Abstract: The Devonian fenestrate bryozoan, SchischcatellaWaschurova, 1964, possessed colonies in the form of low, erect bifoliate fronds that grew from an encrusting sheet-like base with autozooecia arranged in biserial, bifurcating rows. This growth habit is unique in fenestrates, which normally had unilaminate arborescent colonies. Originally, Schischcatella was described from the Lower Devonian of Tajikistan. This article describes a new species, S.heinorum sp. nov., from the Middle Devonian of the Eifel (western Rhenish Massif, Germany) with additional material from the Lower Devonian of the Kellerwald (eastern Rhenish Massif, Germany). External and internal morphologies of this bryozoan have been studied using abundant material. The growth habit of Schischcatella suggests a completely different pattern of feeding currents than that in the normal fenestrate colony. The outflow of the filtered water occurred only on edges of colonies between rami. In the absence of chimneys (areas of vertical water expelling), such a functional morphology may have restricted extension of the colony in a distal direction. The evolution of Schischcatella is apparently an example of paedomorphosis, the genus evolved from an unknown semicosciniid species by the early ontogenetic interruption of colony development and further changes in the mode of growth.
F enestrate bryozoans were abundant during the Middle – Late Palaeozoic (Morozova 1987) after first appearing in the Lower Ordovician (Taylor and Curry 1985) and became extinct in the Upper Permian. They became increasingly important during the Devonian, replacing trepostomes and cystoporates in bryozoan communities (Cuffey and McKinney 1979). Fenestrate bryozoans develop erect colonies, which are attached in different ways to a hard or firm substrate. Characteristically, budding of autozooecia produces unilaminate arborescent colonies in which the autozooecial apertures open only on one side of the branch (Morozova 1987; McKinney and Jackson 1989). Such a construction elicited the evolutionary success of fenestrates in the Palaeozoic. Compared with other erect growth forms, unilaminate arborescent colonies have the advantages of a relatively light skeleton and a continuous feeding surface (McKinney and Jackson 1989). Fenestrate colonies act as a pumping system, in which self-induced feeding currents flow towards the obverse (apertural) surface, whereas filtered water outflows between branches to the reverse side (Cowen and Rider 1972; Cook 1977).
However, one fenestrate genus represents an exception to the rule. Waschurova (1964) described the genus Schischcatella from the Lower Devonian of Tajikistan. She placed this genus tentatively in the family Fenestellidae King, 1849, stressing its similarity to Semicoscinium. Subsequently, Morozova (1987, p. 81) included Schischcatella in her new family Semicosciniidae. The peculiarity of Schischcatella is its growth form, unique in fenestrates: encrusting, comprising erect, low, bifoliate fronds. This construction lacks the open spaces between adjacent branches characteristic of fenestrate colonies.
Until now, Schischcatella was only known from the type locality in Tajikistan (Waschurova 1964). However, in the course of an investigation of Devonian bryozoan faunas of Europe (DFG project ER 278/4.1 and 2), abundant material of Schischcatella was discovered in the Middle Devonian of the Eifel (western Rhenish Massif, Rhineland-Palatinate, Germany). Additional material (two colonies) comes from a Lower Devonian (lower Emisan) locality in the Kellerwald (eastern Rhenish Massif, Hesse, Germany). The German material represents a new species, Schischcatella heinorum sp. nov. This article aims to describe the new species and to discuss its morphology and palaeoecology.
Stratigraphy: ‘Type Eifelian’ vs. Regional Lower Givetian Stratigraphy of the Gerolstein Syncline
The Middle Devonian sequence of the Eifel is subdivided into several formations, subformations and members (Text-fig. 2), which are separated by lithological as well as faunal criteria. The reference profile is the so called ‘Type Eifelian Profile’ within the Hillesheim Syncline (Struve 1982; Struve and Werner 1982). Because of the complexity of facies within the Limestone Synclinorium, several members have a restricted regional extent and detailed stratigraphical nomenclature that differs from the type Eifelian.
The Gerolstein Syncline, from which the bryozoans described here were discovered, is dominated by Lower Givetian deposits, which clearly differ from the Type Eifelian. Therefore, several local members of the Loogh and Cürten formations were established by Winter (1965) (Text-fig. 2). He also differentiated a Gerolstein south-western from a north-eastern regional facies, with differences clearly visible in the Cürten Formation.
The bryozoans come from the Lower Givetian (hemiansatus Conodont Biozone) of the Gerolstein Syncline. The Lower Givetian is represented by poorly exposed strata of the uppermost Ahbach Formation and by the lime-marl successions of the overlying Loogh and Cürten formations, which together with younger dolomitised formations (Text-fig. 2) dominate the syncline.
