligned in rows. Anticlinal walls straight to slightly bent, up to 15 μm high. Periclinal walls perforated. Pores of periclinal walls distributed regularly, up to 4 μm in diameter, without a thickened rim. The surface of the cells shows a pattern of fine, short longitudinal striae and grooves (Plate I, 3). Remarks: The two fragments apparently belong together, because (1) they show the same cell pattern that continues from one into the other fragment, and (2) the broken margins fit perfectly. Together they form a part of an almost complete leaf. This classifies this bryophyte type as a unistratose leafy moss, but a more precise assignment is not possible. The periclinal layers are pressed together and the anticlinal walls are collapsed. As a result, the anticlinal walls appear as double parallel lines, particularly the longitudinal and oblique ones; the transverse walls look much thinner. The double lines are the lines of attachment of the anticlinal walls to the lower and upper periclinal layers. The distance between both lines indicates the height of the periclinal wall, i.e., the cell height. 4.1.2. Type B-Ib (Plate I, 2, 4, 5) Material: 1 specimen: Glösa-Profil KS11-031. Description: Double-layered, unistratose structure. Cells polygonal, slightly elongate, more or less rectangular to diamond-shaped, 58–89 μm long and 22–49 μm wide. Anticlinal walls straight. Periclinal walls perforated. Pores with thickened rim (Plate I, 5); pore diameter not exceeding 4 μm. Remarks: Type B-Ia and Type B-Ib are quite similar and have been differentiated on the basis of differences in cell size and the more regular alignment of the cells in Type B-Ia. The pores in Type B-Ib have a thin rim which seems to be lacking in Type B-Ia. However, it cannot be excluded that both types represent the same species. 4.1.3. Type B-IIa (Plate II, 4–6) Material: 4 specimens: Glösa-Profil KSgrPr-64, -068; Glösa-Profil KS27-115, -135. Description: Laminar, unistratose structure with dimorphic cells, i.e., thicker-walled elongate cells and hyaline cells. Elongate cells, usually being slightly darker, forming a more or less regular polygonal network consisting of quadrangular, pentangular or hexangular, isodiametric fenestrae that contain hyaline cells. Elongate cells generally 24–65 μm long (in KSgrPr-64 up to 111 μm) and 12–26 μm wide, often with more or less longitudinal folds, rounded at the endings, fenestrae generally 18–69 μm (in KSgrPr-64 up to 95 μm) in diameter. The hyaline periclinal walls are strongly wrinkled.

46

M. Hübers et al. / Review of Palaeobotany and Palynology 193 (2013) 38–56

Remarks: The dimorphic nature of the cells is very similar to that of the Protosphagnales (Neuburg, 1960; Ignatov, 1990), an artificial group of leafy mosses that is hitherto only known from the Permian. Hence, the remains from Chemnitz-Glösa would be the earliest record of this extinct group of leafy mosses.

and the fibrils strongly resemble those of leucocysts of extant Sphagnum species. Such structures have not been demonstrated from the rich moss floras from the Permian of Russia that include a number of taxa assigned to the Protosphagnales (e.g., Neuburg, 1956, 1960; Fefilova, 1973, 1978; Ignatov, 1990).

4.1.4. Type B-IIb (Plate II, 7, 8) Material: 4 specimens: Glösa-Profil KSgrPr-008, -054, -070; Glösa-Profil KS11-093. Description: Laminar unistratose structure with dimorphic cells. Elongate dark, thicker-walled cells forming a more or less regular network with quadrangular, pentangular or hexangular isodiametric fenestrae similar to Type B-IIa, and a zone with more elongate fenestrae. Fenestrae containing hyaline cells. Elongate polygonal fenestrae, up to 137 μm long and 42 μm wide. Cells forming the elongate fenestrae up to 118 μm long and 16 μm wide. Polygonal, isodiametric fenestrae 28–77 μm in diameter. Cells forming polygonal fenestrae up to 45 μm long and 19 μm wide. Anticlinal walls of all cell types straight. Periclinal walls of the hyaline cells rather thick, the isodiametric ones usually with one or two small thickenings (papillae?) per cell. Remarks: This type is quite similar to Type B-IIa, but it has two different types of fenestrae; the elongate ones probably being closer to or part of a costa. Moreover, most of the hyaline cells seem to bear papillae. However, it is possible that these thickenings are only artefacts of preservation. Hence, it cannot be excluded that Type B-IIa and Type B-IIb represent the same species.

4.1.7. Type B-IIIb (Plate III, 3, 4) Material: 2 specimens: Glösa-Profil KS27-048; Glösa-Profil KS11-100. Description: Cells short, rectangular or with oblique end walls, 40–110 μm long and 14–37 μm wide, with one pore or rarely two pores per cell. Anticlinal walls thick. Pores slightly elevated, bordered by a narrow rim. Pores lemniscate to rounded; pores 10–30 μm in diameter. Remarks: These remains resemble the surface of the stem cortex of certain Sphagnum species, e.g., S. quinquefarium (Lindberg ex Braithwaite) Warnstorf, S. russowii Warnstorf (cf. Lange, 1982, fig. 62, 91). The similarity in pore shape may suggest that Type B-IIIa and Type B-IIIb could represent the same taxon. However, differences in cell shape and size and the presence of fibrils in Type B-IIIa that are absent in Type B-IIIb for the moment preclude unifying these remains in a single type.

4.1.5. Type B-IIc (Plate II, 1–3) Material: 1 specimen: Glösa-Profil KS11-090. Description: Bistratose structure with three layers of periclinal walls. Cells dimorphic. Two superimposed layers of narrow, thicker-walled cells forming a more or less regular network with elongate pentangular or hexangular fenestrae. Elongate cells forming a network 41–211 μm long and 18–37 μm wide. Anticlinal walls straight. Fenestrae 201– 330 μm long and 31–50 μm wide, containing thinner-walled cells. Periclinal walls of the larger cells filling the fenestrae smooth. Remarks: The bistratose nature is evident from the presence of three layers of periclinal walls (Plate II, 2 arrows) and the two superimposed obliquely lying elongate cells (Plate II, 3 arrow); the dark lines in the elongate cells probably mark the boundaries between the two superimposed layers. Regarding the cell dimorphism and the resulting fenestrate structure, an assignment to the Protosphagnales seems obvious. The order Protosphagnales has been established on the basis of vegetative shoots (Neuburg, 1956, 1960); reproductive structures are still unknown. The leaves of Protosphagnales are comparable to those of Sphagnales but differ in having a distinct costa. The often narrower dark-coloured cells form a network enclosing larger light-coloured cells that have been interpreted as hyaline. Fenestrae are either absent or one or two per hyaline cell; in the latter case they form a cross dividing the cell into more or less equal parts. The bistratose nature of this fragment suggests that it is not part of the lamina but of the costal region. The fenestra/cell pattern is comparable to that of the left part of Type B-IIb, but fenestral cells are larger. However, it cannot be excluded that this type, Type B-IIa and Type B-IIb represent the same species. 4.1.6. Type B-IIIa (Plate III, 1, 2, 5) Material: 2 specimens: Glösa-Profil KS11-002, -029. Description: Cells elongate, rectangular or with oblique end walls, 95–213 μm long and 23–54 μm wide, with one pore or rarely two pores per cell and obliquely positioned fibrils (Plate III, 1, 2). Anticlinal walls thick. Pores slightly elevated, bordered by a narrow rim. Pores lemniscate to rounded; pore diameter 16–26 μm. Remarks: Only two specimens, each showing a very characteristic morphology, have been recovered. The slightly elevated water pores

