Abstract: Nautiloids of the superfamily Rutoceratoidea from the late Emsian (late Early Devonian) of the Prague Basin (Czech Republic) are commented upon. Species recognized include the hercoceratids Hercoceras mirum, H.? transiens, Ptenoceras proximum, P. nudum, P. minusculum and Anomaloceras anomalum, as well as the rutoceratids Adelphoceras bohemicum, Homoadelphoceras devonicans, Pseudorutoceras bolli and Goldringia? devonicans. In addition, four new species are described: Parauloceras regulare sp. nov., Roussanoffoceras chlupaci sp. nov., Otomaroceras sp. nov. and Goldringia sp. nov. Morphology and distribution patterns of Pragian and late Emsian rutoceratoid faunas from the Prague Basin are compared. They show that an increased diversity was accompanied by a higher level of specialization of rutoceratoids, which manifested itself in low abundance, increased facies dependence and greater variation in shell size during the Early Devonian. The evolution of sculpture and a contracted aperture, both regarded as protective adaptive features, was also examined, but no adaptive trend towards more pronounced sculpture and constriction of the aperture was found to have occurred in the Early Devonian. A more distinctive sculpture was, however, observed in shallow-water assemblages of P. proximum in comparison with deeper-water faunules, and two additional cephalopod species were examined in order to obtain comparative data. The presence of distinct sculpture patterns in coeval shallow- and deeper-water assemblages suggests limited migration between them and consequently reflects some degree of territoriality in Devonian nautiloids. New data on early shell development in P. proximum are presented. During the Choteč Event, rutoceratoid generic diversity dropped dramatically, one family became extinct and the Early Devonian diversification of the group came to an end. The recovery of nautiloid faunas was slower than that of other cephalopods and associated, unrelated invertebrates. The absence of change in abundance patterns between Pragian and late Emsian rutoceratoid faunas, i.e. prior to and subsequent to ammonoid radiation, suggests that the appearance and radiation of the latter group in the early Emsian did not affect the structure of nautiloid assemblages, i.e. these two clades did not occupy the same niches.
C ephalopods are significant predators in Recent seas; in fact, they already were so as far back as the Early Ordovician. The composition and structure of cephalopod faunas changed substantially during their geological history. The relative stability in cephalopod faunas after the Ordovician radiation ended in the Devonian when these faunas changed markedly and ammonoids, nautilids and many nautiloid families appeared. However, changes in the Devonian marine ecosystems are still poorly understood, and many models and explanations have been put forward. Amongst these is the hypothesis of a ‘Middle Palaeozoic Revolution’, as suggested by Signor and Brett (1984), which has often been discussed subsequently (e.g. Brett 2003; Kröger 2005; Klug 2007; Klug et al. 2008, 2010). The last-named authors collective explained some changes in morphology and abundance of invertebrates as because of an increased diversity and abundance of durophagous predators. The increasing diversity amongst nautiloids, with highly elaborated sculpture, was presented as an example. Without exception, all well-sculptured Devonian nautiloids are assigned to the Rutoceratoidea, which thus represents a model group for testing the present hypothesis. However, the majority of known taxa were described in the late nineteenth century and therefore are insufficiently documented and in need of revision.
The rutoceratoids (superfamily Rutoceratoidea Hyatt, 1884) comprise oncocerid nautiloids with curved and coiled exogastric shells and usually with highly elaborate sculpture and characterized by distinct outgrowths (wings, spines, nodes and collars). High disparity in shell form and growth structure patterns as well as a wide range of intraspecific variation is exceptional amongst nautiloids. Rutoceratoids include both species with small shells as well as the largest of known Palaeozoic nautiloids. With 27 genera and in excess of a hundred described species, they represent the second most diversified group of Devonian nautiloids, the most diverse being the oncocerid family Entimoceratidae Zhuravleva, 1974.
Rutoceratoids arose close to the Lochkovian/Pragian boundary, disappeared in the early Frasnian (for summary, see Manda and Turek 2009a) and formed a characteristic component of the temperate-water faunas of the Early/Middle Devonian Old World (e.g. Zhuravleva 1974, Dzik and Korn 1992, Turek 2007) and Middle Devonian Eastern American realms (Flower 1945, 1957). The Prague Basin (central Bohemia) represents one of the classic areas of fossiliferous Devonian strata and a long tradition of research and fossil collecting (for summary, see Chlupáč 1993, 1998, 1999). Highly refined litho- and biostratigraphic subdivisions and extensive collections thus provide an extraordinary opportunity for detailed studies. Large collections of rutoceratoids have been made from the Pragian Praha Formation and the late Emsian Třebotov Limestone. The Pragian rutoceratoids have recently been revised by Turek (2007) and Manda and Turek (2009a).
Our current knowledge regarding late Emsian rutoceratoids from the Daleje-Třebotov Formation is summarized and evaluated in the present paper, and four new species are described. Morphology, distribution patterns and evolution of rutoceratoids in the Early Devonian are discussed.
The subclass Nautiloidea is used as defined by Teichert (1988), i.e. including the orders Discosorida, Oncocerida, Tarphycerida and Nautilida. According to this concept, straight-shelled cephalopods of the order Orthocerida, previously placed within the Nautiloidea, are excluded from nautiloids and placed in a separate subclass, Orthoceratoidea. Consequently, the nautiloids comprise cephalopods with similar general morphology, embryonic development and ontogeny as the Recent genus Nautilus, whereby we use the term ‘nautiloids’ to have the usual meaning in relation to palaeobiological studies and examination of long-term evolutionary trends (see Manda 2008; Kröger and Zhang 2009; Manda and Turek 2009b).
All specimens, except those figured in Plate 1, figures 1–7, 10–14 and Text-fig. 3A–C, E, F–G, were coated with ammonium chloride prior to photography.
Institutional abbreviations. NM L, National Museum (Prague); CGS, Czech Geological Survey Prague; in particular Š. Manda (prefix SM), I. Chlupáč (ICH) and ‘Palaeontological’ collections (CGS p); MCZ, Museum of Comparative Zoology, Harvard University (Cambridge).
Cephalopods from the Late Emsian Třebotov Limestone: An Overview
Some of these species have not been revised since Barrande’s time. Their modern generic assignment is in some cases unclear, especially in view of their state of preservation. Specimens were usually affected by postdepositional deformation. Moreover, they are often badly corroded, being generally preserved as internal moulds, occasionally with shell remains. However, internal features such as septa and siphuncle are frequently very well preserved. Despite the fact that the number of described species may exceed the real number of species that existed (for discussion see Dzik 1984; Turek and Marek 1986), the diversity of late Emsian nonammonoid cephalopods in the Prague Basin was relatively high, as an outline by Zhuravleva (1972, 1974, 1978) illustrates.
Chlupáč (1959) and Chlupáčet al. (1979) published data on the stratigraphic range of some of Barrande’s species, although the status of certain taxa mentioned is in need of revision. The great majority of nautiloids from the Třebotov Limestone have largely been incorrectly considered in the literature to be of Middle Devonian age (e.g. Kummel 1964; Zhuravleva 1972, 1974; Teichert et al. 1980; Kröger 2005), which has had an impact on conclusions concerning diversity and macro-evolutionary trends amongst nonammonoid cephalopods close to the Early/Middle Devonian boundary. Barrande’s ‘etage G-g3’ includes the Třebotov Limestone (late Emsian) and Choteč Limestone (Eifelian) as currently defined (Chlupáč 1983a). The lithological characteristics of specimens preserved in light grey micritic limestone (exceptionally in dark red micritic limestone) as well as remarks in Barrande’s field books (see Chlupáč 1983a, 1999) testify to the fact that nautiloids assigned to ‘Hlubočep G-g3’ and ‘Holyn G-g3’ originate from the Třebotov Limestone and thus are of late Emsian age. The age of a few specimens preserved in dark grey micritic limestone of Barrande’s ‘etage G-g3’ remains questionable; they may have come from the basal portion of the early Eifelian Choteč Limestone (Text-fig. 2).