The base of the Loogh Formation (Text-fig. 2) comprises the Dachsberg Member (Text-fig. 2B) with homogeneous limestone containing sparsely preserved macrofossils (Winter 1965). The limestone was deposited under quiet conditions (Winter 1965, p. 307). According to Winter (1965, p. 289), the Dachsberg Member is restricted to the south-west of the Gerolstein Syncline. Progressive shallowing led to the inception of biostromal growth and, therefore, to facies differentiation, characterising the Baarley Member of Winter (1965, pp. 289–290). Massive trochite-dominated limestone and ‘matrix limestone’ characterise this member. Local facies differences within the member often modify this basic lithologic characterisation. As an example, the oldest discussed bryozoan type locality, the ‘Mühlenwäldchen’ (locality 1 of Text-fig. 1), is characterised both by fine-grained homogenous limestone and marl banks. More regular limestone-marl interbeds increase towards the boundary with the Hustley Member (Winter 1965, pp. 290–292). The second bryozoan type locality, the ‘Berlinger Bach’ (locality 2 of Text-fig. 1), is positioned near this boundary, within the uppermost part of the Baarley Member. Irregularly distributed marls between localised stromatoporoid-coral biostromes yielded a macrofossil fauna comparable to that of type locality 1. The superposed strata represent a temporary decrease in sedimentation, which was marked by the new appearance of stromatoporoid-coral biostromes that distinguish the Hustley Member. These biostromes, with partly limy, partly marly deposits, mark a maximum level of facies differentiation.
The locus typicus of the Hustley Member is close to the north-eastern slope of the railway cut near Gerolstein Station and, thus, next to the third and youngest discussed bryozoan type locality (locality 3 of Text-fig. 1). The Hustley Member is generally dominated by thin encrinitic limestone, limestone banks, massive ‘matrix limestone’ and greenish to brighter ochre or brownish marl. All rock types can be found side by side because of interfingering of facies. In the investigated area, local marl packages of several metres thickness are interrupted by isolated stromatoporoid-coral biostromes. The biostromes locally interfinger with limestone banks or limestone-marl interbeds. A diverse macroinvertebrate fauna was recovered, especially within the marly sediments, including the bryozoan species described in this article.
From the base to the top of the Loogh Formation, continuous facies complications and increase in biostromal developments are observed, and a progressive shallowing of the sea can be inferred. This corresponds with the increase in the numbers of species and individuals noted by Winter (1965, p. 309).
In contrast to the Loogh Formation, the overlying Cürten Formation in the south-western part of the Gerolstein Syncline can be compared to the Type Eifelian. However, because of facies peculiarities within the north-eastern part of the Gerolstein Syncline, Winter (1965, pp. 292–304) recognised a restructuring of the Cürten Formation in this region (Text-fig. 2). He attributed the north-east/south-west differentiation of the Cürten Formation mainly to a lack of the typical limestone-marl interbeds (Felschbach and Forstbach members) in the south-western part of the Gerolstein Syncline. He discussed two possible reasons: (1) a sedimentary hiatus in the north-east; (2) the Felschbach and Forstbach members of the north-eastern part could be developed in a Hustley- and/or Meerbüsch-like facies.
Geography, Stratigraphy and Significant Macrofossils of Bryozoan Type Localities within the Gerolstein Syncline
Type locality 1: ‘Mühlenwäldchen’ (Rhineland-Palatinate, Germany)
Type locality 1 is the so-called ‘Mühlenwäldchen’ in south-west Gerolstein (Gerolstein Syncline, Eifel, Rhenish Massif, north-western Rhineland-Palatinate, Germany; UTM 50°13′16.14′′N, 6°39′01.00′′E) (Text-fig. 1). Stratigraphically, the bryozoan outcrop is presumably positioned within the lower Baarley Member (regional valid denomination of the uppermost Wotan Member (Hotz et al. 1955) within the Gerolstein Syncline (sensu Winter 1965)) of the middle Loogh Formation, Lower Givetian (Middle Devonian; hemiansatus Conodont Biozone). This historic fossil locality is most famous for the well preserved crinoids comprising melocrinids (M. pyramidalisGoldfuss, 1839 and M. gibbosus gibbosusGoldfuss, 1831), strongly sculptured eucalyptocrinids (Eucalyptocrinites rosaceusGoldfuss, 1831), rhodocrinitids (Rhipidocrinus crenatus crenatus (Goldfuss, 1831), R. c. reticularis and R. aculeatus (both Schultze, 1866)), megaradialocrinids (M. crispus (Quenstedt, 1861)) and the huge aboral cups of the hexacrinid Hexacrinites pateraeformis (Schultze, 1866) – as reported by Bohatý (2006a, p. 472). Also different echinoids (Lepidocentrus muelleri, Xenocidaris clavigera and X. cylindrica– all Schultze, 1866) are characteristic of type locality 1, which is also famous for the brachiopods Mimatrypa insquamosa (Schnur, 1853) and especially M. flabellata (Roemer, 1844), Schnurella schnuri (De Verneuil, 1840) and the terebratulitid genus StringocephalusDefrance, 1825. Further macrofossil groups are rugose and tabulate corals, platyceratid gastropods, phacopid trilobites and cyclostome, trepostome as well as a diverse spectrum of fenestrate bryozoans.
Material from type locality 1 includes eight colonies of Schischcatella heinorum sp. nov. on brachial valves and three colonies on pedicle valves of the atrypacean spiriferid species Mimatrypa insquamosa. Two colonies of S.heinorum sp. nov. encrust disarticulated crinoid columnals and a tabulate coral. Furthermore, c. 100 colonies were found free, unattached.