4.1.8. Type B-IV (Plate IV, 1, 2) Material:1 specimen: Glösa-Profil KS27-143. Description: Cells in lateral parts of the fragment mostly quadratic, sometimes rectangular or pentagonal to nearly triangular in the most basal part of the structure. Cells in central part of the fragment generally narrower and more elongate. Cell walls thin; in all cell types thin septae (cross-walls) may occur, either longitudinal ones, transverse ones or both (Plate IV, 2, black arrows). Cell size variable, up to 257 μm long and 66 μm wide. Anticlinal walls straight. Periclinal walls smooth. Remarks: Although we have no doubt that the other types described above can be assigned to bryophytes, it is with some reluctance that we classify this specimen among this group. Type B-IV shows no or only very superficial similarities to known pteridophyte and gymnosperm cuticles. The provisional attribution to the bryophytes is based on the basis of the form, size and arrangement of the cells, and on the nature of the cell walls. This type shows the presence of septae (Plate IV, 2, black arrows). Such septae have been illustrated for Pallaviciniites devonicus (Hueber) Schuster (Hueber, 1961, pl. 2, 7) and Muscites sp. (Gomankov and Meyen, 1986, pl. 2, 7–8, 12–14). 4.1.9. Type B-V (Plate IV, 3–6) Material: 5 specimens: Glösa-Profil KS27-144, -145, -147, -149, -157. Description: Thickness varying within each fragment, but mostly thin. Two types of cells occur: the first type is represented by polygonal to sometimes more or less rectangular cells, more or less arranged in rows. Size ranges from just a few microns to 44 μm length and 25 μm width. Cells of second type intercalated more or less regularly, uniformly directed, round to oval, about 30–70 μm long and 20–40 μm wide. The periclinal walls of these large cells are much thinner than in the cells of the first type. Ornamentation absent, except for some folds or small thickenings on the periclinal walls of the larger cells. Remarks: Also in this case we are reluctant to ascribe the fragments to the bryophytes. A pteridosperm, sphenophyte or lycopsid affinity seems unlikely, because the cell pattern in our Type B-V is quite different from that reflected in the cuticles of these latter groups. The lack of stomata, even in the largest specimens, is remarkable. Therefore it is unlikely that these are lycopsid or sphenopsid cuticles. Pteridosperm cuticles may lack stomata but they usually show a different cell pattern. Especially the cell division of polygonal cells into two or into four cells with more or less perpendicular walls appears to be typical for our material. The general cell pattern (Plate IV, 3) is somewhat reminiscent of the liverwort Pallaviciniites devonicus as described by Hueber (1961), and to a lesser degree, of Metzgeriothallus sharonae (Hernick et al., 2008), but cells in our

M. Hübers et al. / Review of Palaeobotany and Palynology 193 (2013) 38–56

specimens tend to be better arranged in rows. Remarkable in our specimens is the differentiation in two distinct cell types that can be distinguished by their colour and size. The larger cells have much thinner periclinal walls than the smaller (normal) ones. Several of these larger cells are damaged, but the intact ones show at higher magnification that these are not stomata but cells with comparatively thin walls (Plate IV, 5, 6). Also P. devonicus and M. sharonae show a differentiation into different cell types, but these can rather be distinguished by their colour than by their size. In both taxa the aberrant cells are darker. Hueber (1961) interpreted the darker cells in P. devonicus as storage or secretory cells, and Hernick et al. (2008) interpreted the darker ones in M. sharonae as storage cells. The difference between the latter two taxa and our specimens is that in Pallaviciniites and Metzgeriothallus the periclinal and particularly the anticlinal walls of the “abnormal” cells are thicker than those of the normal cells, whereas in our specimens the larger cells have normal anticlinal walls and thinner periclinal walls (Plate IV, 3–6). However, our largest specimen (Plate IV, 3, 4) also shows some conspicuous darker cells interspersed among the normal and lighter ones (Plate IV, 3, 4). Type B-V can only be assigned to bryophytes provisionally.

4.1.10. Classification of bryophytes and putative bryophytes Altogether nine different types of putative bryophytes have been recovered from bulk macerations from Chemnitz-Glösa. Although some forms may look very similar, we decided to treat them as separate forms because no definite proof can be given that they indeed represent the same species. Because of the fragmentary nature of the material we refrain from formally naming them. Although several types are compared to fossil or extant mosses, assignments are difficult because the early, pre-Permian history of mosses is still very poorly known. We provisionally classify these nine forms in five categories. The types of the first category comprising Type B-Ia, Type B-Ib and possibly also Type B-IV show the anticlinal walls as double (parallel) lines (Fig. 3; Plate I, 1, 4, 6; Plate IV, 2, grey arrow), either very prominently or just in a few cells. This feature is very common in bryophytes. It has been illustrated for various fossil forms (e.g., Walton, 1925, 1928; Hueber, 1961; Busche, 1968; Gomankov and Meyen, 1986; Ignatov, 1990). The anticlinal walls of the unistratose laminae or thalli with two periclinal layers are laterally compressed. The two lines that are visible are the intersections of the anticlinal walls with the lower and the upper periclinal walls. The distance between the two lines is the height of the cell. Type B-IV also shows a few double walls, although this is not as clear as in the other types. The types of the second category, comprising Types B-IIa–c, show a clear cell dimorphism (Fig. 4). Nevertheless, these types show some similarity to Types B-Ia,b and Type B-IV because the elongate cells forming the network may superficially look like anticlinal walls with

47

Fig. 4. Schematic drawing of the cell pattern in Type B-II. HC=hyaline cells, EC=elongate cells, CW=cell wall.

“double lines” as they occur in Types B-Ia,b and Type B-IV. Although the darker specimens (e.g., Plate II, 7, 8) do not clearly show that the network enclosing the hyaline cells consists of rings or polygons of elongate cells, the lighter-coloured specimens leave no doubt. The elongate cells forming the network are rounded at the endings and not folded like the compressed anticlinal walls in Types B-Ia,b. Moreover, in Types B-IIa–c all cells forming the network are equally wide, whereas the width (= height) of the collapsed and folded anticlinal walls varies depending on the direction in which they are collapsed or folded. A third category of bryophyte remains, comprising Type B-IIIa and Type B-IIIb, is characterised by the presence of very prominent, slightly elevated, narrowly rimmed water pores. Moreover, Type B-IIIa shows the presence of fibrils. Particularly this latter type shows a striking similarity to leucocysts of extant Sphagnales. Such features have not been observed before in other late Palaeozoic mosses. The resemblance to Type B-IIIb, particularly regarding the occurrence and shape of the water pores, indicates that the latter type belongs in the same category and it may even represent the same species. Although the general similarity to some bryophytes illustrated in the literature is striking, the incompleteness of the specimens at hand and the limited amount of material, do not allow a further classification or identification. Types B-I–III apparently belong to mosses, the nature of Type B-IV and Type B-V is unclear. 4.2. Sphenophyte cuticles Sphenophytes, notably Archaeocalamites, are common constituents of Visean floras of Saxony. Although sphenophyte cuticles are usually very rare, except for those of Sphenophyllum, we refer several cuticles to the sphenophytes. Sphenophyllum cuticles often have strongly sinuous cell walls and very simple stomata consisting of two reniform guard cells. Cuticles of calamitaleans/equisetaleans also have simple stomata with two subsidiary cells; the guard cells frequently show radial striae, the so-called wood lamellae; the stomata are normally arranged in longitudinal rows; the normal epidermal cells are more or less rectangular, often elongate, and arranged in longitudinal rows.

Fig. 3. Schematic drawing of the cell pattern in Type B-Ia,b and Type B-IV with two cell layers and high anticlinal walls (AW).

