Articulated ophiocistioids are rare as fossils, enigmatic in terms of their phylogenetic position, and have been considered to be an exclusively Palaeozoic echinoderm group. Apart from articulated body fossils, the fossil record of Ophiocistioidea is otherwise dependent on isolated microfossils. However, it has been demonstrated that disarticulated jaw elements (goniodonts) of ophiocistioids can be diagnostic at genus and species level and can be used to fill gaps in the non‐articulated fossil record. The Late Triassic Cassian Formation of Italy is regarded as one of the most important strata of early Mesozoic invertebrates worldwide. A fossil assemblage from a locality near Misurina, South Tyrol, yielded various echinoderms, including a stem group echinoid representative and a new ophiocistioid, Linguaserra triassica sp. nov. The new taxon is described and placed within linguaserrid ophiocistioids. The evolutionary history of the group is briefly discussed in relation to other taxa of the Ophiocistioidea and Echinoidea. Linguaserra triassica not only represents the first documented occurrence of this ‘Palaeozoic’ echinoderm class from Mesozoic strata but also establishes the stratigraphically youngest record of goniodonts or ophiocistioid teeth in the world.
Members of the Ophiocistioidea are small pentaradiate, free‐moving globular echinoderms, related to sea urchins and sea cucumbers (Reich & Haude 2004; Smith 2004; Smith & Reich 2013). These organisms have a large, rather depressed test, and long (skeletonized) ventral podia. The test is either plated or ‘naked’, both with distinct plates and/or ossicles (Haude & Langenstrassen 1976a, b; Haude 2004; Reich 2010). The mouth in ophiocistioids is central and downward facing with a complex jaw apparatus including goniodonts (‘serrated teeth’), in general similar to Aristotle's lantern in echinoids. Like early echinoids, ophiocistioids were presumably active predators, using their jaws to capture small benthic prey (Smith 2004).
The ophiocistioid test and skeleton is rather fragile and readily disintegrates upon death. Consequently, they have left a relatively sparse fossil record, reported from the Palaeozoic (Ordovician–Permian) only. However, given their propensity to disarticulate (e.g. Reich & Haude 2004), ophiocistioid diversity in the Palaeozoic is certainly underestimated, and thus every single record is important. Besides articulated body fossils from various Ordovician, Silurian and Devonian fossil Lagerstätten (e.g. Ubaghs 1966; Haude & Langenstrassen 1976b; Jell 1983; Haude 2004; Reich & Haude 2004; Haude & Reich 2009), there are a few published reports of questionable Ordovician (Reich 2001; Reich & Smith 2009) as well as unequivocal Silurian (Franzén 1979; Reich & Kutscher 2001, 2014), Devonian (Romanek 1984; Prokop & Petr 1987, 2002; Boczarowski 2001), Carboniferous (e.g. Maliva et al. 1983; Schraut 1992, 1993, 1995; Weber 1997; Reich & Mostler 2002; Sevastopulo 2002) and Permian (Reich 2007) disarticulated ophiocistioid material; they often represent goniodonts, which are distinct for the entire group (Haude & Langenstrassen 1976a).
The purposes of this paper are to describe the youngest ophiocistioid goniodont, to compare it with other known members of the Ophiocistioidea, and to discuss the geological history of these rare fossil echinoderms.Geological and stratigraphic setting
The new ophiocistioid goniodont described herein was collected from the Late Triassic Cassian Formation s. l. near the Misurina section, South Tyrol, northern Italy (Fig. 1). Several samples of greyish to brownish claystones were taken at a landslide scar near Misurina (46°35′41.6″ N, 12°15′34.6″ E), a locality which corresponds to outcrop number 4 of Bizzarini & Laghi (2005, figs 2, 7). Exposures at landslides are common in the marly sediments of the Cassian Formation and encompass both, synsedimentary mass flow transport and recent landslides.Figure 1 Open in figure viewerPowerPoint Geographical (and stratigraphical) position of the locality (‘Misurina Landslide’) near Misurina, South Tyrol, Italy. Subdivision of the Ladinian/Carnian stages after Ogg et al. (2016). Abbreviations: Lad., Ladinian; Longob., Longobardian; Cor., Cordevolian; Austrotrach., Austrotrachyceras; Protrach., Protrachyceras. CaptionGeographical (and stratigraphical) position of the locality (‘Misurina Landslide’) near Misurina, South Tyrol, Italy. Subdivision of the Ladinian/Carnian stages after Ogg et al. (). Abbreviations: Lad., Ladinian; Longob., Longobardian; Cor., Cordevolian; Austrotrach., Austrotrachyceras; Protrach., Protrachyceras.