Type locality 2: ‘Berlinger Bach’ (Rhineland-Palatinate, Germany) (= locus typicus)
Type locality 2, the northern slope of the ‘Berlinger Bach’ (Text-fig. 1), is located west of Berlingen, north-east of Pelm (Gerolstein Syncline, Eifel, Rhenish Massif, north-western Rhineland-Palatinate, Germany; UTM 50°14′20.20′′N, 6°42′19.70′′E). It is stratigraphically positioned near the boundary of the Baarley and Hustley members, within the uppermost part of the Baarley Member (Loogh Formation, Lower Givetian, Middle Devonian; hemiansatus Conodont Biozone). Irregularly distributed marls between the local occurring stromatoporoid-coral biostromes yielded crinoid remains of megaradialocrinids (M. brevis, M. hieroglyphicus (both Goldfuss, 1839) and M. winteriBohatý, in press) and the huge aboral cups of the hexacrinid Hexacrinites pateraeformis, sculptured eucalyptocrinids (Eucalyptocrinites rosaceus), rhodocrinitids (Rhipidocrinus crenatus crenatus) and isolated columnals of the melocrinid species Melocrinites pyramidalis. Also different echinoids (Lepidocentrus muelleri and Xenocidaris clavigera) and blastoids (Hyperoblastus acutangulus (Schultze, 1866) and H.gilbertsoni (Etheridge and Carpenter, 1886)) are characteristic for type locality 2, which is also famous for the brachiopods Mimatrypa insquamosa, Schizophoria schnuri schnuriStruve, 1965, and especially Schnurella schnuri. Further common macrofossil groups are tabulate corals (especially thamnoporoids), platyceratid gastropods and cyclostome, trepostome as well as fenestrate bryozoans.
One colony (the holotype) of Schischcatella heinorum sp. nov. is an epibiont on the brachial valve of Mimatrypa insquamosa.
Type locality 3: North-east slope of the railway cut near Gerolstein Station (Rhineland-Palatinate, Germany)
Type locality 3 is on the north-eastern slope of the railway cut near Gerolstein Station (Gerolstein Syncline, Eifel, Rhenish Massif, north-western Rhineland-Palatinate, Germany; UTM 50°13′28.20′′N, 6°40′13.93′′E) (Text-fig. 1). The outcrop yielded well-preserved macrofossils, which are mainly embedded between localised stromatoporoid-coral biostromes, in the marly deposits of the Hustley Member of the upper part of the Loogh Formation (Lower Givetian, Middle Devonian; hemiansatus Conodont Biozone). Type locality 3 displays a wide spectrum of epibionts encrusting every available hard surface: microconchids, bryozoans, tabulate and rugose corals, chaetitids, crinoid holdfasts and brachiopods. Encrusting bryozoans occur preferentially on corals and brachiopods.
The locality is most famous for well preserved crinoid aboral cups. Megaradialocrinus elongatus (Goldfuss, 1839), Eucalyptocrinites rosaceus, Rhipidocrinus crenatus crenatus, and cupressocrinitids are most characteristic – as reported by Bohatý (2006b, p. 263, in press). Other typical macrofossils are brachiopods like Mimatrypa insquamosa, Schnurella schnuri, Primipilaria primipilaris (von Buch, 1834) and Schizophoria schnuri schnuri as well as the terebratulitid genera Stringocephalus and Stringomimus sensu Struve, 1965. Also characteristic are huge rugose corals (e.g. Mesophyllum (Mesophyllum) maximum (Schlueter, 1882)) and isolated favositid colonies (Tabulata), platyceratid gastropods and phacopid trilobites like Nyterops nyter (Struve, 1970). Besides scutellid remains (Trilobita) and blackish plant fossils, cyclostome, trepostome (?Eostenopora sp.; see Bohatý 2009, fig. 11.6) and fenestrate bryozoans are associated with Schischcatella heinorum sp. nov.
Material from the type locality 3 includes three Schischcatella colonies on brachial valves and two colonies on pedicle valves of Mimatrypa insquamosa.
Type locality 4: northern part of the ‘Rauheck Quarry’, North-east of Berndorf (Rhineland-Palatinate, Germany)
Type locality 4 is positioned at the northern part of the ‘Rauheck Quarry’, north-east of Berndorf, west of Kerpen (Hillesheim Syncline, Eifel, Rhenish Massif, north-western Rhineland-Palatinate, Germany; UTM 50°18′36.67′′ N, 6°42′40.83′′ E; not figured here). Biostratigraphically, the crinoid outcrop is positioned within the Rech Member (upper part of the Loogh Formation, Lower Givetian, Middle Devonian; hemiansatus Conodont Biozone). The locality is famous for the well preserved crinoid remains of melocrinids (M. pyramidalis), eucalyptocrinids (Eucalyptocrinites rosaceus), rhodocrinitids (Rhipidocrinus crenatus crenatus, R. c. reticularis and R. aculeatus), megaradialocrinids (M. cf. brevis and the rare M. globohirsutusBohatý, in press) and the huge aboral cups of Hexacrinites pateraeformis. Also different brachiopods, like Mimatrypa insquamosa, M. flabellata and Schnurella schnuri are famous for type locality 4. Further characteristic macrofossils are well preserved rugose and tabulate corals and platyceratid gastropods.
