Pentamerus Sowerby, Reference Sowerby1813 is one of the most widely reported early Silurian pentameride brachiopods worldwide, often regarded as a good paleobiogeographic indicator for relatively cosmopolitan marine benthic shelly faunas in the early Silurian, in contrast to the strong endemism of Late Ordovician shelly benthos. True Pentamerus, however, occurs predominantly in middle–upper Llandovery strata, especially in Laurentia, Baltica, Avalonia, Siberia, and Kazakhstan. In South China, Pentamerus occurs only rarely in Aeronian rocks, whereas the common forms previously reported as Pentamerus have been assigned to a different genus, Sulcipentamerus Zeng, Reference Zeng1987 (see Rong et al., Reference Rong, Jin and Zhan2007). Thus, the occurrence of ‘Pentamerus’ gothlandicus Lebedev, Reference Lebedev1892 in the upper Wenlock Slite beds of Gotland (Bassett and Cocks, Reference Bassett and Cocks1974; Bassett, Reference Bassett1977) has been one of the well-known exceptions outside the common stratigraphic range of the genus, and on Gotland this species forms a widely distributed marker bed known as the Pentamerus gothlandicus layer (e.g., Hede, Reference Hede1927; Calner et al., Reference Calner, Jeppsson and Munnecke2004; Bassett, Reference Bassett2005)
Lebedev (Reference Lebedev1892) initially proposed the species Pentamerus gothlandicus based only on ventral valves from Gotland (see Bassett, Reference Bassett1977). Among the large number of genera of the family Pentameridae, the ventral valve usually has a rather conservative morphology, with a consistently developed median septum and spondylium. As a result, the differentiation of pentamerid genera often relies on the differences in internal structures of the dorsal valve, such as the various configurations of hinge plates and crura, skeletal structures that supported lophophores for suspension filter feeding and respiration. The lack of information from the original description of P. gothlandicus, in addition to its unusual stratigraphic position, cast some doubt on the affinity of this species to Pentamerus. In terms of shell shape and outline, relative convexity of the ventral and dorsal valves, and the configuration of ventral and dorsal umbo and beak, P. gothlandicus is nearly identical to the type species Pentamerus oblongus Sowerby, Reference Sowerby1839, although the Gotland species generally has considerably larger shells than other known species of the genus on average (Fig. 1.1–1.5). In subsequent studies, Bassett and Cocks (Reference Bassett and Cocks1974) and Bassett (Reference Bassett1977) provided two important clarifications on this species: first, P. gothlandicus is confined to the upper Wenlock Slite beds or its coeval strata on Gotland; second, the semi-transparent shell reveals two discrete inner hinge plates (i.e., outer brachial plates in old usage) at their junctions with the dorsal valve floor. The latter point, in particular, was used by Bassett (Reference Bassett1977) as a convincing basis for confirming the identity of P. gothlandicus as Pentamerus.
After several recent revisions of Pentamerus (for a summary, see Boucot and Johnson, Reference Boucot and Johnson1979; Sapelnikov, Reference Sapelnikov1985; Jin and Copper, Reference Jin and Copper2000), most species now seem to occur in strata from mid-Aeronian to Telychian; only a small number of species are found in the Wenlock–Ludlow, notably in the Ural Mountains and in the Racine Dolomite of North America. In many cases, the preservation of purported Wenlock–Ludlow species of Pentamerus is rather poor, with shells commonly showing distortion, disarticulation, or breakage. This is true also for P. gothlandicus, which occurs predominantly as disarticulated or broken valves in skeletal packstone or grainstone facies, although conjoined or even complete shells can be found. Such preservation bias has undoubtedly contributed to the difficulty of taxonomic identifications of these pentamerids. In this study, all the relatively well-preserved shells available from museum collections, supplemented by newly collected specimens, are assembled in an attempt to elucidate the morphology and taxonomic affinity of the Gotland species.