The Cassian Formation and its slightly younger continuation, the Heiligkreuz Formation, represent marly sediments deposited between the vast Carnian carbonate platforms of the Dolomites. They both host more or less autochthonous basin assemblages and transported assemblages from the nearby carbonate platforms (e.g. Fürsich & Wendt 1977; Hausmann & Nützel 2015).
The present assemblage yielding the new ophiocistioid is of comparatively low diversity and is strongly dominated by gastropods. Bivalves, scaphopods and ammonites are much less common. A smaller part of the fauna contains echinoderms (disarticulated ossicles of crinoids, ophiuroids, holothuroids and echinoids). Particularly noteworthy among the recorded echinoderms are proterocidarid (stem group) echinoids (?Pronechinus). A study of the entire fauna yielding the present taxon is currently prepared for publication.
In accordance with Bizzarini & Laghi (2005), this assemblage is interpreted as essentially autochthonous, as is obvious from the low diversity and the lack or scarcity of shallow water material (e.g. algae, ooids, corals). It corresponds to the Ampullina Association sensu Fürsich & Wendt (1977), which was reported as an autochthonous basin dwelling association from the vicinity of Misurina, Costalaresc and Rudavoi. Its age is the same as that of the nearby locality ‘Misurina Skilift’: Carnian age at the Julian to Tuvalian transition (Bizzarini & Laghi 2005; Nützel et al. 2010). Our whole assemblage contains species typical of the Cassian Formation respectively the Heiligkreuz Formation, such as the gastropods Prostylifer paludinaris (Münster, 1841), Coelostylina conica (Münster, 1841) and Promathildia elongata Leonardi & Fiscon, 1959 or representatives of the bivalve genus Cassianella. Not a single species or specimen has been recovered that would indicate a mixing with non‐Carnian sedimentary material.Material and method
The marly bulk samples, collected in June 2016, were first dried, and fossils were isolated using a hydrogen peroxide solution (c. 7%). After wet‐sieving (>5.0, 0.5, 0.1 mm), the residues were dehydrated (c. 40°C), and meso‐ and microfossils were studied under a binocular microscope. Specimens were photographed using a digital microscope (Keyence VHX 5000) with stacking function first, and later mounted on a stub and coated with Au for investigation and documentation using scanning electron microscopy (Tescan Vega 2). The figured specimen is deposited at Museo di Scienze Naturali dell'Alto Adige (Museum of Nature South Tyrol) Bolzano, Italy (PZO).Systematic palaeontology
For terminology of ophiocistioid skeletal elements and overall morphology, see Haude & Langenstrassen (1976a, b) and Haude (2004). Goniodont parameters used in our study are explained in Figure 2. In addition, we used the ‘denticle index’ (DI), which describes the goniodont serration (number of denticles at flank per 1 mm) (Haude 2004). Due to the minute size of many linguaserrid goniodonts, the DI is modified here to the number of denticles at flank per 0.1 mm (in small forms) and per 1.0 mm/10 (in large forms).Figure 2 Open in figure viewerPowerPoint Schematic view of goniodont (Sollasina) measurements and parameters used in the text. B, width of goniodont/tooth; D, width of main denticle; F, length of tooth flank; H, length of main denticle; L, length of tooth lamina; N, max. length of lateral denticles; OS, length of distal (aboral) facet formed by an open stereom mesh; S, distance between tips of two consecutive goniodonts; V, frontal angle of serrated flanks; Z′, length of goniodont/tooth. CaptionSchematic view of goniodont (Sollasina) measurements and parameters used in the text. B, width of goniodont/tooth; D, width of main denticle; F, length of tooth flank; H, length of main denticle; L, length of tooth lamina; N, max. length of lateral denticles; OS, length of distal (aboral) facet formed by an open stereom mesh; S, distance between tips of two consecutive goniodonts; V, frontal angle of serrated flanks; Z′, length of goniodont/tooth.