Notes Regarding Late Emsian Rutoceratoidea from the Prague Basin
Barrande (1865b–1877) described 11 species which subsequent authors referred to the rutoceratoids, from the lower part of his ‘etage G-g3’, which corresponds to the late Emsian, Daleje-Třebotov Formation (see Chlupáč 1998). Ten of these are considered valid, and each of them is briefly discussed below. Species that have never been previously photographed are now illustrated in that way. Anomaloceras anomalum is redescribed, and the generic diagnosis is emended. Four new species are described in the systematic section, but only two of them are formally named.
Hercoceras? transiens (Barrande, 1865b). Holotype, by monotypy, is NM L 8061, which was originally assigned to the genus Trochoceras and illustrated by Barrande (1865b, pl. 30, figs 13–17) from the Třebotov Limestone at Prague-Hlubočepy (see Pl. 1, figs 3–4). The specimen is an internal mould, only moderately affected by deformation, but strongly corroded. In the plane-coiled shell, with two whorls, the adapertural part of the last whorl markedly diverges. The aperture is slightly contracted. Hyatt (1894) considered T. transiens to be assignable to Hercoceras. The absence of lateral outgrowths in T. transiens is probably due to poor preservation, but the general morphology of the shell is clearly identical with Hercoceras. Zhuravleva (1974) supported Barrande’s original combination. Nevertheless, Trochoceras transiens differs from T. davidsoniBarrande, 1865b (Early Devonian, Pragian of Bohemia) in having a markedly depressed cross-section and in lacking intrasiphonal deposits and lateral outgrows near the aperture.
Ptenoceras proximum (Barrande, 1865b). Holotype, by monotypy, is NM L 10085, originally assigned to the genus Gyroceras, illustrated by Barrande (1865b, pl. 103, figs 12–14) from the Třebotov Limestone at Prague-Hlubočepy. The specimen is a corroded internal mould, of which the ventral surface was superficially artificially abraded to expose the siphuncle; the embryonic shell is preserved. In addition to the holotype, about 60 specimens were available for study; these are mostly undeformed internal moulds, but occasionally shell remains are preserved. The diameter of fully grown specimens ranges between 30 and 50 mm. This is an easily recognized species because of the characteristic whorl section and sculpture on the border area between the dorsal and ventral sides and one pair of small lateral processes near the aperture in adult specimens.
Hyatt (1894) was the first to place this fairly common species in his newly established genus Ptenoceras (type species: Ptenoceras alatum of Pragian age) in the family Hercoceratidae. Although Barrande (1865b) figured only a single specimen, about 60 specimens are available for study (National Museum and Czech Geological Survey collections).
Ptenoceras nudum (Barrande, 1865b). Three specimens of Gyroceras nudum were illustrated by Barrande (1865b, pl. 43, figs 8–9 (NM L 9085), figs 10–11 (NM L 9088) and fig. 12 (NM L 9086)); all are from the Třebotov Limestone at Prague-Hlubočepy. As a result of the preservation of Barrande’s specimens, it was not possible to determine with certainty whether all specimens do belong to the same species. They differ especially in the shape of shell and cross-section. The largest specimen (NM L 9085) is an internal mould, strongly affected by deformation. In its strongly depressed shell and shell size, it resembles Hercoceras mirum, although no traces of ventrolateral outgrows are visible. An artificially abraded fragment of a shell (NM L 9086) displays the siphuncle. Its systematic position is questionable especially in view of its extreme lateral expansion rate. Specimen NM L 9088 illustrated by Barrande (1865b, pl. 43, figs 10, 11) is designated lectotype herein. It is not deformed; the only slightly superficially artificially abraded shell represents the fully grown stage with an enrolled adapertural part of the body chamber. A pair of ventrolateral nodes is preserved. As far as shell morphology is concerned, it is similar to a well-preserved specimen illustrated by Turek (2009, fig. 3a), which preserves the colour pattern. In total, four specimens are known (National Museum, CGS p4912).
Gyroceras nudum was considered to belong to Hercoceras by Hyatt (1894). Turek (2009) transferred it to Ptenoceras on account of its similarity both to the type species of that genus, P. alatum, and to P. proximum.Turek (2007) and Manda and Turek (2009) pointed out that Ptenoceras and Hercoceras may be distinguished not only on general shell shape and cross-section but also on the appearance of lateral outgrowths rather than the mode of coiling or character of outgrowths (wings vs spines); in Ptenoceras, one or two pairs of outgrowths appear just before the end of shell growth (wings or spines), while in Hercoceras, outgrowths (spines, rarely wings) are already visible on the first whorl. However, on internal moulds, these features are frequently poorly visible or even missing.
Ptenoceras minusculum (Barrande, 1865b) comb. nov. Holotype, by monotypy, is NM L 8060. The type specimen, originally assigned to Gyroceras, was illustrated by Barrande (1865b, pl. 30, figs 18–21) from the Třebotov Limestone at Prague-Hlubočepy (see Pl. 1, fig. 2). A strongly corroded internal mould ventrally artificially abraded to expose the empty siphuncle. The position of the siphuncle was noted to be eccentric by Barrande (1865b), but this cannot be confirmed. The shell is loosely coiled, expands moderately and is strongly depressed. A pair of lateral outgrowths (not shown in Barrande’s original figure) is present near the aperture, and there is a faint indication of a node sinistrally. Their shape was similar to those in P. proximum and P. nudum. All these features indicate a close relationship between this species and the genus Ptenoceras. In addition to the holotype, only two additional specimens are known (CGS SM 338, National Museum collections).
Holotype, by monotypy, is NM L 21496, illustrated by Barrande (1870, pl. 459, figs 1–4) from the Třebotov Limestone at Prague-Hlubočepy. Adelphoceras Barrande, 1870 is a monospecific genus and its type species being poorly known. The holotype, which is the sole specimen known, represents half of the last whorl and is preserved as a slightly deformed internal mould (Pl. 1, figs 11, 12, 14). Hyatt (1884, 1894) classified the latter genus within the Rutoceratidae, and this assignment was largely accepted (Flower 1950a; Ruzhencev et al. 1962; Kummel 1964; Zhuravleva 1972; Dzik 1984 and others). Although the aperture in the holotype is moderately contracted, the apertural margin is not preserved, and Barrande’s reconstruction with a T-shaped aperture cannot be considered anything else than hypothetical (Dzik 1984). Flower (1945) mentioned actinosiphonate deposits; however, this observation has not been confirmed. The apparently slightly torticonic shell of A. bohemicum mentioned by Furnish and Glenister (inKummel 1964) may be a reflection of diagenetic processes.