Material from the type locality 4 includes two colonies of Schischcatella heinorum sp. nov. on brachial valves of Mimatrypa insquamosa and one colony on brachial valves of M. flabellata.
Type locality 5: Densberg, Kellerwald (Hesse, Germany)
Additional material of Schischcatella heinorum sp. nov. comprises two colonies embedded in rock, from which five thin sections were made. This material comes from the Kellerwald near Densberg (Hesse, Germany; UTM 50°59′34 N, 9°6′36 E; not figured here). The rock was sampled from the ‘Erbsloch-Grauwake’ (Lower Devonian, Lower Emsian) by Jahnke (1971). It represents greywacke containing skeletal debris of various fossils, such as brachiopods, corals, crinoids and trepostome bryozoans. The colony of Schischcatella heinorum sp. nov. is attached to a branched trepostome bryozoan.
Within the central European Variscan fold belt, the Rhenish Massif and the Ardennes are separated by a north-south trending axial depression, the ‘Eifel Limestone Synclinorium’. Deposits of the Middle Devonian and, in part, of the Upper Devonian are preserved within the synclines, and the anticlines between them elevated the Lower Devonian strata. The Eifel Limestone Synclinorium is bordered in the north-west and north by the older Palaeozoic ‘Stavelot-Venn Massif’, and in the north-east by the ‘Mechernich Triassic Bight’. The eastern boundary is characterised by the western limb of the ‘Siegerland-Eifel Anticlinorium’. The southern boundary is the older Lower Devonian of the ‘Manderscheid Anticlinorium’, in which the ‘Trier Triassic Bight’ adjoins to the south.
The Devonian marine realm of the Eifel was bordered in the north by the ‘Old Red Continent’, which was the source area for the clastic sedimentary input. The sedimentary input accumulated from the Early to the Late Devonian with a retreating coastline towards the north. Because of massive sedimentary input during the Early Devonian, essentially only clastic sediments were deposited. At the beginning of the Middle Devonian, carbonate sedimentation occurred in the area of the later Eifel Limestone Synclinorium as well as to the north of the Venn Massif in the Ardennes. The Moselle area, the deepest and most distal part of the sedimentary realm, is characterised by fine-grained siliciclastic sediments. In this palaeogeographical setting, a lithostratigraphical/facies trichotomy of the Devonian sequence arises in the region north of the ‘Venn Anticline’, the extent of the Eifel Limestone Synclinorium and the ‘Moselle Trough’ (Meyer 1986).
Struve (1961, 1963) proposed the first palaeogeographical reconstruction of the Eifel Middle Devonian. He considered the depositional region as an isolated north-south trending sea basin, surrounded by landmasses, which he denoted as ‘Eifel Sea Street’.
Reef growth occurred to the west of the eastern mainland called ‘Istaevonia’ (= ‘Siegen Block’) and on the ‘Middle Eifel Barrier’ (‘Krömmelbein Structure’ of Struve 1961, p. 98). The so-called ‘Manderscheid Barrier’ was positioned to the south and connected the land of Istaevonia with the mainland of ‘Arduennia’ in the west and separated the comparative flat Eifel Sea from the deeper Moselle Trough to the south. Struve also presumed that a huge island, on the Venn Massif, divided the Eifel Sea Street in the north-west.
Struve (1961, 1963) provided an important foundation for all later palaeogeographical work. Today, particularly, the isolated palaeogeographical position of the depositional basin as well as the accentuation of distinctively developed boundaries in the form of barriers and islands is interpreted differently. Research within the Venn area has shown that the Middle Devonian of the Aachen area (north-west of the Eifel Limestone Synclinorium) is dominated by coarse clastic material up to the Middle Givetian. This sequence was interpreted as coastal sedimentation along the southern edge of the ‘Brabant Massif’ (Kasig and Neumann-Mahlkau 1969, p. 381). The finer clastic sediments of the Eifelian in the northern part of the Eifel Limestone Synclinorium document a gradual ablation of the coast towards the Limestone Synclinorium. Therefore, Struve′s island cores are no longer necessary. Research within the ‘Manderscheid Barrier’ (Krebs 1970) to the south of the Salmerwald Syncline demonstrates a transition between a carbonate and a fine-grained siliciclastic facies without an intervening barrier. An explanation for this may be the topography of the sea bottom, perhaps a distal ramp. According to Struve (1961, 1963), the west coast of Istaevonia was dominated by a carbonate platform or respectively by a reef barrier. Admittedly, within the eastern part of the Eifel Limestone Synclinorium, there is no evidence of any siliciclastic input. Also because of palaeotectonic reasons, an emergent area that would correspond with the Siegerland Block is implausible.