So far monospecific, with one genus only.
Linguaserra ligula Langer, 1991 from the Devonian (Givetian) of Germany.
Early Silurian (Telychian) to Late Triassic (Carnian) of Europe (Germany, Italy, Poland, Sweden).
Reference to the type stratum.
One goniodont (holotype, PZO 11827).
Near Misurina, close to the Lago Antorno, near Cortina d'Ampezzo, South Tyrol, North Italy.
Cassian Formation; Triassic, early Carnian.
Test: unknown. Masticatory apparatus: complete jaw apparatus unknown. Goniodonts with the following characteristics: one short main denticle with 11 secondary lateral short denticles (minute teeth); abaxial face of lateral denticles ridged; the lateral sides form a frontal angle of 50/55°. The distal (aboral) part of the goniodont is slightly narrower than the rest of the goniodont, with more than 150 fine perforations of an open stereom mesh.
Z′ = 484 μm, B = 284 μm, Z′/B = 1.7, D = 28.3 μm, F = 253 μm, OS = 179 μm, H = ~23.2 μm, L = 484 μm, S = ?, N = 8.1–16.2 μm, DI = 4.4.
The new lance‐shaped goniodont is slightly convex at the outer (abaxial) face whereas the inner (adaxial) face is concave. Main denticle broken but presumably short/medium in length. Tooth aborally slightly narrower and shouldered, with stereom distally forming a facet (Fig. 3A, C), which is (adaxially) the insertion site for the ligament that bound together the goniodont elements (V‐shaped ‘battery’). Tooth batteries probably with a low number of goniodonts only. Inner side (adaxial face) grooved (see Fig. 3B, filled in part with sediment) to accommodate the ridged denticles of outer goniodont face. Serrate flanks with about 11 denticles therefore showing a denticle index of about 4.4 (Fig. 4).Figure 4 Open in figure viewerPowerPoint A plot of denticle indices (DI) against max. length/max. width (Z′/B) in goniodonts of various ophiocistioid species, including all known Linguaserra representatives. CaptionA plot of denticle indices (DI) against max. length/max. width (Z′/B) in goniodonts of various ophiocistioid species, including all known Linguaserra representatives.
The new species differs from all other Linguaserra species in having ridged lateral denticles (abaxial face) as well as in its overall morphology. Denticle index of L. triassica sp. nov. is high (4.4), like in the related L. spandeli (4.1). Denticle index of linguaserrids is always >1.5, in all other ophiocistioid genera 1.0–0.2 (Fig. 4).
The new species occurs in the Carnian Cassian Formation, South Tyrol, Italy.Linguaserrid echinoderms: ophiocistioid versus echinoid teeth
Linguaserrid echinoderms are known only from isolated tooth elements and have been reported from the Early Silurian (Telychian) through to the Late Permian (Wuchiapingian) (Langer 1991, 1997; Boczarowski 2001; Reich 2007; Reich & Kutscher 2014) of Europe. Another related tooth element, recorded from Late Ordovician strata (Katian/Hirnantian), was preliminary described and figured as ‘Linguaserra?’ by Reich (2001), ‘gen. et sp. nov.’ by Reich & Haude (2004) and later tentatively named ‘Rogeriserra’ (nomen nudum) by Reich & Smith (2009). Even if these teeth (or mouth spines) were originally included in the family Linguaserridae by Reich & Haude (2004), due to their very distinct morphology Reich (2007) and Reich & Kutscher (2014) excluded ‘Rogeriserra’ from linguaserrids.