‘Adelphoceras’ secundum (Barrande, 1877)
Holotype, by monotypy, is NM L 21897, illustrated by Barrande (1877, pl. 461, figs 4–6; pl. 474, fig. 1) from Hlubočepy G-g3. The holotype is a deformed internal mould preserved in dark grey micritic limestone, which corresponds to either the upper portion of the Třebotov Limestone or the basal levels of the Choteč Limestone. It represents half a whorl; the diameter of the whorl is 163 mm, the maximum height and width 54 mm and 93 mm, respectively. The shell is convolute, exogastric, in cross-section strongly depressed, with a shallow, broad impressed zone; the siphuncle is slightly shifted from the ventral side. We have been unable to observe the growth lines and recurrent growth ridges with a shallow, broad ventral lobe and lateral lobes, which are so clearly shown in Barrande’s illustrations. Barrande (1877) originally assigned this species to Adelphoceras, which, however, is characterized by an irregularly coiled shell, nodes, contracted aperture and growth structures without lateral lobes. No diagnostic feature of rutoceratoids is visible in ‘A.’secundum.
The holotype, NM L 15414, was illustrated by Barrande (1866, pl. 240, figs 1, 2, as Cyrtoceras devonicans); it is from the Třebotov Limestone at Prague-Hlubočepy (see Pl. 1, fig. 1). The specimen is an internal mould of the body chamber with a single phragmocone chamber. In spite of strong corrosion, conspicuous longitudinal ribs combined with distant transverse ribs are clearly visible. The length of the longitudinal ribs is markedly less than that of the transverse ones. Thus, the surface sculpture is reminiscent of that of the Early/Middle Devonian genus GoldringiaFlower, 1945, in particular of the Givetian G. cyclops (Hall, 1861) from New York State (see Hall 1892, pl. 54, fig. 1). The siphuncle is very probably situated close to the ventral side, but the internal structure is unknown. On account of the very poor preservation, taxonomic assignment remains uncertain although its assignment to the family Rutoceratidae is well supported.
Homoadelphoceras devonicans (Barrande, 1866)
Holotype, by monotypy, is NM L 10320. Barrande (1866, pl. 240, figs 16, 17) illustrated it from the Třebotov Limestone at Prague-Hlubočepy (Pl. 1, figs 5, 10). The specimen is an internal mould of the phragmocone with incomplete body chamber, which is obliquely longitudinally dislocated, preserved in light grey biomicritic limestone, with the phragmocone being strongly deformed by compaction. The diameter of the shell is 36 mm, its maximum length and width being 53 mm and 70 mm, respectively. The planispirally coiled, depressed shell appears expand laterally rapidly, but the apical angle may have been considerably influenced by deformation and corrosion. The body chamber was probably less coiled than the phragmocone so that the shape of the shell was probably similar to that in Adelphoceras. Six rows of prominent tubercles (two laterals on each side and one ventral pair) are the most characteristic feature of the sculpture seen on the corroded surface of the internal mould. They were probably situated on transverse ribs. The first artificially abraded preserved phragmocone chamber reveals a ventrally situated siphuncle with radial plates within the septal foramen. Flower (1950a) placed Homoadelphoceras in the family Rutoceratidae, and this assignment has been largely accepted (Ruzhencev et al. 1962; Kummel 1964; Zhuravleva 1972 and others). According to Dzik (1984), the type specimens of H. devonicans and A. bohemicum are conspecific. Because of poor preservation of both specimens, we can neither confirm nor refute this opinion. With the exception of the extreme expansion rate in H. devonicans, which may in part be the result of secondary diagenetic processes, we have not found any important difference between H. devonicans and A. bohemicum.
Pseudorutoceras bolli (Barrande, 1877)
Lectotype (designated by Manda and Turek 2009a) is NM L 24212 as illustrated by Barrande (1865b, pl. 42, figs 3, 4, as Cyrtoceras bolli) from the Třebotov Limestone at Prague-Hlubočepy. This species is the type of PseudorutocerasManda and Turek, 2009a; the surface sculpture (undulating frills) that resembles a colour pattern was recently discussed by Turek (2009). In total, four shells of P. bolli are known, all of them being figured types.
Type species. Cyrtoceras pupus Barrande, 1877, designated by Manda and Turek (2009a), Early Devonian (Pragian), Prague Basin.
Discussion. To date, Parauloceras is known only from Early Devonian strata of the Prague Basin where it is represented by an evolutionary lineage containing two closely related species, Parauloceras pupus (Barrande, 1877) from the Early–Middle Pragian (Dvorce-Prokop and Loděnice limestones) and P. regulare sp. nov. from the late Emsian (Třebotov Limestone).
1877 Cyrtoceras pupus Barr.; Barrande (partim), pl. 464, figs 8–10.
1877 Cyrtoceras pupus Barr.; Barrande (partim), pp. 41–42.
2001 Uloceras sp. nov. Manda, p. 270.
2009 Parauloceras sp. nov. Manda and Turek, p. 134.
Derivation of name. From the Latin adjective regulare (regular).
Types. Holotype is NM L 34506 (Text-fig. 3A–C); paratype is an unregistered body chamber with two phragmocone chambers in Barrande’s Collection (National Museum, Prague).
Type locality and horizon. Prague-Holyně (Prague Basin, Bohemia); Třebotov Limestone (Daleje-Třebotov Formation), late Emsian.
Material. Only the types are known to date.
Diagnosis. Parauloceras with relatively high phragmocone chambers, four raised growth ridges on the body chamber (in fully grown specimens), moderately depressed cross-section and thin siphuncle.
Description. The holotype is an internal mould of a slightly curved, exogastric and longiconic shell. Maximum shell length, height and width are 63, 14 and 17 mm, respectively. The angle of expansion is c. 15 degrees. In cross-section, the shell is depressed (height/width ratio 0.8). The septa are moderately concave, the suture straight and oblique. The height of the phragmocone chambers increases from 2 to 5 mm. The siphuncle is thin, with a maximum diameter of 2 mm; it is in contact with the shell wall, and the connecting rings are very moderately vaulted. Four straight growth ridges with ventral lobes are visible on the body chamber whose length is 26 mm. The aperture is open. The hyponomic sinus is shallow and broad. A second available specimen (see Barrande 1877, pl. 464, figs 8–10) is a body chamber with two phragmocone chambers; it exhibits the same morphological features as the holotype.
Type locality and horizon. Prague-Hlubočepy (Prague Basin, Bohemia); Třebotov Limestone (Daleje-Třebotov Formation), late Emsian.
Description. CGS SM 339 is an internal mould of a near-complete shell, lacking the apical part. The shell is a planispirally tightly coiled exogastric shell. In cross-section, it is depressed (height/width ratio 0.8). The length of the phragmocone chambers varies between 2 and 3 mm (measured on the lateral side). The suture is straight with broad ventral lobes. The body chamber is expanded and less coiled than the phragmocone. The shell is preserved only on the body chamber; fine growth lines are intercalated with recurrent growth ridges, the distance between them being c. 5 mm; growth structures are straight, oblique with shallow, narrow ventrolateral lobes and a broad ventral lobe. The body chamber is 56 mm in length. The aperture is open and oblique with a lateral sinus. The maximum shell diameter, height and width are 60, 21 and 27 mm, respectively.
Discussion. Otomaroceras sp. nov. resembles O. tardum from Pragian strata, but differs in having more densely packed and more oblique recurrent growth ridges with minute ventrolateral lobes and a less coiled body chamber than phragmocone in the fully grown shell. Otomaroceras sp. nov. represents the first record of this genus from strata of Emsian age.
Diagnosis (emended). The shell is coiled, evolute, with a broad, shallow impressed zone, the whorl section being reniform and strongly depressed. The thin empty siphuncle is slightly expanded within the phragmocone chambers; it is moderately shifted dextrally from the sagittal plane. The shell has prominent growth lines, with accentuated ridges forming a ventrolateral sinus indicating the presence of ventrolateral outgrowths adaperturally.