4.2.1. Type S-I (Plate V, 7) Material: 1 specimen: Glösa-Profil KS11-022. Description: Epidermal cells of different size and shape, from isodiametric, ~17 μm in diameter, to elongate, 25–30 μm long and 15–20 μm

48

M. Hübers et al. / Review of Palaeobotany and Palynology 193 (2013) 38–56

wide. Anticlinal walls strongly sinuous with an amplitude of up to 3 μm. Periclinal walls smooth, cutinisation increasing towards the middle of each epidermal cell. Stomata absent. Remarks: The cell structure of Type S-I is similar to the cell structure of several Sphenophyllum species (e.g., Abbot, 1958; Batenburg, 1977; Barthel, 1997; Yao et al., 2000). We are not aware of any other Carboniferous plant group, showing this kind of structure and therefore assign Type S-I to Sphenophyllum.

4.2.2. Type S-II (Plate V, 1–3) Material: 1 specimen: Glösa-Profil KS27-050. Description: Epidermal cells more or less rectangular, except for those directly bordering the stomata, 13–27 μm long and 15–27 μm wide, anticlinal walls straight, sometimes slightly curved. Periclinal walls not ornamented, except for the subsidiary cells. Stomata sunken, intercalated, densely arranged, longitudinally oriented. Stomatal complexes consisting of two guard cells encompassed by a pair of subsidiary cells forming a narrow ring, 23–43 × 13–29 μm. Guard cells poorly preserved, apparently with short wood lamellae. Subsidiary cells with conspicuous crescent-shaped ledges. Remarks: A sphenophytalean nature for these cuticles is based on the distribution and the longitudinal orientation of the stomata, the rather simple organisation of the stomatal apparati, and the presence of wood lamellae on the guard cells (Plate V, 3). Particularly, the presence of wood lamellae is an important feature; such bands are common in equisetaleans (Hauke, 1957; Pant and Kidwai, 1968) and have also been demonstrated for calamitaleans (Good, 1971, pl. 49, 28). It should be noted that calamitalean and equisetalean cuticles are rarely preserved.

4.2.3. Type S-III (Plate V, 4–6) Material: 3 specimens: Glösa-Profil KS27-159, -160, -162. Description: Cuticle differentiated into a stomatiferous zone and an adjacent non-stomatiferous zone. Epidermal cells of stomatiferous zone more or less rectangular, except for those directly bordering the stomata, generally 15–25 μm long and wide, anticlinal walls straight, sometimes slightly curved. Periclinal walls not ornamented, except for the subsidiary cells. Stomata sunken, intercalated, densely arranged, longitudinally oriented. Stomatal complexes consisting of guard cells encompassed by a pair of subsidiary cells forming a narrow ring, 30–41 × 21–31 μm. Subsidiary cells with very prominent, strongly sinuous ledges. Adjacent zone consisting of elongate cells, 24–95 μm long, arranged in rows, with intercalated round cells, about 20–30 μm in diameter, with decreasing cutinisation from the margin to the centre. Remarks: Cuticle Type S-III is very similar to Type S-II. It differs in the occurrence of an adjacent stomata-free zone, but the lack of such a zone in Type S-II is probably an artefact. The main difference is the ornamentation of the subsidiary cells, Type S-II having crescent shaped ledges, whereas Type S-III has strongly sinuous ledges. It is possible that at least one of the two cuticles types belongs to Archaeocalamites, which is probably the most common sphenophyte found in Mississippian deposits.

4.3. Pteridosperms The flora of the Hainichen Basin comprises a wide variety of pteridophylls (Hartung, 1938). Most of these taxa probably belong to the pteridosperms, some are ferns and the natural affinity of some is still unclear. Regarding the great diversity in pteridosperm species, their common occurrence and the fact that most pteridosperm cuticles have an excellent preservation potential, it is not surprising that the majority of plant mesofossils obtained by bulk maceration are pteridosperm cuticles, although the diversity seems not very high. Although many Carboniferous floras yield well-preserved cuticles (e.g., Krings and Kerp, 1997; Krings et al., 2001; Cleal et al., 2007; Šimůnek, 2008, 2010), only very few cuticles of Mississippian plants have been described to date, e.g., Lacey (1962), Thomas (1970, 1972). We here describe three cuticle types assigned to the pteridosperms. Some of the cuticles classified under “Plant remains incertae sedis” may also represent pteridosperms, but they are too small to show enough features to justify an assignment to this group. 4.3.1. Type P-I (Plate VI, 1–3) Material: 27 specimens: Glösa-2324/2004-013, -019, -20, -021, -22, -024, -029, -034, -040, -041, -47, -051, -072, -073, -074, -075, -079, -080, -089, -093, -094, -096, -098, -101, -119, -120; Glösa-Profil KS27-170. Description: Cuticle thick. Epidermal cells elongate, c. 70–125 μm long and 5–12 μm wide. Periclinal walls bearing dark round papillae and trichomes. Papillae generally 15–20 μm in diameter. Trichomes 16 μm in diameter at the base, c. 4 μm at the apex, recurved. Cuticle of one leaf surface bearing more stomata than the other. Stomata arranged more or less regularly, longitudinally oriented, usually between 45 and 50 μm long and c. 15 μm wide. Guard cells with wood lamellae and two stomatal ledges. Subsidiary cells differ from normal epidermal cells in being stronger cutinised. 4.3.2. Type P-II (Plate VII, 1–5) Material: 17 specimens: Glösa-Profil KS27-024, -040, -046, -049, -053, -064, -066, -071, -073, -081, -082, -089, -091, -117, -118, -130, -139. Description: Cuticle thin, sometimes forming glandular structures. Epidermal cells indistinct, variable in form and size. Anticlinal walls straight to slightly curved. Periclinal walls smooth. Stomata irregularly arranged, 25–40 μm long and 10–15 μm wide. Guard cells very thin. Stomatal apparati rectangular, with four trapezoidal subsidiary cells; anticlinal walls of subsidiary cells thick, particularly the outer and inner ones; periclinal walls of subsidiary cells thinner than those of normal epidermal cells. 4.3.3. Type P-III (Plate VII, 8) Material: 8 specimens: Glösa-Profil KSgrPr-024, -040; Glösa-Profil KS34/35-046; Glösa-Profil KStonPr-001, -002, -003, -005, -020. Description: Cuticle thick, with differentiation into longitudinal (?costal and intracostal) zones. Epidermal cells aligned in rows, polygonal– elongate in form, generally about 100–160 μm long and 40 μm wide. Anticlinal walls more or less straight. Periclinal walls rough, unornamented. Stomata absent.

Plate V. Sphenopsida.

1 2 3 4 5 6 7

Type S-II, cuticle showing the distribution and orientation of the stomata. Accession no. Glösa-Profil KS27-050; scale bar = 100 μm. Type S-II, detail of Plate V, 1; scale bar = 100 μm. Type S-II, detail of Plate V, 1 showing stomata with crescent-shaped ledges on the subsidiary cells (arrows) and the wood lamellae of the guard cells in the upper stoma; scale bar = 10 μm. Type S-III, overview showing a stomatiferous zone and part of an adjacent non-stomatiferous zone (right below). Accession no. Glösa-Profil KS27-159; scale bar = 100 μm. Type S-III, detail of Plate V, 4 showing lower part of Plate V, 4 with stomatiferous and non-stomatiferous zones; scale bar = 100 μm. Type S-III. Stoma with sinuous ledges on subsidiary cells and well-developed wood lamellae. Accession no. Glösa-Profil KS27-160; scale bar = 10 μm. Type S-I with strongly sinuous anticlinal walls. Accession no. Glösa-Profil KS11-022; scale bar = 100 μm.

M. Hübers et al. / Review of Palaeobotany and Palynology 193 (2013) 38–56

49

50

M. Hübers et al. / Review of Palaeobotany and Palynology 193 (2013) 38–56

Plate VI. Pteridospermopsida. 1. 2. 3.