However, the Late Triassic (Carnian) tooth element, Linguaserra triassica sp. nov., reported here, shares all the distinctive morphological characteristics of previously described linguaserrid teeth. As Linguaserra triassica occurs together in our material with ambulacral and interambulacral plates of a proterocidarid echinoid (?Pronechinus) and linguaserrids have never been found in situ associated with partial or complete skeletons, we have to discuss the biological affinities of linguaserrid (ophiocistioid) tooth elements in comparison to proterocidarid (and related) echinoid tooth elements in more detail:
Albeit different goniodont morphology in ‘true’ ophiocistioids (V‐shaped goniodonts of Eucladidae, Rhenosquamidae, Rotasacciiidae and Sollasinidae found in situ) and linguaserrid echinoderms (lance to tongue‐shaped goniodonts), all share characteristics clearly different from echinoid teeth yet known. Furthermore, based on the material (>25 000 Linguaserra goniodonts) reported and figured by Boczarowski (2001, p. 86, fig. 30F), linguaserrid teeth also show evidence of being part of a multielement battery. Therefore, there is strong evidence that linguaserrids represent a very poorly known clade of ophiocistioid echinoderms, ranging from Silurian to Triassic times.Discussion and wider significance
The geological history and diversity of ophiocistioids are summarized in Figure 5A. The earliest known records we have are isolated ‘mouth spines’ (of a questionable stem‐group member) from late Ordovician strata (Katian/Hirnantian; Reich 2001; Reich & Smith 2009) and early Silurian goniodonts (Telychian; Reich & Kutscher 2014) of Baltica.Figure 5 Open in figure viewerPowerPoint A, Palaeozoic/Mesozoic stratigraphic range of the extinct echinoderm group Ophiocistioidea; the widths of the bars are proportional to the taxic (species‐level) diversity; black bars mark species/taxa based on isolated goniodonts and the grey bars mark species/taxa based on articulated body fossils; note the gaps in Pennsylvanian and Early/Middle Triassic strata; red dots mark the occurrence of linguaserrids: 1, Linguaserra franzenae Reich & Kutscher, 2014; 2, L. ligula Langer, 1991; 3, L. sp. sensu Weber (1997); 4, L. spandeli Reich, 2007; 5, L. triassica sp. nov.; length of goniodonts is between 0.25 mm (L. sp.) and 0.82 mm (L. franzenae) (from Reich & Kutscher 2014, modified). B, the fossil record of selected echinoderm groups (classes) plotted through geological time for comparison (summarized after Sprinkle 1983; Horowitz et al. 1985; Kolata et al. 1991; Smith 2004; Sumrall 2009; Lefebvre et al. 2013; Song et al. 2013; Zamora et al. 2013); all the purely Cambrian (Cincta, Ctenocystoidea, Helicoplacoidea) and Ordovician/(Silurian) (Coronoidea, ?Ctenocystoidea, Edrioblastoidea, Parablastoidea, Paracrinoidea, Somasteroidea) echinoderm groups were excluded. Abbreviations: C., Cambrian; Cr., Cretaceous. CaptionA, Palaeozoic/Mesozoic stratigraphic range of the extinct echinoderm group Ophiocistioidea; the widths of the bars are proportional to the taxic (species‐level) diversity; black bars mark species/taxa based on isolated goniodonts and the grey bars mark species/taxa based on articulated body fossils; note the gaps in Pennsylvanian and Early/Middle Triassic strata; red dots mark the occurrence of linguaserrids: 1, Linguaserra franzenae Reich & Kutscher, ; 2, L. ligula Langer, ; 3, L. sp. sensu Weber (); 4, L. spandeli Reich, ; 5, L. triassica sp. nov.; length of goniodonts is between 0.25 mm (L. sp.) and 0.82 mm (L. franzenae) (from Reich & Kutscher , modified). B, the fossil record of selected echinoderm groups (classes) plotted through geological time for comparison (summarized after Sprinkle ; Horowitz et al. ; Kolata et al. ; Smith ; Sumrall ; Lefebvre et al. ; Song et al. ; Zamora et al. ); all the purely Cambrian (Cincta, Ctenocystoidea, Helicoplacoidea) and Ordovician/(Silurian) (Coronoidea, ?Ctenocystoidea, Edrioblastoidea, Parablastoidea, Paracrinoidea, Somasteroidea) echinoderm groups were excluded. Abbreviations: C., Cambrian; Cr., Cretaceous.