Discussion. Hyatt (1884) established the present genus based on Barrande’s Nautilus anomalus in the family Hercoceratidae; no comments were added to the generic diagnosis. Later, Hyatt (1894) supported his previous statement and suggested that Anomaloceras was remarkable in the eccentric position of its siphuncle, but in its shell form and sculpture resembled Hercoceras. Zhuravleva (1974) transferred Anomaloceras to the Late Palaeozoic family Aipoceratidae Hyatt, 1883 (Nautilida). Dzik (1984, pp. 86, 91), albeit with a query, synonymized Anomaloceras anomalum with Nothoceras bohemicum (Barrande, 1856). Nothoceras, however, differs from Anomaloceras in having a thicker siphuncle with actinosiphonate deposits and a suture with deep ventral saddles. Nevertheless, in the same paper, Dzik (1984, p. 156) also synonymized Anomaloceras with Hercoceras. Manda and Turek (2009a) did not find any substantial rutoceratoid diagnostic feature in Anomaloceras. During the latest revision of Barrande’s collection, a specimen with preserved growth structures, including lateral recurrent growth lines with lateral lobes, was examined (see Pl. 1, fig. 8). This feature clearly demonstrates that Anomaloceras belongs to the Hercoceratidae. With respect to the reniform cross-sectional, widely open aperture, narrower dextrally shifted siphuncle and appearance of ventrolateral outgrowths in the late growth stage, it differs markedly from Hercoceras.
The wide evolute shell of Anomaloceras is rather exceptional amongst Early Palaeozoic nautiloids and resembles that of some Late Palaeozoic nautilids (see Zhuravleva 1974). A planispirally coiled shell usually exhibits a siphuncle positioned in the median plane; however, all A. anomalus shells available for study exhibit an eccentric siphuncle shifted to the right. The function of an eccentrically situated siphuncle is questionable. It may represent a nonadaptive feature, inherited from an ancestor having a slightly torticonic shell. Asymmetry in the position of the siphuncle has also been ascertained in some specimens of Ptenoceras alatum, P. nudum, Hercoceras mirum and H.? transiens; in Anomaloceras, however, this feature is the most striking.
1865 Nautilus anomalus Barr., Barrande, pl. 34, figs 3–6.
1884 Anomaloceras anomalus Barr; Hyatt, p. 283.
1889 Anomaloceras anomalum; Hyatt, pp. 494, 599, pl. 8, figs 16–20.
non 1895 Nautilus anomalus Barr.; Katzer, pp 7, 8, pl. 2, figs 8–10.
1926 Anomaloceras anomalum (Barrande, 1865); Foerste, p. 382.
1962 Anomaloceras anomalum (Barrande, 1865); Ruzhencev et al., p. 382.
1964 Anomaloceras anomalum (Barrande, 1865); Kummel, p. 416.
1974 Anomaloceras anomalum (Barrande, 1865); Zhuravleva, pp. 137, 138.
1984 Nothoceras bohemicum; Dzik (partim), pp. 86, 91.
Lectotype. NM L 8057, illustrated by Barrande (1865, pl. 34, figs 3–5), designated herein (see Pl. 1, figs 1, 7, 9); paralectotype is NM L 8066 (illustrated by Barrande 1865b, pl. 34, fig. 6).
Type locality and horizon. Prague-Hlubočepy (Prague Basin, Bohemia); Třebotov Limestone (Daleje-Třebotov Formation), late Emsian.
Material. In addition to the types, eight additional specimens, all in the Barrande Collection. A specimen figured by Hyatt (1889) and deposited in the Schary Collections at the Museum of Comparative Zoology could not be traced during our recent visit.
Description. An exogastric shell, evolute with a maximum of two and a quarter whorls; the adapertural part of the shell in fully grown specimens may be enrolled, with the aperture widely opened. The imprint zone is broad and shallow. The angle of expansion in a lateral view is c. 12 degrees, whereas in ventral view, it is 21 degrees. In cross-section, the shell is reniform and strongly depressed (height/width ratio 0.5). The siphuncle is ventral, not in close contact with the outer shell wall, empty and thin; the segments are fusiform with short septal necks, being orthochaanitit to subortochoanitic. Septa very shallow. The suture is very slightly undulating, with a broad, shallow ventral lobe, narrow, lateral and dorsal saddles are faintly visible. The phragmocone chambers are of low height, 15 per adapertural half of the whorl in the paralectotype, 12 in the lectotype. There is some indication of recurrent ventrolateral nodes in later growth stages. Growth lines, oblique to the axis laterally, but ventrolaterally may form a small deep sinus indicating the presence of two pairs of ventrolateral outgrowths in fully grown specimens; ventrally they form a shallow, broad hyponomic sinus. The length of the body chamber is about half a whorl. Maximum diameter, width and height of the shell are 100, 79 and 38 mm, respectively.
Remarks. All specimens available for study are internal moulds. In one specimen only (Pl. 1, fig. 8), a trace of shell sculpture is preserved ventrolaterally in the adapertural part of the phragmocone. It documents the primary presence of ventrolateral outgrowths in this part of the shell, morphologically similar to or identical with outgrowths appearing in Hercoceras and an important feature for classifying Anomaloceras amongst rutoceratoids. The specimen belongs to the unfigured original type series on which the genus was established. This is confirmed by Barrande’s inscription on the specimen (‘Hlubočep An 5’), which means ‘5th anomalum specimen’, as well as by the original label attached to the specimen.
A specimen from the Koněprusy Limestone, which was identified by Katzer (1895) as Nautilus anomalus, in fact most probably belongs to Ptenoceras alatum.
Occurrence. Prague-Hlubočepy and Prague-Holyně (single shell).
Discussion. Recurrent raised growth walls (megastriae) in Roussanoffoceras resemble those of GoldringiaFlower, 1945 from which genus it probably is derived. However, the present genus differs in having a greater angle of shell expansion (laterally as well as ventrally). In cross-section, it is compressed and ventrally flattened, the cross-section in Goldringia being subcircular or only slightly compressed. The new species constitutes the first record of the genus outside Novaya Zemlya.
Derivation of name. After Ivo Chlupáč (1931–2002) for his contributions to stratigraphy and palaeontology of the Early Palaeozoic.
Types. Holotype is NM L 40788 (see Text-fig. 3E, G); paratypes are NM L 13489, NM L 27420 and NM L 27421.
Type locality and horizon. Prague-Hlubočepy (Prague Basin, Bohemia); Třebotov Limestone (Daleje-Třebotov Formation), late Emsian.
Diagnosis. Roussanoffoceras with larger shell, closely spaced transverse ribs and a slightly to moderately depressed cross-section.
Description. A gyroceraconic exogastric shell with two whorls. The angle of expansion is c. 15 degrees in lateral view and 20 degrees in ventral view. In cross-section, the shell is depressed and dorsally flattened (height/width ratio 0.7) or subcircular and only very slightly depressed. The siphuncle is marginal, without contact with the shell wall, empty and thin; the connecting rings are weakly developed, and the septal necks are short. The septa are very shallow. The suture is straight, oblique with a shallow dorsal lobe. The length of the phragmocone chambers increases from 3 mm (height 18 mm) to 7 mm (height 45 mm). Recurrent ribs are intercalated with gentle growth lines. The course of the ribs is straight and oblique to the axis of the shell; on the ventral side, shallow ventral saddles may be seen. The distance between individual ribs increases from 9 mm (height 22 mm) to 21 mm (height 55 mm). The body chamber is 48 mm in length, with a shell height of 37 mm. The aperture is open. Maximum shell length, height and width are 140, 55 and 49 mm, respectively.