Type P-I. Thick cuticle with stomata, papillae and trichomes. Accession no. Glösa-2324/2004-093; scale bar = 50 μm. Type P-I, detail of Plate VI, 1 showing stoma with wood lamellae of guard cells and two stomatal ledges; scale bar = 10 μm. Type P-I. Lower and upper leaf surface cuticles. Accession no. Glösa-2324/2004-120; scale bar = 100 μm.

Remarks: The cuticles of Type P-III, with an estimated number of a few hundred specimens, are by far the most frequent ones in the samples from Glösa. It is therefore most likely, that these cuticles belong to Lyginopteris bermudensiformis, which is the most common macrofossil found at Glösa. Cuticles of L. bermudensiformis have been isolated from the surface of several samples. They are mostly very brittle and hardly show any structure. We managed to isolate a single leaf fragment showing clear cell outlines (Plate VII, 6), which can be long and narrow

with tapering endings (Plate VII, 7, right hand side) or polygonal– elongate (Plate VII, 7, left hand side). The shape of the latter corresponds to that of the cells in Type P-III. 4.4. Plant remains incertae sedis A number of cuticles that cannot be attributed to any group of plants have been encountered. These are briefly characterised below.

Plate VII. Pteridospermopsida.

1 2 3 4 5 6 7 8

Type P-II. Cuticle showing stomata, anticlinal walls of epidermal cells very weakly cutinised. Accession no. Glösa-Profil KS27-040; scale bar = 50 μm. Type P-II showing multicellular glandular structure. Accession no. Glösa-Profil KS27-082; scale bar = 100 μm. Type P-II, detail of Plate VII, 2; scale bar = 100 μm. Type P-II. Stomatal apparatus with four trapezoidal subsidiary cells; scale bar = 10 μm. Type P-II, same specimen as on Plate VII, 4, SEM micrograph of lower surface; scale bar = 10 μm. Leaf cuticle of Lyginopteris bermudensiformis; scale bar = 500 μm. Detail of Plate VII, 6 showing elongate cells with tapering endings (right) and polygonal–elongate cells (left); scale bar = 100 μm. Type P-V. Cuticle with polygonal–elongate cells. Accession no. Glösa-Profil KStonPr-002; scale bar = 100 μm.

Plate VIII. Plant cuticles incertae sedis. (see on page 52)

1 2 3 4 5 6

Type Type Type Type Type Type

U-I, overview showing outlines of epidermal cells and a single stoma. Accession no. Glösa-2324/2004-099; scale bar = 30 μm. U-I, detail of Plate VIII, 1 showing stoma with strongly cutinised guard cells; scale bar = 20 μm. U-II, overview showing thin cuticle with hair bases (arrows) and stoma. Accession no. Glösa-Profil KS27-056; scale bar = 100 μm. U-II, detail of Plate VIII, 3 showing stoma with slightly striated guard cells; scale bar = 30 μm. U-III, overview showing epidermal cells and stoma from the inner side of the cuticle. Accession no. Glösa-Profil KS11-012; scale bar = 30 μm. U-III, detail of Plate VIII, 5 showing guard cells; scale bar = 10 μm.

M. Hübers et al. / Review of Palaeobotany and Palynology 193 (2013) 38–56

Plate VII.

51

52

M. Hübers et al. / Review of Palaeobotany and Palynology 193 (2013) 38–56

Plate VIII (caption on page 50).

M. Hübers et al. / Review of Palaeobotany and Palynology 193 (2013) 38–56

Furthermore, microsporangia, pollen sacs and other small plant remains were found. Because the latter lack distinct features, they are not described here. 4.4.1. Type U-I (Plate VIII, 1, 2) Material: 1 specimen: Glösa-2324/2004-099. Description: Epidermal cells either rectangular, 86–102 μm long and 23–34 μm wide, or triangular, 43–52 μm long and 12–28 μm wide. Anticlinal walls straight or curved. Stoma 68 μm long and 35 μm wide. Guard cells 55 μm and 22 μm wide, strongly cutinised, having a bulging appearance. Subsidiary cells and neighbouring cells absent. 4.4.2. Type U-II (Plate VIII, 3, 4) Material: 1 specimen: Glösa-Profil KS27-056. Description: Cuticle thin. Epidermal cells elongate, more or less rectangular. Cell outlines not clearly visible, except for stoma and hair bases. Anticlinal walls apparently straight. Cuticle bearing few hair bases, about 20 μm diameter. Stoma 122 μm long and 58 μm wide, partly covered with organic remains. Guard cells slightly striated with stronger cutinisation along stomatal pit. Subsidiary cells and neighbouring cells absent. 4.4.3. Type U-III (Plate VIII, 5, 6) Material: 1 specimen: Glösa-Profil KS11-012. Description : Cuticle thick. Form of epidermal cells ranging from elongate polygonal to almost triangular, 44–52 μm long and 13–18 μm wide. Anticlinal walls straight. Periclinal walls unornamented. Stoma 18 μm long and 6 μm wide. Guard cells almost semi-circular, about 17 μm long and 9 μm wide, partly covering the subsidiary cells. Two elongated polar cells are oriented in the same direction as the stomatal pore. Two lateral subsidiary cells at each side, almost triangular, 24–25 μm long and 14–19 μm wide; one of the subsidiary cells has subdivided to form an encircling cell. Remarks: Plate VIII, 5, 6 show the inner side of the cuticle revealing the whole dimensions of the guard cells, which are normally partly covered by the subsidiary cells. 5. Discussion Although it is very difficult to assign the specimens to particular genera or species, because the fragments are small and only a few Mississippian cuticles are known that could be used for comparison, an assignment to higher groups of plants appears to be possible for most of the dispersed remains. As was to be expected, gymnosperm cuticles, here represented by pteridosperm cuticles, appear to be most frequent. Lyginopteris bermudensiformis and Lepidodendron lossenii are the most common macrofloral elements in Glösa and it is likely that some of the cuticle fragments belong to these species. The macroflora from other localities of the Hainichen Basin includes a great variety of taxa, mainly pteridophylls (Hartung, 1938). This is reflected in the number of pteridosperm cuticles in the Glösa samples, but not in the diversity, which is rather low, even if it is taken into account that some of the pteridophylls are probably ferns and it is very unlikely that fern cuticles are preserved. However, it is remarkable that two types of putative calamitalean cuticles were found, because these have only rarely been reported in the literature. The Glösa cuticles were collected from alluvial sediments and therefore might represent a mixture of elements from different habitats. Some pteridosperm cuticles show xeromorphic features, but these do not necessarily reflect drier habitats, because plants growing in peat bogs and climbers often show xeromorphic features (e.g., Krings and Kerp, 1997; Krings et al., 2001). The most surprising finds in the Glösa samples are several bryophytes. With some 20,000 extant species, bryophytes are the second largest group of plants after the flowering plants, and important constituents of most terrestrial ecosystems, from the tropics to boreal