Materials and methods
In a recent survey of available collections at the Gotland Museum (Visby), the Swedish Museum of Natural History (Stockholm), and the Natural History Museum (London), a fairly large number of specimens of Pentamerus gothlandicus were examined in order to clarify the phylogeny and evolutionary patterns of Pentamerus because this Silurian hallmark taxon can provide important information on the faunal evolution and paleobiogeography of the period. Examination of museum collections, as well as new collections from Tjeldersholm, one of the P. gothlandicus localities on Gotland, through naturally exposed dorsal interiors and transversely cut sections, immediately reveals that the dorsal internal structures of P. gothlandicus are distinctly different from those of the typical Pentamerus. Rather, the ventral sides of the outer hinge plates show various degrees of fusion to form a ventrally projecting plate, resembling a medially crested arch, usually around the mid-length of the hinge plates (Fig. 1.7–1.11). The inner hinge plates and the dorsal portions of the outer hinge plates, however, are clearly discrete, as is typical of Pentamerus. Such a tent-like structure, called a “closed brachiophorium” by Sapelnikov (Reference Sapelnikov1972), has been reported in several pentamerid genera of Wenlock–Ludlow age, such as Pentamerifera Khodalivich, Reference Khodalevich1939; Capelliniella Strand, Reference Strand1928; and Brooksina Kirk, Reference Kirk1922, which are most common in the Ural Mountains, Alaska, and western North America (e.g., Kirk, Reference Kirk1922, Reference Kirk1925; Sapelnikov, Reference Sapelnikov1972, Reference Sapelnikov1985; Johnson et al., Reference Johnson, Boucot and Murphy1976; Boucot and Johnson, Reference Boucot and Johnson1979; Rong and Zhang, Reference Rong and Zhang1988; Sheehan and Harris, Reference Sheehan and Harris1997). Thus, it appears that this brachial structure was exclusively a post-Llandovery evolutionary trend among some genera of the Pentameroidea, since such a structure is unknown in Llandovery or older members of the superfamily. Sections of other well-preserved dorsal interiors of the Gotland species have also confirmed the presence of a brachiophorium with a fused crest, thus warranting the establishment of a new genus, to be called Karlsorus n. gen. in this study. The main purpose of this study, therefore, is to clarify the morphology of Karlsorus gothlandicus (n. comb.) and to discuss the implications of its revised taxonomy for pentamerid evolution and paleobiostratigraphy in the Silurian.
A recent summary and update of the geology of Gotland and the litho- and biostratigraphy of the Slite beds can be found in Calner et al. (Reference Calner, Jeppsson and Munnecke2004). Details of the geology of the Pentamerus gothlandicus-bearing localities (Tjeldersholm 1, Stora and Lilla Karlsö, Valle 1, and Valle 2) can be found in Hede (Reference Hede1927), Laufeld (Reference Laufeld1974), Calner et al. (Reference Calner, Jeppsson and Munnecke2004), and Bassett (Reference Bassett2005). The Karlsorus-bearing upper Slite beds lie in the upper Wenlock Cyrtograptus lundgreni Biozone, where a minor extinction event has been recognized, but it affected mainly planktonic fossil groups (e.g., graptolites; see Jaeger, Reference Jaeger1991; Lenz et al., Reference Lenz, Noble, Masiak, Poulson and Kozłowska2006) rather than the shelly benthos (e.g., brachiopods; Copper, Reference Copper2004).
Repositories and institutional abbreviations
Specimens examined and cited in this study, including figured specimens, are from the following institutions: Gotland Museum (LMG), Visby, Gotland, Sweden; Swedish Museum of Natural History (Riksmuseum, Br), Stockholm, Sweden; Natural History Museum (B), London, UK; Paleontological Institute (VSEGEI), St. Petersburg, Russia.
Evolutionary and biostratigraphic significance
Historically, Pentamerus has been a taxonomic dump bag for a variety of smooth-shelled pentameride genera. This has greatly diminished the biostratigraphic utility of this globally widespread taxon and obscured its evolutionary trend during the early Silurian, although true Pentamerus has been thought largely as a middle Aeronian–late Llandovery genus.
The first appearance of Pentamerus during the middle Aeronian has been widely recorded in Laurentia, Baltica, Avalonia, Siberia, and some of their adjacent small terranes (see summaries of Boucot and Johnson, Reference Boucot and Johnson1979; Sapelnikov, Reference Sapelnikov1985; Jin and Copper, Reference Jin and Copper2000). The early forms of Pentamerus developed a nearly equibiconvex shell, were most abundant from high tropical to subtropical regions, and adapted to a vertical life position on a muddy, level-bottom substrate in tightly packed clusters (Jin, Reference Jin2008). By the late Aeronian, Pentamerus diversified into the paleoequatorial regions to give rise to Sulcipentamerus, a strongly ventribiconvex shell with an arched ventral valve and a flattened and reduced dorsal valve. This has been interpreted as an adaption for a recumbent life position in the generally low-energy paleoequatorial environment devoid of frequent severe storms, with the large and deep ventral valve partly embedded in the sediments, and the reduced dorsal valve behaving like a mobile lid (Rong et al., Reference Rong, Jin and Zhan2007; Gushulak et al., Reference Gushulak, Jin and Rudkin2016).