Discussion. All available specimens are similar in general shell shape, but differ in cross-section, two specimens being slightly depressed (Text-fig. 3H) and two others markedly depressed (Text-fig. 3E, G). Intraspecific variability in cross-section within the Rutoceratoidea is considerable, usually being relatively high. In four of the available shells of R. chlupaci sp. nov., it is not possible to examine the variability in more detail. Variation in cross-section (wider vs narrower shell) may also reflect sexual dimorphism (for summaries see Teichert 1964; Ward 1987).
Remarks. Two specimens deposited in the National Museum were labelled as Gyroceras nude Novák, ‘Gyroceren böhmens’. However, O. P. Novák (1851–1892) never published any paper with this classification.
Occurrence. Early Devonian (late Emsian); Třebotov Limestone (Daleje-Třebotov Formation) of the Prague Basin, Bohemia (localities Praha-Hlubočepy and Praha-Holyně).
Type species. Gyroceras cyclopsHall, 1861, designated by Flower (1945), Middle Devonian (Eifelian), New York State, USA.
Discussion. Goldringia is a common rutoceratoid in the Middle Devonian of New York State and adjacent areas in the Eastern American Realm (Flower 1945, 1957; Baird and Brett 2008). The Early Devonian record of Goldringia is poorly documented. Goldringia gondolaManda, 2001, the oldest known species of the genus, occurs in Pragian and early Emsian strata of the Prague Basin (Manda 2001; Turek 2007; Manda and Turek 2009a). Slightly younger is G. valnevensisZhuravleva, 1996 from the late Pragian of Novaya Zemlya (Russia) and Goldringia sp. from the early Emsian of the Robert Mountains, Nevada (unpublished specimen, CGS SM 343). All other species known to date are of Middle Devonian age (Flower 1945; Zhuravleva 1974). Consequently, Goldringia sp. nov. (and perhaps G.? devonicans, see above) from the late Emsian of the Prague Basin fills a gap.
Type locality and horizon. Prague-Hlubočepy (Prague Basin, Bohemia); Třebotov Limestone (Daleje-Třebotov Formation), late Emsian.
Description. The single shell available is part of a whorl with a partially preserved sculptured body chamber; on the right side, the shell is deeply corroded. The shell was probably loosely planispirally, exogastrically coiled and slightly expanding with its maximum diameter and shell height being 54 and 18 mm, respectively. In cross-section, it is subcircular. The siphuncle is thin and ventral. The phragmocone chambers are low in height (observed on the right-hand corroded side). Septa are very slightly convex. The suture is straight and on the lateral side. The maximum preserved length of the body chamber is 27 mm. Sculpture consists of very fine growth lines and recurrent raised growth walls, their distance increases from 2 mm (in diameter 10 mm) to 5 mm (in diameter 18 mm); growth structures are straight with a shallow, narrow ventral lobe.
Discussion. Goldringia sp. nov. differs from all congeners in having relatively low recurrent growth ridges and undulated growth lines. Transverse ridges have not been observed.
The Daleje-Třebotov Formation: Geological Setting and Preservation Conditions
In the Prague Basin, the latest Emsian Daleje-Třebotov Formation is represented by three principal facies, the Daleje Shale, Třebotov Limestone and Suchomasty Limestone. For a summary, reference is made to Chlupáč (1998). The Daleje Shale (Barrande’s ‘etage G-g2’) comprises green, grey and reddish shale and is found in the north-eastern part of the Prague Synform, passing laterally into the Třebotov Limestone. Chlupáč (1959) suggested that this unit alternated with Třebotov Limestone in areas with a high siliciclastic influx. So far, no nautiloid has been described from this facies, but flattened shells of ammonoids and orthoceratoids are quite common (Barrande 1865b–1877; Chlupáč 1959; Chlupáč and Turek 1983).
The lower part of the Třebotov Limestone (thickness c. 16 m) is developed as platy red-coloured limestones intercalated with shales (Polygnatus serotinus and Nowakia richteri zones) in the Prague-Hlubočepy area. Only a few rutoceratoids housed in old collections originate from these beds. The upper part of the Třebotov Limestone (thickness c. 18 m) is developed as light grey, coarsely bedded biomicritic limestones (latest P. serotinus–early P. costatus partitus and N. richteri–N. holynensis zones; see Chlupáč 1959, 1993, Chlupáčet al. 1977, 1979, 1980). The vast majority of cephalopods assigned to ‘Hlubočep Gg3’ originate from these beds. The early Eifelian Choteč Limestone at Prague-Hlubočepy (thickness c. 6 m) is represented by well-bedded limestone, differing from the underlying Třebotov Limestone by subordinate, darker grey, fine-grained bioclastic intercalations and a slightly darker colour of biomicrites (Chlupáč 1959, 1993; Berkyová 2009). Consequently, the stratigraphic provenance of specimens in old collections may be evaluated upon rock character.
Cephalopods are usually preserved as more or less corroded internal moulds. However, the shells and their internal moulds are usually complete with preserved septa, only the apex is frequently missing (e.g. Pl. 1, figs 2–8). Large shells are commonly affected by limestone dissolution on their surface (e.g. Pl. 1, fig. 1). In addition, some larger shells are moderately deformed (e.g. Pl. 1, figs 12–14). Better preserved specimens were collected in particular from strongly weathered limestones, ‘white beds’, from which fossils were extracted by washing (e.g. Pl. 2, figs 2, 3) at Prague-Holyně, ‘George’ (Bouček 1931; Kříž 1999) and an unknown site at Prague-Hlubočepy (Text-fig. 1). Miners collected the vast majority of available cephalopods in the late nineteenth and early twentieth centuries (see Hanuš 1923; Kříž 1999) from large active quarries in the vicinity of the village of Hlubočepy (i.e. locality Hlubočep G-g3 in Barrande) and less frequently at the village of Holyně (i.e. locality Holín G-g3 in Barrande). Detailed descriptions of these quarries were provided by Barrande (1865a), Wahner (1916), Storm (1935) and especially by Chlupáč (1959) and Chlupáčet al. (1979, 1980). The assumed stratigraphic distribution of rutoceratoids based on lithological characteristics is shown in Text-figure 2. Interestingly, the Třebotov Limestone outside the areas of Hlubočepy and Holyně yielded only few nautiloids (Chlupáč 1959). This limited distribution of rutoceratoids in the Daleje-Třebotov Limestone suggests that most late Emsian rutoceratoids (similar to others nautiloids) inhabited a relatively narrow facies (depth) zone even within the depositional area of the Třebotov Limestone (Text-fig. 5).
A relatively small-sized benthic fauna, comprising gastropods, trilobites, hyolitids and rare brachiopods, co-occurs with nautiloids in the Třebotov Limestone; it indicates a well-oxygenated, firm muddy bottom, but below wave base. Some better preserved shells (Hercoceras) contain common crinoid holdfasts and in one case also Microconchus tubes. Nevertheless, the large crinoid reefs observed on some bedding planes of nodular limestones suggest the occasional presence of hardgrounds. Common nowakiids and stylolinids, juvenile orthoceratoids and bactritids are indicative of open marine conditions, while common, relatively complete cephalopod shells with preserved inner structures reflect a low-energy environment. Straight orthocerid shells are weakly orientated (Petránek and Komárková 1953). Consequently, it is assumed that rather weak, yet stable, bottom currents ventilated the lower level.