53

regions. Unlike vascular land plants, bryophytes are predominantly poikilohydric, which means that their state of hydration is controlled by the environment and not by the plant itself. Moreover, the freeswimming sperm requires water for fertilisation. Most bryophytes are bound to constantly moist habitats. Some can survive in temporarily drier habitats but they must complete their life cycle to the production of dormant, desiccation-resistant spores prior to the dry season, or be desiccation tolerant. In modern wetland ecosystems bryophytes play a preeminent role in nutrient cycling, water retention, and water availability. Peat bogs can only survive due to the enormous water storage capacity of peat mosses (Sphagnales). Bryophytes are commonly regarded as the oldest land plants, based on phylogenetic reconstructions (e.g., Kenrick and Crane, 1997; Heinrichs et al., 2007; Clarke et al., 2011; Shaw et al., 2011). Molecular clock estimates indicate minimum ages of 449 Ma for the origin of liverworts and 420.4 Ma for the divergence of mosses (Clarke et al., 2011). They predate the earliest tracheophyte macrofossils that are known from the Wenlock (Edwards et al., 1983). If the earliest record of the spore Ambitisporites (Steemans et al., 2009) is accepted as proof for the first appearance of tracheophytes, their fossil record can even be traced back to the Hirnantian. Wellman et al. (2003) described aggregates of cryptospores from the middle Ordovician of Oman, which they interpreted as sporangia of liverworts. Dispersed cryptospores, regarded as evidence for land vegetation, have been described from the lower middle Ordovician of Argentina (Rubinstein et al., 2010). The oldest unequivocal liverwort macrofossils have been described as Metzgeriothallus sharonae Hernick et al. (Hernick et al., 2008) from the upper Middle Devonian of New Brunswick, Canada. Slightly younger is Hepaticites devonicus Hueber from the lowermost Upper Devonian of New York (Hueber, 1961). The claim that the enigmatic plant fossil Prototaxites would be rolled liverwort mats (Graham et al., 2010) is unsubstantiated (Boyce and Hotton, 2010; Taylor et al., 2010). Carbon isotopic analyses show that Prototaxites is probably a saprophytic fungus that relied on deposits of algal-derived organic matter (Hobbie and Boyce, 2010). Nowadays, bryophytes play an important role in shaping and maintaining wetland ecosystems. It has recently even been suggested that the expansion of early non-vascular land plants accelerated chemical weathering and may have drawn down enough atmospheric CO2 to trigger the growth of ice sheets in the Late Ordovician (Lenton et al., 2012). These conclusions were based on weathering experiments with the extant moss Physcomitrella patens (Hedwig) Bruch et Schimper. Several bryophytes have been described from the Pennsylvanian, but they are extremely rare in pre-Permian deposits. Three liverworts showing cell structure, Treubites kidstonii (Walton) Schuster, Blasiites lobatus (Walton) Schuster and Metzgeriothallus metzgerioides (Walton) Schuster, have been described from the Middle Coal Measures, Yorkian Series, Upper Carboniferous (= Duckmantian/Westphalian B) of Shropshire, U.K. (Walton, 1925, 1928). One liverwort not showing cell structure has been described from the Westphalian D of The Netherlands, as Hepaticites tortuosus (Van Amerom, 1996). A single specimen of putative moss has been described as Muscites plumatus Thomas from the Mississippian of Drybrook, Forest of Dean, Gloucestershire, U.K. (Thomas, 1972). However, the microscopic preparation of this specimen does not show a clear cell pattern and a moss affinity is very unlikely. A few other species of mosses were described, two from the uppermost Carboniferous (Stephanian) of central France, i.e., Muscites polytrichaceus Renault et Zeiller, a not fully convincing compression fossil from Commentry (Renault and Zeiller, 1885, 1886–88) and Muscites bertrandii Lignier, an anatomically preserved form from Grand Croix, central France (Lignier, 1914). In recent years, two mosses were described from the Upper Carboniferous of South America. Muscites amplexifolius Ottone et Archangelsky from the western Paganzo Basin, San Juan Province, Argentina (Ottone and Archangelsky, 2001) is a very convincing compression,

54

M. Hübers et al. / Review of Palaeobotany and Palynology 193 (2013) 38–56

partially still showing cell structure. Another compression from the Paraná Basin, Brazil (Amaral et al., 2004) was identified as Dwykea sp. The single specimen shows less detail but is interpreted as a female gametophyte. Other remains from the Carboniferous interpreted as bryophytes are considered to be too poorly preserved to justify assigning them to the mosses (Krassilov and Schuster, 1984; Oostendorp, 1987; Ignatov, 1990). A plant that shares characters of liverworts, hornworts and pteridophytes has been described from the Lower Carboniferous of the Russian Platform (Mosseichik et al., 2007); however, a taxonomic assignment is not possible. Bryophytes appear to be more common in the Permian and they have been reported from several floral provinces. From Euramerica two occurrences have been reported, i.e., a compression of a presumed liverwort from the lower Permian of the Autun Basin, France (Nectoux et al., 1968), and four types of leafy mosses obtained by bulk maceration from the lower Permian of the Saar-Nahe Basin, Germany (Busche, 1968). A few finds have been reported from the upper Permian of Gondwana, i.e., a liverwort and two mosses preserved as compressions from the lower Permian of India (Chandra, 1994) and at least three genera of compression fossils from the Permian of South Africa (Plumstead, 1966; Lacey, 1969; Anderson and Anderson, 1985). Two genera were recently described from the Guadalupian (Permian) of the Paraná Basin, Brazil (Cortez Christiano De Souza et al., 2012). Most remarkable of these is Capimirinus, a pleurocarpous moss of which sporophytes and gametophytes are known. A structurally preserved moss has been described from a silicified peat from Antarctica (Smoot and Taylor, 1986). Bryophytes are most common in the Permian of Angaraland, Russia. Over twenty genera of mosses have been described to date from the Pechora, Tunguska and Kuznetsk basins and from the Russian Platform. Neuburg (1956, 1958a, 1958b, 1960) and Fefilova (1973, 1978) described very rich and diverse floras with exceptionally well-preserved leafy mosses from the upper Permian of the Pechora Basin (Russia). Gomankov and Meyen (1986) and Ignatov (1990) described twelve genera of excellently preserved mosses from the upper Permian of the Russian Platform. Ignatov and Shcherbakov (2009) described a moss preserved as adpressions from the lower Permian of the Russian Far East. Although only part of our material can be assigned to a particular group of bryophytes, and none can be assigned to a genus or species, the presence of nine forms shows that bryophyte taxa are present, including the oldest representatives of the Protosphagnales (Types B-IIa–c), an order that was hitherto known only from the Permian of Russia (Neuburg, 1960; Fefilova, 1973, 1978; Ignatov, 1990). Types B-IIIa and B-IIIb show remarkable resemblances to the Sphagnales or peat mosses, an early diverging lineage of mosses (Shaw et al., 2010a; Chang and Graham, 2011), which according to recent molecular clock estimates diverged from the ancestral line between 319 and 128 Ma (Shaw et al., 2010b). However, Sphagnales have so far not been found in pre-Cenozoic rocks (Miller, 1984). The two forms of our Type B-III are not only the oldest known representatives of the order Sphagnales but they also predate current molecular clock estimates. The paucity of bryophytes in the Carboniferous is striking, especially because Carboniferous floras have been extensively studied and this time interval is marked by the occurrence of extended wetland ecosystems resulting in an unequalled accumulation of plant biomass. The rareness of bryophytes in the Carboniferous has often been discussed (e.g., Lacey, 1969; Krassilov and Schuster, 1984; Hemsley, 2001) and several possible explanations have been given. A major problem is that bryophytes are difficult to recognise without having information on the cellular organisation. Bryophyte remains may have often been overlooked or misinterpreted. Several compression/ impression specimens without cellular structure, which were originally attributed to mosses, have later been assigned to other groups or considered to be unidentifiable or unassignable (Lacey, 1962). A moss from the Permian of Russia, Polyssaievia spinulifolia (Zalessky) Neuburg, was originally described as a conifer based on the gross morphology (Zalessky,