The precise morphology, especially the presence of perradial pores and a masticatory apparatus in the Middle Ordovician ?ophiocistioid Volchovia (Hecker 1938, 1940; Regnéll 1948) still remains uncertain (e.g. Haude & Langenstrassen 1976b; Reich & Haude 2004; Reich 2010). By middle Silurian/Middle Devonian times, there is evidence for the highest diversity of the group (based on the amount of taxa and articulated fossils; Reich & Haude 2004, Reich & Kutscher 2014) to suggest that ophiocistioids went into decline towards the end of the Palaeozoic, with reports of unpublished articulated material (F.H.C. Hotchkiss pers. comm. November 2008) or isolated goniodonts only (Fig. 5A). However, when the chapter on Ophiocistioidea of the Traité de Paléontologie (Ubaghs 1953) and the Treatise on Invertebrate Paleontology (Ubaghs 1966) were published, the youngest known ophiocistioid was Middle Devonian.
The discovery of several echinoderm taxa (Fig. 5B) in strata significantly younger than previously thought is a common pattern in Palaeozoic echinoderms (e.g. Kolata et al. 1991; Reich 2007; Sumrall 2009). Unnoticed by other echinoderm authors, Song et al. (2013, table S1) for instance listed ‘unnamed blastoid species’ from the earliest Triassic Hindeodus parvus Conodont Zone of the Huangzhishan section in Zhejiang Province, China. However, bulk abundance studies based on uncertain systematic data (e.g. Twitchett & Oji 2005) should be replaced by studies with a better systematic treatment of recorded taxa (see also Hunter & McNamara 2017a; Thuy et al. 2017a).
For all biota, the end‐Permian mass extinction was the most dramatic crisis in the history of life (e.g. Raup 1979; Erwin 1994; Benton & Twitchett 2003), and this mass extinction is still a subject of intense debate concerning its timing, causation and number of extinction events (e.g. Wignall & Hallam 1992; Jin et al. 2000; Shen et al. 2011; Song et al. 2013). The Permian–Triassic mass extinction affected a wide variety of groups, including echinoderms (e.g. Twitchett & Oji 2005). However, an intensive debate (Blake 2017; Hunter & McNamara 2017b; Salamon & Gorzelak 2017; Thuy 2017; Thuy et al. 2017a, b) on Palaeozoic echinoderm ‘hangovers’ or ‘holdovers’ discovered in Triassic strata (e.g. Hunter & McNamara 2017a; Thuy et al. 2017c; Hagdorn 2018; Thompson et al. 2018) started very recently.
Our discovery of Triassic ophiocistioids is significant for several reasons. Recent reports of ‘Palaeozoic’ echinoderms, at least echinoids, asteroids and ophiuroids, from Triassic sediments (e.g. Hagdorn 2012, 2018; Thuy et al. 2017c; Thompson et al. 2018) find additional confirmation. Ophiocistioids did not go extinct at the end of the Palaeozoic, as previously thought, and appear to have coexisted with crown‐group members of modern echinoderms as well as echinoid stem‐group representatives. Like stem‐group echinoids (Thompson et al. 2018), members of the Ophiocistioidea also exhibit the Lazarus effect, having disappeared from Changhsingian, Induan and Olenekian stages (latest Permian and Early Triassic). However, given that specimens are rather fragile and often mesoscopic/microscopic in size, alternatively these gaps in the fossil record are probably a result of sampling and preservational biases worldwide or are connected to neglected, overlooked or misinterpreted fossil material.
Altogether, the occurrence of ophiocistioids in the Late Triassic Cassian Formation of Italy has novel implications for understanding the Permian–Triassic fossil record of echinoderms, and shift existing paradigms that ‘Palaeozoic’ echinoderms are exclusively found in Palaeozoic strata.Acknowledgements
MR appreciates the stimulating discussions on our new material with Hans Hagdorn (Ingelfingen, Germany), Andreas Kroh (Vienna, Austria), and Jeffrey R. Thompson (Los Angeles, CA). Evelyn Kustatscher and the Museum of Nature South Tyrol, Bolzano (Naturmuseum Bozen) are kindly acknowledged for their support of this project. This study was supported in part by the German Research Foundation (DFG NU 96/13‐1, DFG KI 806/14‐1). We thank Andrew S. Gale as well as Andrew B. Smith for insightful comments that improved our paper.