The current palaeogeographical and facies models for the Middle Devonian of the Eifel, especially the Eifelian, were initially developed by Winter (see Meyer, Stoltidis and Winter 1977, p. 327), who defined three characteristic facies realms (facies types A–C). Facies type A, distinguished by clastic sediments, is developed within the northern Eifel Limestone Synclinorium. Carbonates are proportionally rare. In the northern part of the synclinorium, the sediments were not deposited under normal marine conditions. Normal marine conditions occurred towards the south. At about the axis of the Dollendorf Syncline, the changeover to facies type C occurred. Type C is characterised by limestone and marl. Clastic components are sparse. Towards the south, the clay content increases and type C facies passes into the clay rich facies of the Moselle Trough (= ‘Wissenbach Slate’). The third facies type (type B) is developed within the eastern part of the Eifel Limestone Synclinorium. It is characterised by pure, commonly biostromal limestone; marly as well as silty sediments is secondary. This facies type characterises a shallow water area, which lay close to a shallow water barrier at the eastern Eifel. Type B facies dominates the eastern parts of the Salmerwald, Gerolstein, Hillesheim, Ahrdorf and Sötenich synclines as well as parts of the Rohr and the middle and eastern part of the Blankenheim synclines. This basic division of facies types applies at least to the Junkerberg Formation (Eifelian), but within some time slices, it was modified, for example, at the Niederehe Subformation, biostromal beds were established nearly at the complete northern Eifel sea area during a transgression. Alternatively, facies type A expanded towards the south during times of low sea level. In this case, sedimentation within the upper part of the Nohn and Junkerberg formations was dominated by clastic input. Beginning with the Freilingen Formation (Upper Eifelian), facies differences disappear. Because of a transgression, facies type C was established all over the depositional area. In the Givetian, stromatoporoid-coral biostromes extended all over the Eifel Sea.
By accentuating the validity of the three facies types, Faber (1980, p. 112) modified Winter's model. Faber differentiated two palaeogeographical situations within the Lower Eifelian: (1) a relatively undifferentiated open shelf, which is characterised by southwest–northeast trending facies belts; and (2) a carbonate platform, which was developed twice within the eastern part of the Limestone Synclinorium, while the western synclinorium was still dominated by ‘normal’ shelf sedimentation. Thus, a second structural control developed, trending north–south. In the Lower Givetian, the whole Eifel region was bounded by a tectonic high within the southern part of the synclinorium (Krebs 1974) (taken from Bohatý 2005a, pp. 384–388).
Material and Methods
Colonies of Schischcatella heinorum sp. nov. were investigated both through transmitted light microscopy using thin sections and by SEM. Thin sections were made from the hand specimens and from colonies embedded in epoxy resin SpeciFix-40. The terminology of the morphological characters is adopted and modified from Snyder (1991a, b) and Hageman (1991a, b). The following morphological characters were measured (Text-fig. 3): WR (width of ramus), DRC (distances between rami centres), TR (thickness of ramus), AW (aperture width), ADR (distance between aperture centres along ramus), AAR (distance between aperture centres across ramus), ABR (distance between aperture centres between rami), MAW (maximum autozooecial chamber width), CL (autozooecial chamber length). Statistics are summarised in the tables accompanying the species description, giving number of measurements (N), arithmetic mean (X), sample standard deviation (SD), coefficient of variation (CV) and minimum (MIN) and maximum (MAX) values.
Studied material is housed at the Senckenberg Museum (Frankfurt am Main, Germany). Twenty-one free colonies were studied using SEM. Colonies encrusting brachiopods (including the holotype) are numbered SMF 20.199–SMF 20.200. SEM processing was performed at the Institute of Geosciences, University of Kiel (CAMSCAN-Serie-2-CS-44). SEM samples (34 colonies) are numbered SMF 20.172–SMF 20.193, SMF 20.201–SMF 20.212. Eight thin sections were prepared from five colonies (SMF 20.194–SMF 20.198). Additional material from a single colony is housed at the Geological Centre Göttingen (GZG), numbers GZG.IN.0.010.519c–d, f–g, GZG.IN.0.010.520c (five thin sections).
Type species. S. concretaWaschurova, 1964. Lower Devonian; Shishkat, Kshtut River basin, Zeravshan Mountains, Tajikistan.
Diagnosis: Colonies forming erect low bifoliate fronds with a continuous, non-perforated median skeletal sheet. Fronds having wide encrusting bases. Autozooecia on each side of median skeletal sheet arranged in two rows on rami (from Latin ramus = branch). Rami originating from basal encrusting wall and extending essentially perpendicularly from the base to the upper edge of the bifoliate frond. Rami may bifurcate with bifurcation point preceded for short distance by increase in number of rows of autozooecia to four; separated laterally by a space approximately equal to the width of a ramus. Obverse surface of each ramus extended medially as a high keel, club-shaped in cross section with a narrow base and broader towards the outer edge, which commonly is rounded. Keels wide in basal parts, disappearing in distal parts of rami, slightly undulating longitudinally. Superstructure absent. Hemisepta and diaphragms absent. Heterozooecia absent.
Endozonal autozooecial chambers somewhat variable in shape with pentagonal to irregularly polygonal cross sections in deep endozonal tangential sections, grading to elongate trapezoidal or pentagonal in shallow endozonal tangential sections; axial wall zigzag in deep endozonal tangential sections grading to linear and with a clearly defined granular core in shallowest endozonal tangential sections. In longitudinal sections, autozooecial chambers moderately high and short, with short vestibules.
Autozooecial walls with more or less developed internal granular skeleton. Extrazooecial skeleton dense, obscurely laminated, microstyles numerous, incipient, represented as small deflections contributing to a generally granulose appearance of laminations. (Diagnosis after Frank K. McKinney, pers. comm. 2008; modified by authors).