Contraction of the aperture is a reliable feature indicative of a fully grown shell (e.g. Flower and Teichert 1957; Stridsberg 1981, 1985, 1988; Manda 2008), which enables assessment of variability in shell size in cephalopod populations. In some cases, apertural modifications supposedly accompanied changes in the mode of life of nautiloids (Prell 1921; Flower 1957; Stridsberg 1981, 1985; Manda 2008). Configuration of the aperture depends totally on the accretion mode of shell material by the mantle, and any disturbance of the mantle’s outer edge would have produced variations in the normal apertural pattern of the species (Stridsberg 1981, 1985). The apertural shape is a feature that has been widely used for taxonomic purposes in nautiloids. However, as pointed out by Stridsberg (1985), the minor differences in morphology have occasionally been overestimated.
A contracted aperture is commonly developed in Early Palaeozoic nautiloids with straight or slightly curved breviconic shells (i.e. with a more or less downward-oriented aperture), but is exceptional in coiled nautiloids (in which the aperture was usually oriented anteriorly in life). It indicates a protective function of the contracted aperture, previously suggested by Teichert (1964). Hercoceras mirum and Adelphoceras bohemicum (Pl. 1, figs 11, 12, 14) are the only known Devonian rutoceratoids with a markedly contracted aperture. The shape of the fully grown aperture in the latter species is poorly known, because it is not completely preserved in the specimen available (Pl. 1, fig. 14). An apertural contraction of the aperture in Hercoceras was originally described and illustrated by Barrande (1865b, 1867) and recently mentioned by Turek (2007). Its character strongly resembles apertural modifications in the Ordovician coiled tarphyceratid MoreauocerasCullison, 1944, and this feature has also been reported from the Ordovician coiled tarphyceratid PilotocerasCullison, 1944 and the trocholitid GraftonocerasFoerste, 1925b. It may also be fairly common in other tarphyceratids and in the family Trocholitidae Chapman, 1857 (see Furnish and Glenister 1964), but this assumption has not yet been fully tested.
During a revision of Hercoceras mirum, 68 specimens with a constricted aperture were found. Other specimens available for study (mainly from the Barrande Collection) were either incomplete or did not represent the fully grown stages of the shell. The diameter of fully grown shells ranges between 68 and 125 mm (Turek 2007). Shell material near the aperture is usually missing; consequently, the character of the aperture has to be derived from the morphology of usually incomplete internal moulds (Text-figs 4, 5). If there is shell material preserved, it is markedly thickened near the constricted aperture (as in Silurian oncocerids and discosorids; compare Stridsberg 1985; Manda 2008).
The constriction is the result of differential growth of the aperture. While the accretion of shell material on the ventral margin had almost ceased, it continued dorsally but in a markedly different direction. In lateral view, the resulting aperture looks like a partially laterally closed visor (Text-fig. 5A–C) or, more commonly, the dorsal side is bent under a right angle (Text-fig. 5D). Ocular sinuses, if present, are only vaguely indicated. The result of this process is a markedly restricted apertural opening. In frontal view, almost two-thirds of the aperture may be closed. The dorsal side in this case is vaulted, sometimes with a faint, wide median groove or simply flat. The apertural margin forms a very shallow dorsal sinus. The boundary between the dorsal side of the shell adjacent to the previous whorl and the free part adjoining the aperture is usually rounded. In a few cases, when shell growth continued at right angles, this boundary is sharp. Owing to retarded growth of the shell on the ventral side and presence of a wide, shallow hyponomic sinus, the apertural opening remains fairly large. Exceptionally, the aperture in fully grown specimens is widely opened with only a slightly modified shape on the dorsal side (Text-figs 4E, 5E, F). From a deep, narrow sinus in the aperture, lateral spines originated, which morphologically (Text-fig. 4B–D) strongly resemble the spines of the Jurassic ammonite Aspidoceras (Checa and Martin-Ramos 1989).
The shape of the aperture in nautiloids reflects the morphology of the soft body close the aperture. Accordingly, in the fully grown adult stage of H. mirum, the majority of tentacles could not point directly forwards but faced obliquely downwards. In this growth stage, the animal’s tentacles were in closer contact with the sea floor. A similar effect was achieved in nautiloids with an uncoiled body chamber (e.g. some tarphyceratids; see Flower 1955). Eyes were located at about one-third of the dorsoventral diameter of the whorl from the ventral side, i.e. far lower than in the Recent Nautilus and Triassic GermanonautilusMojsisovics, 1902 (see Klug and Lehmkuhl 2004 and further references therein). The shape of the aperture, in addition to the general morphology of the shell, supports the assumed nektobenthic mode of life (Turek 2007). The advantage of a partially closed, downward-oriented aperture in the fully grown stage may be enhanced protection of the soft parts against predators close to the bottom.
In addition to the type, Hercoceras also includes some other Emsian–Eifelian species (Dzik 1984; Manda and Turek 2009a). However, a contracted aperture has been demonstrated only in H. mirum. This is interesting, because the contracted aperture is usually shared by all species of a genus or even family in the Nautiloidea, for instance in the Silurian families Hemiphragmoceratidae Foerste, 1926, Mandaloceratidae Flower, inFlower and Teichert, 1957 and Trimeroceratidae Hyatt, 1900. Similar to sculpture strengthening, the contracted aperture is considered to be an adaptive protective feature (see Teichert 1964). Consequently, a constricted aperture should be a progressively evolving feature if the Devonian radiation of durophagous predators is taken into account (see Signor and Brett 1984). However, the limited occurrence of a contracted aperture in rutoceratoids suggests that the adaptive pressure to retain this feature was relatively low.
Comparison of Pragian and Late Emsian Rutoceratoid Faunas in the Prague Basin
Comparison of Pragian and late Emsian rutoceratoid faunas is interesting with respect to possible evolutionary trends in the Early Devonian (see Signor and Brett 1984; Brett 2003; Kröger 2005; Klug 2007; Klug et al. 2008). Therefore, we comment here briefly on distribution patterns, abundance, shell morphology (mode of coiling, sculpture) and range of shell size in both faunas. Pragian rutoceratoids (similar to other nautiloids) inhabited various environments, ranging from reefs to deeper-water settings on carbonate slopes below storm base; maximum diversity, however, is found in deeper-water settings just below storm wave base (see Manda and Turek 2009a). In comparison with Pragian faunas, Emsian nautiloid assemblages were restricted to a narrow facies belt in a deeper-water environment. It is remarkable that, despite the facies restriction, the total diversity of rutoceratoids (as in other nautiloids) increased during the latest Emsian, when compared to Pragian strata (see Text-fig. 6).
Nautiloids are, as a rule, relatively rare in both Pragian and Emsian strata in the Prague Basin. Amongst them, rutoceratoids are the commonest (Text-fig. 7). Both faunas exhibit a similar pattern of abundance: one or two common species (exclusively members of Ptenoceras and Hercoceras; see Dzik and Korn 1992; Manda 2001; Turek 2007) are accompanied by several markedly rare taxa. A marked predominance of one cephalopod species is a typical feature of both faunas. Ptenoceras alatum is the dominant species in Pragian faunas, while in late Emsian assemblages, it is H. mirum. Calculated coefficients of dominance have relatively high values for both faunas. However, the latter value for late Emsian fauna (0.80) is significantly higher than that of the Pragian fauna (0.53). Dzik (1984, p. 187) concluded in a summary of his analysis of cephalopod phylogeny that ‘relatively variable compositions of the Devonian nautiloid faunas’ reflected the early radiation of ammonoids that ‘occupied many niches utilized previously by the nautiloids’. Nevertheless, the unchanged abundance pattern between the Pragian and late Emsian rutoceratoid faunas, i.e. prior to and subsequent of ammonoid radiation (e.g. Chlupáč and Turek 1983; Klug et al. 2008) suggests that the appearance and radiation of ammonoids did not affect the structure of nautiloid assemblages (see also Kröger 2008), i.e. the two cephalopod clades did not occupy the same niches. In addition, the very low abundance of rutoceratoids should be taken into account in palaeobiogeographic analyses, simply because there is a strong sampling effect between traditional and poorly known terrains.