1936; Neuburg, 1956). This example shows the difficulties of recognising fossil mosses without having anatomical information. However, the recognition of bryophytes in thin sections and cellulose acetate peels is equally problematic, because bryophytes normally consist, except for the stem and leaf costa, of a single layer of cells. Such cell files can easily be confused with various kinds of deteriorated plant tissues, algal filaments or even fungal hyphae. A common explanation for the paucity of bryophytes in the fossil record is that the preservation potential of bryophytes would be very limited, due to their delicate nature and the absence of cuticles and highly resistant compounds like lignin (Erickson and Miksche, 1974; Hébant, 1977). However, lignin-like substances, including phenolic cell wall contents have been reported (Erickson and Miksche, 1974; Glime, 2007), and some bryophytes have a thin leaf cuticle that seems to be similar to that of vascular land plants (Proctor, 1979). Hemsley (2001) carried out experiments to test the decay of bryophytes compared to that of selected vascular plants. He found that bryophytes have a similar preservation potential as vascular plants and therefore he concluded that during the Carboniferous bryophytes were indeed less abundant than might be expected on the basis of palaeoenvironmental reconstructions. Bryophytes generally have an even slower decomposition rate than tracheophytes due to the presence of many kinds of secondary metabolites, which function as antiherbivore and antibiotic compounds (Glime, 2007). Decomposition rates vary depending on water availability, pH and temperature and they are lowest in temperate, and particularly in boreal regions. Modern peats can contain abundant moss remains, but nothing is known about their fate during coalification, when physico-chemical changes (particularly gelification, caused by increasing temperature) occur. Tracheids, cuticles and spores are likely to be preserved in a morphologically unaltered state, due to the presence of lignin, cutin/cutan and sporopollenin, as is shown by the occurrence of macerals that are still well distinguishable in medium-volatile bituminous coals (Taylor et al., 1998). However, the preservation of tissues and organs lacking these compounds is unlikely. Moss spores are likely to be preserved, but dispersed moss spores may be difficult to recognise as such. It cannot be excluded that wildfire played a role in the preservation of the moss remains, although one would expect that such fragments would be darker. The bryophytes from Glösa have only been found in bulk-macerated sediment samples, not as macroscopic remains. The samples have yielded many very small plant fragments and some arthropod remains. Every tiny fragment has been examined under high magnification and slides have been made of all promising pieces of organic matter, regardless of their size, resulting in many hundred slides with minute plant remains. Carefully scrutinising this material revealed the bryophyte remains here described. Normally, only the larger (macroscopically recognisable) remains are picked out when bulk macerations are sorted out; the fine debris is normally not examined. It is interesting to note that the best known and generally accepted Carboniferous bryophytes have also been described from bulk-macerated sediments (Walton, 1925, 1928). Similarly, the only records of mosses from the lower Permian of Euramerica were found in bulk macerates (Busche, 1968). Regarding these finds, we anticipate that bryophyte remains may actually be much more common than usually thought. Bulk macerates yielding material with cellular preservation are most promising as has already been advocated by Lacey (1962), even though only tiny fragments may be found. Nonetheless, these delicate fragments unequivocally demonstrate that bryophytes were part of Mississippian palaeotropical vegetations. We therefore assume that bryophytes were also more common in Pennsylvanian palaeotropical wetland ecosystems than is generally believed. 6. Concluding remarks Careful analysis of plant mesofossils obtained by bulk maceration appears to add a new dimension to our knowledge of late Palaeozoic

M. Hübers et al. / Review of Palaeobotany and Palynology 193 (2013) 38–56

terrestrial ecosystems. It reveals the presence of several types of bryophytes, a group of plants that is extremely rare in the Devonian and Carboniferous, although much older. The fragments we found are small, smaller than most dispersed cuticle fragments, and they may easily be overlooked. The moss remains from Glösa are the oldest unequivocal mosses known to date. They include the earliest representatives of the Protosphagnales, a group so far only known from the Permian of Angaraland, and probably the oldest known representatives of the Sphagnales, which predate current molecular clock estimates for this group. The occurrence of bryophytes, especially mosses, in the Carboniferous is of particular interest because they play a major role in modern wetland ecosystems in nutrient cycling, water retention, and water availability. All bryophytes are bound to at least seasonally humid habitats, varying from bare rock and sediment surfaces to peat bogs, and it is difficult to be more specific about the forms encountered in the Glösa flora. Only the Sphagnum-like Types B-IIIa,b may point to a peat-bog habitat, because the extant genus Sphagnum occurs in peat bogs and moist tundras, generally in nutrient-poor habitats with a low pH. The representatives of the Protosphagnales (Types B-IIa–c) may point in the same direction regarding the similarities in overall morphology, notably the presence of large water-storage cells, and the fact that they are often associated with coal occurrences, e.g., in the Kuznetsk Basin (Neuburg, 1960). The calamitalean cuticles probably represent riparian vegetation along streams and lakes. Although the remains from Glösa are fragmentary and cannot be named, they unequivocally demonstrate that several different types of bryophytes formed part of Carboniferous wetland ecosystems. We expect that a systematic application of bulk maceration, a method now rarely used for studying Carboniferous floras, will show that bryophytes were much more widespread than commonly thought. Acknowledgements Financial support was provided by the Deutsche Forschungsgemeinschaft (DFG grant KE 584/17-1). Stephan Schultka (Berlin) kindly made part of the Glösa material available. We thank Benjamin Bomfleur (Lawrence, KS), Fred Daniels (Münster) and Hagen Hass (Münster) for fruitful discussions. Edith L. Taylor and Michael Krings are acknowledged for the constructive reviews. References Abbot, M.L., 1958. The American species of Asterophyllites, Annularia and Sphenophyllum. Bulletin of American Paleontology 38, 285–390. Amaral, P.G.C., Bernardes de Oliveira, M., Ricardi-Branco, F., Broutin, J., 2004. Presencia de Bryopsida fértil en los niveles Westfalianos del subgrup Itararé, Cuenca de Paraná, Brasil. Tropical Bryology 25, 101–110. Anderson, J.M., Anderson, H.M., 1985. Paleoflora of Southern Africa Prodromus of South African megafloras Devonian to Lower Cretaceous. Ed. A.A. Balkema, Rotterdam (423 pp.). Barthel, M., 1997. Epidermal structures of sphenophylls. Review of Palaeobotany and Palynology 95, 115–127. Batenburg, L.H., 1977. The Sphenophyllum species in the Carboniferous flora of Holz (Westphalian D, Saar Basin, Germany). Review of Palaeobotany and Palynology 24, 69–99. Boyce, C.K., Hotton, C.L., 2010. Prototaxites was not a taphonomic artifact. American Journal of Botany 97, 1073. Busche, R., 1968. Als Laubmoosreste gedeutete Pflanzenfossilien aus den Lebacher Schichten (Autunien) von St. Wendel, Saar. Argumenta Palaeobotanica 2, 1–14. Chandra, S., 1994. Bryophytic remains from the Early Permian sediments of India. The Palaeobotanist 43 (3), 16–48. Chang, Y., Graham, S.W., 2011. Inferring the higher-order phylogeny of mosses (Bryophyta) and relatives using a large, multigene plastid data set. American Journal of Botany 98, 839–849. Clarke, J.T., Warnock, R.C.M., Donoghue, P.C.J., 2011. Establishing a time-scale for plant evolution. New Phytologist 192, 266–301. Clayton, G., Coquel, R., Doubinger, J., Gueinn, K.J., Loboziak, S., Owens, B., Streel, M., 1977. Carboniferous miospores of western Europe: illustration and zonation. Mededelingen Rijks Geologische Dienst 29, 1–71. Cleal, C.J., Zodrow, E.L., Šimůnek, Z., 2007. Pennsylvanian-aged medullosalean Odontopteris cantabrica Wagner. Acta Palaeobotanica 47 (2), 327–337.