Comparison. Schischcatella is the only known genus of the Order Fenestrata that forms colonies comprising bifoliate fronds that lack open space between adjacent branches (rami in Schischcatella) through which water could flow. In its internal morphology (autozooecial chamber shape) and presence of wide, club-shaped keels, it is similar to SemicosciniumProut, 1859.
Occurrence and geological age. Two species are known: Schischcatella concretaWaschurova, 1964 from the Emsian (Lower Devonian) of the Zeravshan Mountains, Tajikistan, and S.heinorum sp. nov. from the Lower Emsian (Lower Devonian) of Hesse, Germany, and the Lower Givetian (Middle Devonian) of the Eifel (both Rhenish Massif, Germany).
Additional material. Two colonies GZG.IN.0.010.519c–d, f–g and GZG.IN.0.010.520c (five thin sections).
Type locality. Type locality 2, northern slope of the ‘Berlinger Bach’, west of Berlingen, north-east of Pelm, Gerolstein Syncline, Eifel, Rhenish Massif, north-western Rhineland-Palatinate, Germany.
Type strata. Transition between Baarley and Hustley members, within the uppermost part of the Baarley Member, middle Loogh Formation, Lower Givetian (Middle Devonian; hemiansatus Conodont Biozone).
Diagnosis. Colonies forming erect bifoliate fronds; autozooecia arranged in biserial fascicles on each side of median skeletal lamina; autozooecia short, relatively high, triangular to roughly rhombic in mid-tangential section; hemisepta and diaphragms absent; median keel wide in basal part, disappearing in distal part; heterozooecia absent; superstructure absent.
Description. Colonies forming erect bifoliate fronds, comprising bifoliate sheets, 0.9–3.0 mm high and 2.3–15.0 mm long. Fronds can bi- and trifurcate, or fuse together (Pl. 1, figs 1–3, Pl. 2, figs 7–8, Pl. 4, figs 2, 4). Autozooecia on each side of the median skeletal lamina arranged in two rows on rami, originating from basal encrusting wall, short, relatively high, with short vestibules; triangular to pentagonal in shallow tangential sections, trapezoid to roughly rhombic in mid-tangential sections; with axial wall zigzag in deep endozonal tangential sections grading to linear and with a clearly defined granular core in shallowest endozonal tangential sections; hemisepta and diaphragms absent. Median lamina 0.020–0.025 mm thick, fabric microgranular. Maximum eight generations of autozooecia observed on each ramus; most colonies with 4–5 generations. Autozooecial apertures circular to oval, surrounded by 12–15 apertural nodes (Pl. 4, fig. 1). Apertural nodes 0.005–0.010 mm in diameter. Rami semicircular to trapezoid in transverse section, diverging at angles of 68–90 degree, separated laterally by a space approximately equal to the width of a fascicle. Bifurcation of rami not observed. Median keel wide in basal parts of fascicles, disappearing in distal parts, club-shaped, slightly undulating longitudinally. Superstructure absent. Heterozooecia unknown.
Internal granular skeleton well developed, continuous with median lamina, keels, peristomes and microstyles. Microstyles 0.002–0.005 mm in diameter, densely spaced. Extrazooecial skeleton dense, obscurely laminated, protruded by numerous microstyles.
Comparison. Waschurova (1964) reported a comprehensive set of measurements for Schischcatella concreta, although many measurements do not coincide with those used in the present publication. The new species differs from the type species in having closely spaced autozooecial apertures. Waschurova gave spacing of apertures as 5 per 2 mm of the branch length. The same measurement for the new species ranges between 8 and 9. Furthermore, branch widths for S. concreta were given as ‘0.25 mm’, without any range or arithmetic mean. This value is smaller than those in S. heinorum: 0.28–0.50 mm in Eifel material and 0.42–0.54 mm in the sample from Hesse (0.37 mm and 0.42 mm, respectively, at average).
Remarks. The Lower Devonian material from Hesse shows some differences to the Middle Devonian specimens from Rhineland-Palatinate. Colonies from Hesse have better developed extrazooidal skeletons and slightly wider rami (0.42 mm vs. 0.37 mm in Rhineland material on average). Furthermore, keels in Hesse specimens are locally sharpened (Pl. 4, figs 2, 5–6), whereas keels in Rhineland Palatine samples are mainly flattened or rounded (Pl. 5, fig. 5). However, rounded keels are also present in specimens from Hesse (Pl. 4, fig. 3), thus we suggest that the shape of the keel can vary within this species, and the differences between the samples are insufficient to split them into two different species.
Occurrence and geological age. Kellerwald near Densberg (Hesse, Germany); Erbsloch-Grauwake, Lower Emsian (Lower Devonian) (type locality 4). Gerolstein Syncline, Eifel, Rhenish Massif (north-western Rhineland-Palatinate, Germany); type locality 1 (‘Mühlenwäldchen’), South-west Gerolstein; lower Baarley Member of the middle Loogh Formation, Lower Givetian (Middle Devonian; hemiansatus Conodont Biozone). Type locality 3, north-eastern slope of the railway cut near Gerolstein Station; Hustley Member of upper Loogh Formation, Lower Givetian (Middle Devonian; hemiansatus Conodont Biozone).