Another interesting evolutionary feature is the development of shell size (for a summary see Jablonski 1996). Text-figure 8 shows the maximum dimensions of Pragian and late Emsian rutoceratoids in the Prague Basin. The latter exhibit a wider variation in shell size, and the average shell size increased in the late Emsian. A comparison of shell dimensions of nautiloids in Wenlock–early Lochkovian cephalopod assemblages from the Prague Basin (Manda and Turek 2009b; plus unpublished data) is interesting as it shows relative stability of shell size throughout that time period.
Rutoceratoids represent a single monophyletic Devonian clade of nautiloids with highly elaborate sculpture and shell outgrowths. Signor and Brett (1984) suggested that highly elaborate shell sculpture in nautiloids functioned as protection against predators, and they also pointed out that increased diversity of well-sculptured nautiloids during the Devonian represented an adaptive reaction to the radiation of durophagous predators. In fact, the majority of Pragian and late Emsian rutoceratoids exhibit almost identical growth sculptures. The Ptenoceras-Hercoceras line represents a single exception; in the late Emsian, lateral outgrowths in Ptenoceras were reduced, contrary to the situation in the derived species, Hercoceras mirum, in which the number of outgrowths significantly increased. Consequently, no consistent evolutionary trend is visible in the development of sculpture.
Kröger (2005) examined the diversity of tightly coiled forms and suggested an adaptive control of shell coiling. He correlated increased diversity of nautiloids with a nautiliconic shell with the radiation of durophagous predators. In addition, the diversity of rutoceratoids with a tightly coiled shell increased in the Early Devonian of the Prague Basin. Four Pragian rutoceratoids have an openly coiled shell; in the late Emsian, four species have an openly coiled shell, while the shell of seven others is tightly coiled. The trend from an openly to tightly coiled shell is recorded especially in the Ptenoceras-Hercoceras line.
Highly dynamic changes in distribution pattern and morphology in Devonian nautiloids in the Prague Basin may reflect changes in marine communities on a global as well as a regional scale. Comparison of Pragian and late Emsian rutoceratoid faunas in the Prague Basin clearly suggests that rutoceratoids were restricted to a narrow facies zone just below storm wave base, that diversity increased (Text-fig. 9), that there were identical abundance patterns in which one or two species is (are) common while others are very rare, that amongst rutoceratoids an increased in shell size is noted, that sculpture patterns remained relatively stable and that there was an increase in diversity of tightly coiled forms.
Variability of Sculpture in Ptenoceras and Palaeoecological Implications
Ptenoceras proximum is the sole late Emsian rutoceratoid known, which has been found in both the shallow-water Suchomasty Limestone and the deeper-water Třebotov Limestone. The species first occurred in the lowermost portion of the former unit (middle Emsian, latest Zlichovian, N. elegans Zone), in infill of submarine fissures (Neptunian dykes) at the Koněprusy–Voskop Quarry and in the Mramorová stěna Section (Chlupáč 1996). Shells have also been found in the pink- and red-coloured crinoidal limestone, which characterizes the Suchomasty Limestone, i.e. in old material designated ‘Koněprusy’, but with exact localities unknown (see Chlupáč 1983a) and that from the locality Koněprusy-U transformátoru (Chlupáč and Vaněk 1957). Ptenoceras proximum is much commoner in biomicritic limestones such as Třebotov Limestone, at the localities Hlubočepy and Holyně (see Text-fig. 4); the species appeared in the Třebotov Limestone slightly later than in the Suchomasty Limestone, but both populations were primarily coeval.
We have compared shell morphology in assemblages from both units. The mode of coiling, cross-section and direction of growth structures are almost identical. A distinct feature in which the two assemblages differ is sculpture. All seven shells available from the Suchomasty Limestone show strengthening of growth ridges in a narrow zone between the ventral and dorsal side; the growth ridges here are more pronounced and form small nodes. The vast majority of shells from the Třebotov Limestone are poorly preserved. However, shells with fine, or even very fine, growth lines may be distinguished. Shells exhibiting lateral strengthening of growth lines are also present. In addition, the strengthening in sculpture appears later in ontogeny in shells from the Třebotov Limestone than in those from the Suchomasty Limestone (Pl. 2). In summary, shells with fine growth lines predominate.
A smooth shell or gentle growth lines are developed in embryonic shells of Ptenoceras proximum from the Třebotov Limestone (Pl. 2, figs 1–4). Embryonic shells of this species from Suchomasty, however, exhibit a more elaborate sculpture. Fine irregular growth lines are developed on the apex; approximately at the position of the first phragmocone chamber, regular densely packed growth lines appear. With age, the distance between growth lines increases, and a ventral lobe may also be seen (Pl. 2, fig. 9). This change in sculpture probably corresponds with hatching time.
Although the shell morphology of adult specimens and size of embryonic shells are very similar (i.e. suggesting conspecificity), the tightness of coiling varies greatly in early shells of P. proximum. A similarly wide range of intraspecific variation in umbilical perforation has been reported for the Pragian rutoceratoid Ptenoceras (Turek 2007). Specimens from the Suchomasty Limestone show a very small, drop-like umbilical perforation (width 0.5–1 mm; see Pl. 2, figs 9, 11), while shells from the Třebotov Limestone usually have a much larger umbilical perforation (maximum width 2.9 mm; Pl. 2, fig. 4), but specimens with a small and mid-range umbilical perforation can also be found (Pl. 2, figs 3, 4).
The small hatching size in Ptenoceras (just before the shell reached one-quarter whorl) resembles that of the Silurian Phragmoceras and BoionautilusTurek, 2008 (Manda 2008; Turek 2008) and that of Mesozoic nautilids and extant Nautilus (Chirat and Rioult 1998, and further references therein) is much longer. Consequently, hatching size may be useful for higher-taxa grouping within the Nautiloidea, but more data are needed.
Strengthening of growth sculpture in a shallow-water environment is not surprising. Metabolic precipitation of calcium carbonate is more effective in warm and highly oxygenated water, i.e. a shallow-water setting. Many modern tropical gastropod molluscs exhibit strengthening of sculpture in a shallow-water environment (see Graus 1974; Vermej 1987). A similar pattern has been reported for Palaeozoic nautiloids by Hewitt and Watkins (1980), but, for example, a comparison of populations of Silurian nautiloids OphiocerasBarrande, 1865b and Phragmoceras from Gotland (a tropical carbonate platform) and the Prague Basin (a temperate zone) did not reveal any differences in shell sculpture (see Stridsberg and Turek 1997; Manda 2007b, 2008).
Distinct sculpture differences in occurrences of P. proximum in shallow- and deeper-water environments suggest that there was no significant migration between the two populations. Occasional migration from a shallow to a deepwater setting cannot be excluded, as a single shell of P. proximum from the Třebotov Limestone (see Pl. 2, fig. 1) shows; this exhibits a near-identical sculpture to shells from the Suchomasty Limestone. Consequently, Ptenoceras should be regarded as a more or less territorial animal, which, in general, did not migrate during its lifetime between different environmental settings. Findings of early posthatching specimens in the shallow-water Suchomasty Limestone as well as in the deeper-water Třebotov Limestone document the presence of specific hatching places in both environmental settings.