By the middle Telychian, Pentamerus went through a major speciation event and gave rise to Pentameroides Schuchert and Cooper, Reference Schuchert and Cooper1931, which became a common pentameroid in Laurentia and Baltica during the late Telychian (Johnson, Reference Johnson1979; Baarli and Johnson, Reference Baarli and Johnson1988; Jin and Copper, Reference Jin and Copper2000). This cladogenetic event was marked by the formation of a cruralium, through the convergence of inner hinge plates onto a low median septum in the dorsal valve. The Pentamerus-Pentameroides transition has been regarded as a reliable biostratigraphic marker for a middle Telychian age for Laurentia and Baltica (Baarli and Johnson, Reference Baarli and Johnson1988).
The post-Llandovery evolution of Pentamerus has been confusing because the taxonomic validity of many smooth-shelled pentamerids has been a matter of debate. In North America, some large-shelled forms of late Telychian–Wenlock age, such as Apopentamerus and Pentamerus (Supertrilobus) proposed by Boucot and Johnson (Reference Boucot and Johnson1979), have not been studied in detail regarding their internal structures. Apopentamerus, for example, was regarded by Sapelnikov (Reference Sapelnikov1985) as a junior synonym of Harpidium. The relationship between the smooth-shelled Harpidium Kirk, Reference Kirk1925 and Pentamerus also remains to be investigated. The strongly ventribiconvex shells of Harpidium, ranging from late Llandovery to Ludlow, may be similar to the evolution of Sulcipentamerus that adapted to low-energy level bottom or sheltered reefal depositional environments (Gushulak et al., Reference Gushulak, Jin and Rudkin2016).
Pentamerus and Karlsorus n. gen., as well as many other genera of the family Pentameridae showed various diversifications in shell morphology during the Wenlock and Ludlow:
(1) Drastic increase in shell size, sometimes with highly arched ventral umbo to give the ventral valve a somewhat conical appearance (e.g., Harpidium; see Boucot et al., Reference Boucot, Rong and Blodgett2002). Large shells of Rhipidium Schuchert and Cooper, Reference Schuchert and Cooper1931 from Gotland may exceed 135 mm in length (e.g., specimen B5578 in the NHM collection), and large forms of Karlsorus n. gen. are nearly 130 mm long.
(2) Inversed biconvexity, with a flattened ventral valve and reduced ventral beak, in contrast to a strongly convex dorsal valve with its umbo and beak arched over the ventral posterior, as is typical of Brooksina and Capelliniella, which are common in North America and the Ural Mountains.
(3) Increased length of the spondylium and hinge plates, sometimes extending very close to the anterior margin (e.g., Brooksina; Harpidium; Kirkidium Amsden, Boucot, and Johnson, Reference Amsden, Boucot and Johnson1967; and Pentamerifera). These extravagant internal structures are rarely seen in the pentamerids of Llandovery age.
(4) Development of the brachiophorium, as discussed earlier in this study, can be viewed as a dorso-ventrally inversed cruralium, which seems to have first appeared in late Wenlock time, corresponding to the Cyrtograptus lundgreni graptolite Biozone.
The great range of morphological diversifications of large-shelled Pentameridae is a reflection of the success of this family in their ecological expansion in the level-bottom regime, as well as their invasion into the reefal environments during the Silurian. Pentamerid shells with a reversed biconvexity, a flattened ventral valve, and an orthocline ventral beak/palintrope had the delthyrium well exposed, and thus may have retained a functional pedicle to attach to a variety of hard surfaces (e.g., walls of large cavities or other sheltered areas) in reefal environments.
Order Pentamerida Schuchert and Cooper, Reference Schuchert and Cooper1931
Suborder Pentameridina Schuchert and Cooper, Reference Schuchert and Cooper1931
Superfamily Pentameroidea M’Coy, Reference M’Coy1844
Family Pentameridae M’Coy, Reference M’Coy1844
Genus Karlsorus new genus
Pentamerus gothlandicus Lebedev, Reference Lebedev1892, upper Slite beds, upper Wenlock, Cyrtograptus lundgreni Biozone, Gotland, Sweden (Bassett, Reference Bassett1977; Calner et al., Reference Calner, Jeppsson and Munnecke2004).
Type species only. The species has been reported from Wenlock strata of Gotland (Slite beds; Bassett and Cocks, Reference Bassett and Cocks1974), England (Much Wenlock Limestone; Bassett, Reference Bassett1977), and Podolia (Kitaigorod horizon; Nikiforova, Reference Nikiforova1978).