Bryozoans are active suspension-feeders generating feeding currents by coordinated beating of cilia on their tentacles. Growth habits and extrazooidal structures such as keels or nodes play an important role in a hydrodynamic situation of feeding currents. In individual zooids, lophophores generate inhalant currents directed towards their mouths, whereas exhalant currents flow out into gaps between the zooids (Winston 1978, 1979). To facilitate expulsion of exhalant currents structures such as maculae are developed in many erect and encrusting bryozoan colonies (Banta et al. 1974; Taylor 1975, 1979, 1999; McKinney 1990; Shunatova and Ostrovsky 2001, Shunatova 2002). Maculae often consist of heterozooecia or skeletal material. They can be elevated above the colony surface (monticules), or have depressed central areas. Inhalant currents are produced in intermacular areas, whereas filtered water outflow from the colony surface through the maculae (so-called ‘chimneys’). Multizooidal currents in bryozoan colonies play apparently different roles: feeding, cleaning and clearing sediment from the colony surface, or helping expulsion of sperm and larvae (Taylor 1979; Taylor, pers. comm. 2008; McKinney, pers. comm. 2008). Strong currents can remove sediment and larvae of potential epizoans from the colony surface.
The feeding current patterns in colonies of Schischcatella would have been completely different to those of other fenestrate bryozoans. Schischcatella have erect bifoliate fronds arising from a substrate, in which autozooecia are arranged in two rows on each ramus, the equivalent of branches. Such colonies lack open spaces (fenestrules) necessary for outflow of filtered water.
The hydrodynamic function of rami can be compared with that of fascicles inferred in some Jurassic cyclostomes such as Theonoa and Spiropora (Taylor 1979). The incurrent water flowed towards the polypides and excurrent water was expelled over polypide-free spaces between the fascicles (Text-fig. 4). The rami in Schischcatella are arranged closely, so that the spaces between them are roughly equal to their widths. Lophophores of autozooidal polypides apparently extended into gaps between rami, and inhalant currents were directed towards them in the same pattern as in a typical fenestrate colony (Text-fig. 4A). However, there is no space for filtered water to depart to the opposite side as in the model of Cowen and Rider (1972). Instead, colonies of Schischcatella have morphological characters, which may have had hydrodynamic function for directing the exhalant currents. On the one hand, the distal edges of the bifoliate fronds are sharp and smooth. The exhalant currents could outflow here through spaces between the rami, functioning as channels for expelling filtered water (Text-fig. 4A–B). A similar pattern has been described by Nielsen (1981) in the living cheilostome ‘Hippodiplosia’insculpta. This bryozoan produces bifoliate sheets by back-to-back growth of autozooids, and the whole surface of the sheet represents a continuous inhalant area. On the other hand, median keels at the bases of the rami of Schischcatella are often not symmetrical, but curved proximally (Pl. 5, figs 3–4; Text-fig. 4B). Indeed, they resemble the blades of a screw. We infer that exhalant water currents flowed out along these ‘blades’.
In some cases, colonies of Schischcatella heinorum sp. nov. show development of keels, which partly or completely fuse together forming a cylindrical space (Pl. 1, fig. 4, Pl. 5, fig. 5). Inhalant water currents must have entered at the outer part of this ‘cylinder’ and been expelled between the rami on the distal edge of the frond (Text-fig. 4C–D).
It is obvious that colonies of Schischcatella are constrained in their growth, as distal extension of the encrusting sheet increases the distance the exhalant currents must flow. They can flow out only on the base and on the distal part near the sharp edge. The augmentation of the autozooids (power-producing units) does not compensate the increase in distance, and the efficiency of the excurrent system drops because of the friction of the water. Furthermore, the autozooecia in proximal parts of rami are often closed by terminal diaphragms and seem to be inactive. The only way to increase the feeding surface is to grow laterally, which we observe in colonies of Schischcatella. The same restriction of the feeding surface width because of marginal expulsion of filtered water was postulated for living ‘Hippodiplosia’insculpta by Nielsen (1981). This bryozoan, like Schischcatella, does not develop ‘chimneys’, which improve the expulsion of filtered water from the colony surface and thus enable expansion of the feeding surface (Banta et al. 1974; McKinney and Jackson 1989).
Colonies of Schischcatella were found attached to hard substrates or were free. Free, unattached colonies were apparently removed from their substrates after death and did not encrust soft sediments. Attached colonies show the following distribution on substrates: one on a trepostome bryozoan (5%), one on a rugose coral (5%), two on crinoids (10%) and 16 on brachiopod valves, especially Mimatrypa insquamosa (Schnur, 1853) (80%). This bryozoan apparently preferred brachiopod brachial valves over pedicle valves: 14 colonies (73.68%) are attached to brachial valves, whereas only five (26.32%) were found on pedicle valves. On brachiopods, Schischcatella colonies tend to be attached near the edges of the valves, apparently profiting from water currents created by a brachiopod.