We have also compared the palaeogeographic differences in sculpture pattern in Early Devonian nautiloids and their sister clade, the pseudorthocerids, which have straight shells as in orthocerids, but a cup-like embryonic chamber with a cicatrix as in nautiloids. However, the vast majority of taxa are restricted to a certain facies or their abundance strongly changes in different facies, and thus, sufficient comparative material was not available. The two Pragian species selected for consideration were Ptenoceras alatum (Barrande, 1865b), which is the precursor of P. proximum (Manda 2001; Turek 2007) and the pseudorthocerid Suloceras pulchrum (Barrande, 1868).
Ptenoceras alatum exhibits a gyroceracone shell with growth lines and recurrent growth ridges, transforming during ontogeny into megastriae; two pairs of lateral outgrowths (wings) appear in fully grown shells (for details, see Turek 2007; Manda and Turek 2009a). All specimens examined exhibited growth structures of identical shape. Specimens from the Koněprusy and Slivenec limestones have a larger shell size, with a maximum diameter of 60 mm (compared to 53 mm in a specimen from the Dvorce-Prokop Limestone). Specimens from the Koněprusy and Slivenec limestones also show highly elaborate sculpture and lateral outgrowths (contrary to only two specimens from Dvorce-Prokop Limestone, which exhibit lateral outgrowths). In summary, specimens from shallow-water limestones exhibit a more elaborate sculpture, i.e. thicker growth lines (see specimens illustrated by Turek 2007). This corresponds with the observation in populations of P. proximum.
The Pragian Suloceras pulchrum is a relatively common species in the shallow-water Koněprusy and Slivenec limestones (at the Branžovy, Císařský, Homolák and Houbův quarries; see Manda 2001), but occurs rather sporadically in the muddy Dvorce-Prokop Limestone (at the Braník, Černá Gorge and Konvářka sections). It has a straight annulated shell, the annulation being well developed in shells with a diameter up to 15 mm, but in more mature growth stages, the annulae become less pronounced. The sculpture on early shells (up to a diameter of c. 7 mm) consists of a regular reticulate ornament, i.e. combinations of longitudinal ribs and straight growth ridges. Adult specimens from shallow-water limestones exhibit a differentiation of longitudinal ribs into 2 or 3 orders, and an annulation is always developed (Text-fig. 10C). Specimens from biomicritic deeper-water limestone, however, in general show weaker growth sculptures, while longitudinal ribs are differentiated into two orders (Text-fig. 10A), or even reticulate ornament resembling the early shells may be developed. The annulation is weaker in or even absent from larger shells (Text-fig. 10C–E). In summary, S. pulchrum exhibits strengthening of growth sculptures in a shallow-water environment as revealed by both species of Ptenoceras examined. Interestingly, there is a reduction or even lack of annulation in some adult specimens from muddy limestones.
Intraspecific variability in Devonian nautiloids and pseudorthocerids was probably greater than previously assumed. Some differences exist in shell coiling and shell size (as noted by Dzik 1984; Turek 2007; Manda and Turek 2009a), but as shown above, differences also exists in sculpture and annulation, both commonly used as species-diagnostic features, and therefore need careful consideration. ‘Facies dependence’ of some closely related ‘species’, which differ in sculpture or presence of weak annulation, could be an artificial effect of splitting. Hewitt and Watkins (1984) and Evans (1994) suggested that cephalopod taxa in shallow-water settings developed a better developed sculpture. It may be more correct to note, however, that shallow-water populations, not necessarily species, have a more elaborate sculpture. The presence of distinct nautiloid and pseudorthocerid morphotypes (phenotypes) in different environmental settings may be indicative of animal territoriality, i.e. document the limited migration of adult animals between facies/depth zones and thus the presence of local populations.
Effect of the Choteč Event on Rutoceratoids
The Choteč Event was studied in detail by Walliser (1984, 1985), House (1985), Chlupáč and Kukal (1986, 1988) and others. Walliser (1996, p. 230) concluded that, ‘Extinctions during the Choteč Event occurred in nearly all fossil groups of both neritic and pelagic facies’. However, House (2002, p. 14) suggested a rather weak extinction, but did note that the Choteč Event, ‘has the first clear characters of many later Devonian events’. In the Prague Basin, the type area of the Choteč Event, light grey muddy skeletal limestones were replaced by dark grey coloured crinoidal pack-grainstone within lighter mudstones during the Choteč Event (Chlupáč and Kukal 1986, 1988). Changes in facies are usually explained by deepening, coupled with anoxic conditions close to the bottom (for a summary, see Walliser 1996).
Taken as a whole, with the exception of Bolloceras, all nautiloid genera known from the Třebotov Limestone (late Emsian) are absent in the overlying Choteč Limestone (Eifelian). The rutoceratoids Adelphoceras, Anomaloceras, Homoadelphoceras, Parauloceras, Roussanoffoceras and Otomaroceras became extinct worldwide. Hercoceras, Ptenoceras (the most abundant Early Devonian rutoceratoids), Goldringia and Pseudorutoceras ranged up into the Middle Devonian (Text-fig. 9). A high rate of extinctions amongst rutoceratoids (and other nautiloids) probably reflects their specialization in addition to their restriction to a narrow facies zone in a deeper-water setting, just below storm wave base, in the Emsian. This muddy limestone biofacies, which nearly exclusively hosted rutoceratoids, was the only facies strongly affected by dysoxic conditions, which accompanied the Choteč Event (Text-fig. 6).
Recovery of nautiloid faunas took place in the late Eifelian and thus, much slower than ammonoid and benthic faunal recovery. A nautiloid fauna of low diversity, including Aphyctoceras sp. (Text-fig. 11B), appeared in the upper Acanthopyge Limestone (late Eifelian) in the shallow-water Koněprusy elevation area (Chlupáč 1959). Various species of Bolloceras are known from the uppermost Choteč Limestone (late Eifelian) in an old quarry in the Hluboké Valley at Karlštejn (Chlupáč 1959). The rutoceratoid Kophinoceras (Text-fig. 11A), which is a common element in Givetian faunas of the Old World Realm, appeared later in the latest Eifelian Kačák Shale (Srbsko Formation), i.e. just after the Kačák Event (Chlupáč 1960).
It may therefore be concluded that the Choteč Event led to a restructuralization of nautiloid faunas and prominent extinctions. It represents the first significant extinction event affecting cephalopods in Devonian time, following previous extinctions at the Silurian/Devonian boundary (see Manda 2001, 2007a; Kröger 2008). The Lochkovian/Emsian time interval may be considered as a time of nautiloid radiation following the Silurian-Devonian extinction events (see also Zhuravleva 1972, 1974; Kröger 2008; Manda and Turek 2009; Manda and Frdýda 2010). The rutoceratoids represent a well-documented example of this phenomenon.
Acknowledgements. This research was supported by the Czech Grant Agency (project GAČR 205/09/0260, VT) and by the Ministry of Education (project KONTAKT MEO8011l, ŠM). We thank Jessica Cundiff (Museum of Comparative Zoology, Harvard), Petr Budil and Radko Šarič (Czech Geological Survey, Prague) for providing the material used in this study and Radvan Horný (National Museum, Prague), Gill Horalek and journal reviewers Jerzy Dzik, Björn Kröger and an anonymous reviewer for critical reading of the manuscript and improving the English. Handling editor John W. M. Jagt is deeply acknowledged for manuscript revision. Jiří Frýda kindly helped us with statistical methods. Radko Šaric kindly loaned a specimen of P. proximum and V. Frank one of S. pulchrum from collections in their care. Ivan Kolebaba is acknowledged for drawing Text-figure 5.