55

Cortez Christiano De Souza, I., Ricardi Branco, F.S., Vargas, Y.L., 2012. Permian bryophytes of Western Gondwanaland from the Paraná Basin in Brazil. Palaeontology 55, 229–241. Edwards, D., Feehan, J., Smith, D., 1983. A late Wenlock flora from Co. Tipperary, Ireland. Botanical Journal of the Linnean Society 86, 19–36. Erickson, M., Miksche, G.E., 1974. On the occurrence of lignin or polyphenols in some mosses and liverworts. Phytochemistry 13, 2295–2299. Fefilova, L.A., 1973. New genera of Permian bryopsids from the north of Fore-Urals flexure. Geologiya i poleznye iskopaemye severo-vostoka Evropeiskoi chasti SSSR. Institute of Geology of Komi Branch of Academy of Sciences of USSR, Syktyvkar, pp. 85–90 (in Russian). Fefilova, L.A., 1978. Leafy Mosses of the Permian of the European Northern Part of the USSR. Nauka, Leningrad (in Russian). Gaitzsch, B., Egenhoff, S., Hesse, S., Ehling, B.C., 2010. Variscan early molasses in the SaxoThuringian. In: Linnemann, U., Romer, R.L. (Eds.), Pre-Mesozoic Geology of SaxoThuringia; from the Cadomian Active Margin to the Variscan Orogen. Schweizerbart, Stuttgart, pp. 311–322. Gehmlich, M., Linnemann, U., Tichomirowa, M., Gaitzsch, B., Bombach, K., 1998. Geochronologie invers gestapelter Deckenreste im Frankenberger Zwischengebirge. Terra Nostra 98, 52–54. Glime, J.M., 2007. Bryophyte Ecology: Physiological ecology, 1 (http://www.bryoecol. mtu.edu/ last accessed 30 December 2011). Gomankov, A.V., Meyen, S.V., 1986. Tatarina flora (composition and distribution in the Late Permian of Eurasia). Trudy Geologicheskogo Instituta. Akademiya Nauk SSSR 401, 1–174 (in Russian). Good, C.W., 1971. The ontogeny of Carboniferous articulates: calamite leaves and twigs. Palaeontographica B 133, 137–158. Graham, L.E., Cook, M.E., Hanson, D.T., Pigg, K.B., Graham, J.M., 2010. Structural, physiological, and stable carbon isotopic evidence that the enigmatic Paleozoic fossil Prototaxites formed from rolled liverwort mats. American Journal of Botany 97, 268–275. Hartung, W., 1938. Flora und Altersstellung des Karbons von Hainichen-Ebersdorf und Borna bei Chemnitz (Erzgeb. Becken). Abhandlungen des Sächsischen Geologischen Landesamts 8, 1–140. Hauke, R.L., 1957. The stomatal apparatus of Equisetum. Bulletin of the Torrey Botanical Club 84, 178–181. Hébant, C., 1977. The Conducting Tissues of Bryophytes. J. Cramer, Lehre (157 pp.). Heinrichs, J., Hentschel, J., Wilson, R., Feldberg, K., Schneider, H., 2007. Evolution of leafy liverworts (Jungermanniidae, Marchantiophyta): estimating divergence times from chloroplast DNA sequences using penalized likelihood with integrated fossil evidence. Taxon 56, 41–44. Hemsley, A.R., 2001. Comparison of in vitro decomposition of bryophytic and tracheophytic plant material. Botanical Journal of the Linnean Society 137, 375–384. Hernick, L.V., Landing, E., Bartowski, K.E., 2008. Earth's oldest liverworts—Metzgeriothallus sharonae sp. nov. from the Middle Devonian (Givetian) of eastern New York, USA. Review of Palaeobotany and Palynology 148, 154–162. Hobbie, E.A., Boyce, C.K., 2010. Carbon sources for the Palaeozoic giant fungus Prototaxites inferred from modern analogues. Proceedings of the Royal Society B 277, 2149–2156. Hübers, M., Kerp, H., 2012. Oldest known mosses discovered in Mississippian (late Visean) strata of Germany. Geology 40 (8), 755–758. Hübers, M., Bomfleur, B., Kerp, H., 2011a. Dispersed lycopsid cuticles from the Mississippian of Chemnitz (Saxony, Germany). Review of Palaeobotany and Palynology 167, 10–15. Hübers, M., Bomfleur, B., Krings, M., Kerp, H., 2011b. An Early Carboniferous leafcolonizing fungus. Neues Jahrbuch für Geologie und Paläontologie, Abhandlungen 261, 77–82. Hueber, F.M., 1961. Hepaticites devonicus, a new fossil liverwort from the Devonian of New York. Annals of the Missouri Botanical Garden 48, 125–132. Ignatov, M.S., 1990. Upper Permian mosses from the Russian Platform. Palaeontographica 217B, 147–189. Ignatov, M.S., Shcherbakov, D.E., 2009. A new fossil moss from the Lower Permian of the Russian Far East. Arctoa 18, 201–212. Jäger, H., Wierich, F., 2006. Palynostratigraphie. In: Amler, A., Stoppel, D. (Eds.), Stratigraphie von Deutschland VI. Unterkarbon (Mississippium): Schriftenreihe der Deutschen Gesellschaft für Geowissenschaften, 41, pp. 294–318. Kenrick, P., Crane, P.R., 1997. The Origin and Early Diversification of Land Plants—A Cladistic Study. Smithsonian Institution Press, New York/London. Kerp, H., 1990. The study of gymnosperms by means of cuticular analysis. Palaios 5, 548–569. Kerp, H., Bomfleur, B., 2011. Photography of plant fossils—new techniques, old tricks. Review of Palaeobotany and Palynology 166, 117–152. Kerp, H., Krings, M., 1999. Light microscopy of fossil cuticles. In: Jones, T.P., Rowe, N.P. (Eds.), Fossil Plants and Spores: Modern Techniques. The Geological Society, London, pp. 52–56. Kerp, H., Kampe, A., Schultka, S., Van Amerom, H.W.J., 2006. Makrofloren. In: Deutsche Stratigraphische Kommission (Ed.), Stratigraphie von Deutschland VI. Unterkarbon (Mississippium): Schriftenreihe der Deutschen Gesellschaft für Geowissenschaften, 41, pp. 271–293. Krassilov, V.A., Schuster, R.M., 1984. Paleozoic and Mesozoic fossils. In: Schuster, R.M. (Ed.), New Manual of Bryology: Hattori Botanical Laboratory, 2 (Nichinan, Japan). Krings, M., Kerp, H., 1997. Cuticles of Lescuropteris genuina from the Stephanian (Upper Carboniferous) of Central France: evidence for a climbing growth habit. Botanical Journal of the Linnean Society 123, 73–89. Krings, M., Taylor, T.N., Taylor, E.L., Axsmith, B.J., Kerp, H., 2001. Cuticles of Mariopteris occidentalis White nov. emend. from the Middle Pennsylvanian of Oklahoma (USA), and a new type of climber hook for mariopteroid pteridosperms. Review of Palaeobotany and Palynology 114, 209–222.