Shell very large, elongate, nearly equibiconvex with moderate convexity; ventral valve strongly trilobate, longer than dorsal valve, with prominent umbo and beak arched over dorsal umbo; trilobation of dorsal valve inconspicuous. Spondylium and ventral median septum same as for Pentamerus. Inner hinge plates low, subparallel to each other; outer hinge plates inclined ventromedially, discrete at posterior and anterior parts, commonly fused into single plate in middle part to form brachiophorium. Crura attached to medial sides of junctions between inner and outer hinge plates.
Named after the island Stora and Lilla Karlsö, Gotland, where the type species is most abundant and best preserved.
In external shell morphology, the new genus is most similar to Pentamerus in its subpentagonal shell outline and subequibiconvexity (Fig. 1.1–1.5), with the ventral valve slightly deeper than the dorsal valve, especially in the posterior half, and the strongly convex ventral umbo and a beak arched over the dorsal umbo. Internally, Karlsorus n. gen. resembles Pentameridfera Khodalevich, Reference Khodalevich1939 (see Sapelnikov, Reference Sapelnikov1972) in having a tent-like brachiophorium formed by the inner and outer hinge plates (Fig. 1.7, 1.10, 1.11). Pentamerifera, from Ludlow strata of the eastern slope of the Ural Mountains, differs in having an extremely long spondylium and hinge plates that extend for more than three-fourths of the total shell length, whereas in Karlsorus n. gen. those internal structures do not exceed one half of the shell length (Figs. 1.1, 1.2, 2.8). A few other predominantly Ludlow pentameroids, such as Brooksina Kirk, Reference Kirk1922 from Alaska, Nevada, and the Ural Mountains, may also have developed a brachiophorium-like structure inconsistently (Boucot and Johnson, Reference Boucot and Johnson1979, p. 107). Several dorsal valves of Brooksina from the C Fauna (Ludlow) of the Roberts Mountains Formation, central Nevada, illustrated by Johnson et al. (Reference Johnson, Boucot and Murphy1976, pl. 19, figs. 3, 8, 15), possess a brachiophium similar to that of Karlsorus n. gen., in that only the middle part of the outer hinge plates are fused together. However, Brooksina has a distinctly ribbed shell and tends to have a reversed lateral shell profile compared to the new genus and Pentamerifera and has the dorsal valve larger and more convex than the ventral valve, and has the spondylium and hinge plates extending near the shell anterior margin (see Sapelnikov, Reference Sapelnikov1972, Reference Sapelnikov1985).
Karlsorus gothlandicus (Lebedev, Reference Lebedev1892) (new combination)
1892 Pentamerus gothlandicus Reference LebedevLebedev, p. 22, pl. 2, figs. 11a, 11b, 12.
non 1938 Pentamerus (Pentameroides) cf. gotlandicus; [sic] Reference St. JosephSt. Joseph, pl. 5, figs. 7, 8, pl. 6, figs. 13, 15, text-figs. 1, 8. St. Joseph’s material is from the Oslo region, assignable to Pentameroides subrectus (Hall and Clarke, Reference Hall and Clarke1893)
1974 Pentamerus gothlandicus; Reference Bassett and CocksBassett and Cocks, p. 23. No illustrations; localities and stratigraphic occurrences discussed.
1977 Pentamerus gothlandicus; Reference BassettBassett, p. 170, pl. 47, figs. 1–3, text-fig. 19. Lectotype selected.
1978 Pentamerus gothlandicus; Reference NikiforovaNikiforova, p. 170, pl. 1, figs. 1–5.
VSEGEI 102/324, selected by Bassett (Reference Bassett1977; see also Nikiforova, Reference Nikiforova1978), exact locality and stratum unknown. On Gotland, the species is largely confined to the upper Slite beds, corresponding to the Cyrtograptus lundgreni Biozone of the upper Wenlock (Bassett and Cocks, Reference Bassett and Cocks1974; Calner et al., Reference Calner, Jeppsson and Munnecke2004, fig. 2).
On Gotland, K. gothlandicus is most abundant and best preserved in the Slite beds on Lilla Karlsö (see Hede, Reference Hede1927), a tiny island about 5 km off the southwest coast near Djupvik. Other specimens examined are from coeval strata at Tjeldersholm 1 (57°37'56.4''N, 180°46'34.9''E; Laufeld, Reference Laufeld1974), Blåhäll, Fröjel (57°18'53.6''N, 18°09'40.5''E; Laufeld, Reference Laufeld1974), and Valle 1 and 2 (see Laufeld, Reference Laufeld1974 for detailed locality information). There are many other localities in Gotland where the species has been reported, but such material was not examined in this study. More detailed information on K. gothlandicus localities can be found in Laufeld (Reference Laufeld1974) and Bassett and Cocks (Reference Bassett and Cocks1974).