The internal morphology of Schischcatella shows that it belongs within the order Fenestrata, which normally possess unilaminate arborescent colonies. Most probably, a representative of the family Semicoscinidae was an ancestor of Schischcatella. The modification of a ‘standard’ fenestrate colony to a bifoliate frond of Schischcatella should have had some advances, because these bryozoans were found in relatively large quantities, although restricted to a few localities. Accompanying fenestrate fauna includes several genera, many of which have well-developed protective superstructures (Hemitrypa, Loculipora, Unitrypa, Isotrypa).
Devonian fenestrate bryozoans started to produce defensive superstructures, apparently as a response to an increasing predator pressure (Morozova 1987; McKinney et al. 2003). Signor and Brett (1984) suggest that the impact of predators controlled the evolution of shelly fauna during the mid-Palaeozoic (Middle Devonian – Early Carboniferous), because of the appearance of fishes. Armoured benthic taxa had an advantage over those without protective structures. Colonies of large fenestrate species started to develop protective structures, whereas miniaturisation could have been another effective method to limit predator stress. Schischcatella could have been adapted to inhabit hidden microenvironments avoiding exposure to predators.
Another plausible explanation could be a substrate condition, which hindered attachment of bryozoan colonies. Fenestrate bryozoans have encrusting bases formed after larval settlement (McKinney 1983, 1987; McKinney and Burdick 2001). Isotrypa pannosa (Počta, 1894), a semicoscinid species from the Konĕprusy Limestone (Lower Devonian) of the Czech Republic, was found encrusting crinoid columnals (McKinney 1987, p. 163, fig. 2J). Similar encrustations of Fenestrapora sp. (on ephemeral objects) were observed on abundant material from the Rhenish Massif (unpublished data). In many cases, Devonian environments (notably the Lower Eifelian to lowermost part of the Lower Givetian of the Rhenish Massif; see Struve et al. (2008, pp. 297–374)) were locally characterised by soft substrates or moderately stabilised firmgrounds. Crinoids, which are abundant here (e.g. within the Olifant and Zerberus members of the Ahbach Formation (lowermost Lower Givetian), see Bohatý (2005b, p. 205, fig.3A, 2006c)), developed large roots and stabilised the soft substrate as pioneers. In the absence of rooting structures, it was an advantage for the bryozoans to produce encrusting colonies attached on not only other organisms like brachiopods, corals and crinoids (Bohatý 2009) but also algae. Indeed, there is a shift in a composition of bryozoan communities according to the substrate conditions. In the Middle Devonian of Rhenish Massif, there are numerous tubular bryozoan species from localities with soft sediments like marls (Ernst 2008). These bryozoans, which belong to different cystoporate and trepostome genera, encrusted ephemeral substrates, apparently algae, which disappeared after decaying. The peculiar shape of the colony in Schischcatella, also its small sizes, could be an adaptation for occupying of microbiotopes like brachiopod shells, because anchoring in soft substrates was tied with significant energy and material costs for the animal.
Although the ancestral part of Schischcatella colonies is not clearly defined (maybe, except the specimen on Pl. 2, figs 7–8), the colony development shows the stages characteristic for the early growth in ‘normal’ fenestrates (Cumings 1904, 1905; McKinney 1978; McKinney and King 1984). The growth of a fenestrate colony started with an ancestrula, followed by a series of basal zooids developed around the ancestrula. Basal zooids were developed in pairs (consequently from each side laterally) forming a ring around ancestrula. From this ancestrular complex branches were developed. In Schischcatella, the ontogenetic development of the colony stopped apparently after forming of an ancestrula and at least two first basal zooids (progenesis after Gould 1977 or hypomorphosis after Shea 1983). Further basal zooids grown contacting the substrate and the basal wall of an opposite zooecium. Instead of a concentric arrangement of the first series of basal zooids in a normal fenestrate colony, in which autozooecia open outside of the cone (Fenestella and Unitrypa type inCumings 1904), Schischcatella had a bilateral base, from which a vertical bifoliate frond was developed. This basal part represented an expanding ancestrular complex (zone of astogenetic changes), from which following generations of zooecia arisen, arranged in two rows on rami (zone of astogenetic repetition). Interestingly, this bryozoan retained an arrangement of autozooecia in two rows on rami, equivalents of branches, characteristic of an ancestor semicosciniid bryozoan. In general, the colony of Schischcatella can be regarded as a derivate of a cone-shaped growth form, which is extremely flattened so that the reverse sides of branches contact each other building common median lamina. This modification was possible because of the changes in the early colony growth, which resulted in offset of later parts of development into erect branches.
Acknowledgements. We thank Hans-Peter Hein (Wermelskirchen) and Uwe Hein (Solingen) for donating the type material. We are grateful to Ute Schuldt (Kiel) for her assistance in SEM, and Wolfgang Reimers (Kiel) for his assistance in the preparation of thin sections. Frank K. McKinney (Boone) and Paul D. Taylor (London) are greatly thanked for their helpful comments and advices to the manuscript. Patrick Wyse Jackson (Dublin) and an anonymous reviewer are thanked for helpful reviews and comments. Most of the studied bryozoan material was collected during the crinoid research project of the Deutsche Forschungsgemeinschaft (DFG project HE 1610/16-1); JB gratefully acknowledges this financial support. AE thanks the Deutsche Forschungsgemeinschaft for financial support (DFG project ER 278/4.1 and 2). This paper is a contribution to IGCP 499.