56

M. Hübers et al. / Review of Palaeobotany and Palynology 193 (2013) 38–56

Lacey, W.S., 1962. Welsh Lower Carboniferous plants I. The flora of the Lower Brown Limestone in the Vale of Clwyd, North Wales. Palaeontographica 111B, 126–160. Lacey, W.S., 1969. Fossil bryophytes. Biological Reviews 44, 189–205. Lange, B., 1982. Key to the northern boreal and arctic species of Sphagnum, based on the characteristics of stem leaves. Lindbergia 8, 1–29. Lenton, T.M., Crough, M., Johnson, M., Pires, N., Dolan, L., 2012. First plants cooled the Ordovician. Nature Geoscience 5, 86–89. Lignier, O., 1914. Sur une mousse houillière à structure conservée. Bulletin de la Societé Linnéenne de Normandie 6, 128–131. Máyas, G., 1920. Funde neuer Pflanzenreste aus dem Kulm von Chemnitz-Borna. Bericht der Naturwissenschaftlichen Gesellschaft zu Chemnitz 20, 55–73. Miller, N.G., 1984. Tertiary and Quaternary fossils. In: Schuster, R.M. (Ed.), New Manual of Bryology: Hattori Botanical Laboratory, Nichinan, 2, pp. 1194–1232. Mosseichik, Y.V., Ignatov, M.S., Ignatiev, I.A., 2007. A bryophyte-like plant from the Lower Carboniferous of the Moscow Coal Basin. Arctoa 16, 99–122. Nectoux, P., de La Comble, J., Jovet Ast, P., Boureau, E., 1968. Sur une hepatique permienne de Charmoy (S. & L.). Bulletin de la Societé d'Histoire Naturelle d'Autun. Nouvelle Série 48, 8–10. Neuburg, M.F., 1956. The discovery of mosses in the Permian deposits of the USSR. Doklady Akademii Nauk SSSR 107, 321–324. Neuburg, M.F., 1958a. Permian true mosses of Angaraland. Journal of the Paleontological Society of India 3, 22–29. Neuburg, M.F., 1958b. Palaeozoic mosses of Angarida. Congresso Geologico Internacional, XXe Sesión, México, 1956, Sección VII—Paleontología, Taxonomía y Evolución, pp. 97–106. Neuburg, M.F., 1960. Leafy mosses from the Permian deposits of Angaraland. Trudy Geologičeskij Institut Akademija Nauk SSSR 43, 3–64. Nindel, F., 1955. Die tierischen Reste aus dem Karbon von Karl-Marx-Stadt und Hainichen i.S. Geologie 4, 673–694. Oostendorp, C., 1987. The bryophytes of the Palaeozoic and the Mesozoic. Bryophytorum Bibliotheca 34, 1–112. Ottone, E.G., Archangelsky, S., 2001. A new bryophyte from the Upper Carboniferous of Argentina. Ameghiniana 38, 219–233. Pant, D.D., Kidwai, P.F., 1968. Development of stomata in Equisetum. Annals of Botany 32, 601–608. Plumstead, E.P., 1966. Recent palaeobotanical advances and problems in Africa. Symposium on Floristics and Stratigraphy of Gondwanaland. Birbal Sahni Institute, Lucknow, pp. 1–12. Proctor, M.C.F., 1979. Surface wax on the leaves of some mosses. Journal of Bryology 10, 531–538. Renault, B., Zeiller, R., 1885. Sur les mousses de l'époque houillère. Comptes Rendus Hebdomadaires des Séances de l'Académie des Sciences 100 C, 660–662. Renault, B., Zeiller, R., 1886–88. Études sur le terrain houiller de Commentry. Livre deuxième. Flore fossile. Théolier and Cie, St. Étienne, pp. 3–366. Rößler, R., Schneider, J.W., 1997. Eine bemerkenswerte Paläobiocoenose im Unterkarbon Mitteleuropas – Fossilführung und Paläoenvironment der Hainichen-Subgruppe (Erzgebirge-Becken). Veröffentlichungen des Museums für Naturkunde Chemnitz 20, 5–44. Rubinstein, C.V., Gerrienne, P., de la Puente, G.S., Astini, R.A., Steemans, P., 2010. Early Middle Ordovician evidence for land plants in Argentina (eastern Gondwana). New Phytologist 188, 365–369. Schneider, J.W., Hoth, K., Gaitzsch, B.G., Berger, H.J., Steinborn, H., Walter, H., Zeidler, M.K., 2005. Carboniferous stratigraphy and development of the Erzgebirge Basin, East Germany. Zeitschrift der deutschen Gesellschaft für Geowissenschaften 156, 431–466.

Scott, A.C., Galtier, J., Clayton, G., 1984. Distribution of anatomically-preserved floras in the Lower Carboniferous in Western Europe. Transactions of the Royal Society of Edinburgh 75, 311–340. Shaw, A.J., Cox, C.J., Buck, W.R., Devos, N., Buchanan, A.M., Cave, L., Seppelt, R., Shaw, B., Larraín, J., Andrus, R., Greilhuber, J., Temsch, E.M., 2010a. Newly resolved relationships in an early land plant lineage: Bryophyta class Sphagnopsida (peat mosses). American Journal of Botany 97, 1511–1531. Shaw, A.J., Devos, N., Cox, C.J., Boles, S.B., Shaw, B., Buchanan, A.M., Cave, L., Seppelt, R., 2010b. Peatmoss (Sphagnum) diversification associated with Miocene Northern Hemisphere climatic cooling? Molecular Phylogenetics and Evolution 55, 1139–1145. Shaw, J.A., Szövényi, P., Shaw, B., 2011. Bryophyte diversity and evolution: windows into the early evolution of land plants. American Journal of Botany 98, 352–369. Šimůnek, Z., 2008. The Asturian and Cantabrian floral assemblages with Cordaites from the Plzeň Basin Czech Republic. In: Kędzior, A., Cleal, C.J. (Eds.), Pennsylvanian (Carboniferous) Biotas and Palaeoenvironments: Studia Geologica Polonica, 129, pp. 51–80. Šimůnek, Z., 2010. Cuticles of Paripteris Gothan (Pteridospermopsida) from the Westphalian of Poland. Bulletin of Geosciences 85 (2), 353–360. Smoot, E.L., Taylor, T.N., 1986. Structurally preserved fossil plants from Antarctica. II. A Permian moss from the Transantarctic Mountains. American Journal of Botany 73, 1683–1691. Steemans, P., Le Herisse, A., Melvin, J., Miller, M.A., Paris, F., Verniers, J., Wellman, C.H., 2009. Origin and radiation of the earliest vascular land plants. Science 324, 353. Taylor, W.A., Strother, P.A., 2008. Ultrastructure of some Cambrian palynomorphs from the Bright Angel Shale, Arizona, USA. Review of Palaeobotany and Palynology 151, 41–50. Taylor, W.A., Strother, P.A., 2009. Ultrastructure, morphology, and topology of Cambrian palynomorphs from the Lone Rock Formation, Wisconsin, USA. Review of Palaeobotany and Palynology 153, 296–309. Taylor, E.L., Taylor, T.N., 2012. Paleozoic mosses: small, but no longer inconspicuous. Geology 40 (8), 767–768. Taylor, G.H., Teichmüller, M., Davis, A., Diessel, C.F.K., Littke, R., Robert, P., 1998. Organic Petrology. Borntraeger, Berlin/Stuttgart (704 pp.). Taylor, T.N., Taylor, E.L., Decombeix, A.-L., Schwendemann, A., Serbet, R., Escapa, I., Krings, M., 2010. The enigmatic Devonian fossil Prototaxites is not a rolled-up liverwort mat: comment on the paper by Graham et al. (AJB 97: 268–275). American Journal of Botany 97, 1074–1078. Thomas, B.A., 1970. Epidermal studies in the interpretation of Lepidodendron species. Palaeontology 13 (1), 145–173. Thomas, B.A., 1972. A probable moss from the Lower Carboniferous of the Forest of Dean, Gloucestershire. Annals of Botany 36 (1), 155–161. Van Amerom, H.W.J., 1996. A new bryophyte species (Hepaticites tortuosus nov. sp.) from the Westphalian D in borehole De Lutte-6, Twente, the Netherlands. Mededelingen Rijks Geologische Dienst 55, 99–105. Walton, J., 1925. Carboniferous Bryophyta. I. Hepaticae. Annals of Botany 39, 563–572. Walton, J., 1928. Carboniferous Bryophyta. II. Hepaticae and Musci. Annals of Botany 42, 707–716. Wellman, C.H., Osterloff, P.L., Mohiuddin, U., 2003. Fragments of the earliest land plants. Nature 425, 282–285. Yao, Z.-Q., Liu, L.-J., Mapes, G., Rothwell, G.W., 2000. Leaf morphology and cuticular features of Sphenophyllum in the Gigantopteris flora from South China. Review of Palaeobotany and Palynology 110, 67–92. Zalessky, M.D., 1936. Sur quelques plantes nouvelles du système anthracolitique du bassin de Kousnetzk. Problems of Paleontology, 1. Publications of the Laboratory of Paleontology, Moscow-University, USSR, pp. 233–236.