Shell very large, with maximum observed length of 128 mm and width 102 mm (Table 1); subpentagonal in outline, with maximum width located at about two-thirds length from apex; nearly equibiconvex, or with ventral valve slightly deeper than dorsal valve in umbonal area. Anterior commissure rectimarginate to gently sulcate (Fig. 1.1–1.5). Hinge line varying from short to relatively long, about one-third to two-thirds of shell width. Ventral umbo strongly convex, narrow, with apical angles ~90°, relatively low for shell size, raised ~15 mm above hinge line in shells over 100 mm long; palintrope defined by fairly clear ridge; beak strongly incurved, arched over but not touching dorsal umbo; pseudodeltidium present, strongly concave, almost appressed to floor of spondylium in apical area (Fig. 1.9). Ventral valve trilobate, with medial lobe broadening rapidly from valve apex to anterior margin, approximately twice as wide as each lateral lobe, delimited by gentle to prominent furrow on each side to separate from lateral lobes (Figs. 1.2, 2.4, 2.7), forming slightly protruding tongue at anterior margin. Dorsal umbo moderately and evenly convex, not marked by sulcus or fold, with small beak buried in delthyrial area of ventral valve; antero-medial portion of valve gently and faintly depressed in some shells, or marked with faint development of trilobation. Shell essentially smooth, except for randomly developed, concentric growth lines, especially close to anterior margin.
Interior of ventral valve characterized by very small and week teeth; spondylium narrow, deep, extending for less than one-half of shell length; supported by relatively long and high median septum for about one-half of spondylium length (Fig. 2.6, 2.8); thick prismatic layers of median septum forming wedge-shaped junction with prismatic layer of valve floor.
Interior of dorsal valve with correspondingly minute hinge sockets; inner hinge plates extending for about two-fifths of valve length from valve apex, discrete along entire length (Figs. 1.1, 2.2, 2.6), very low, attaining about 1/4 height of outer hinge plates (Figs. 1.7, 1.8, 2.6, 2.8, 3.1); outer hinge plates about same length as inner hinge plates, discrete in posterior and anterior portions, fused in medial portion to form tent-like brachiophorium, with ventro-medial portion forming crest-like structure (Figs. 1.7, 1.8, 1.10, 1.11, 2.1, 2.2); crura slender, rod-like, fused to inner surface of junction between inner and outer hinge plates (Fig. 3.1, 3.3), becoming free only at distal ends. Muscle scars not observed in either ventral or dorsal valves.
The internal structures of K. gothlandicus are partially similar to those of Pentamerus, especially in the relative size and shape of the spondylium, and the configuration of the posterior part of the hinge plates. In addition to the brachiophorium, however, the inner hinge plates of the Gotland species are proportionally much lower than in Pentamerus, in which the inner hinge plates are about half as high as the outer hinge plate (see Jin and Copper, Reference Jin and Copper2000). In the Gotland species, the inner hinge plates are only about one-quarter of the height of the outer hinge plates, with the outer-inner plate junction, marked by the crura, being close to the valve floor. The unusually large shell size of K. gothlandicus, typically exceeding 100 mm in length, is very rarely achieved by various Pentamerus species of Llandovery age (Jin and Copper, Reference Jin and Copper2000).
From the limited number of exposed shell interiors and sectioned specimens available for study, it appears that the fusion of outer hinge plates in the new genus occurs only in the middle portion of the plates. In their anterior and posterior potions, the configuration of the hinge plates is very similar to that in Pentamerus, except for the notably low inner hinge plates (Figs. 1.10, 2.6). There has been some doubt that the medially coalesced hinge plates may have been the result of preservation artifact, which explains why the brachiophorium was not considered an important diagnostic character in the revised brachiopod volumes of the Treatise on Invertebrate Paleontology (Boucot et al., Reference Boucot, Rong and Blodgett2002). In this study, however, a thin section of a well-preserved dorsal valve clearly shows that the lamellar layers of the two outer hinge plates are fused into a single lamellar layer in the coalesced part (Fig. 3.1, 3.2), which strongly suggests that the fusion was not a random compression of the two plates together during post-mortem taphonomic or diagenetic processes.
S. Eliason of the Gotland Museum (Visby, Sweden), C. Skovsted of the Swedish Museum of Natural History (Stockholm), and Z. Hughes of the Natural History Museum (London, UK) kindly arranged loans of specimens for this study. Z. Zhang’s assistance in fieldwork on Gotland is much appreciated. The helpful comments of two journal reviewers are greatly appreciated. Research funding was provided by the Natural Science and Engineering Research Council of Canada (Jin) and by the Swedish Research Council (VR 2012-1658, Holmer).