Tergomyan molluscs from the Early Ordovician of the Llangynog Inlier, South Wales, UK

A low abundance but diverse fauna of tergomyan molluscs is described from the Llangynog Inlier, Carmarthenshire, South Wales. The specimens originate from a single quarry exposing shallow-water siltstone and mudstone beds of the Early Arenig (early Floian) Bolahaul Member of the Ogof Hên Formation. The fauna includes Proplina areniga sp. nov., Proplina ? obtusa sp. nov., Celtopileus calvapex gen. et sp. nov. with preserved muscle scars, Costulaconus mirificus gen. et sp. nov., and Hypseloconus ? sp. Comparable tryblidiid taxa are mainly found in the Lower Ordovician of Midwestern USA among species of Proplina Kobayashi, 1933 and Gasconadeoconus Stinchcomb, 1986. The tentative assignment of the predominantly late Cambrian Hypseloconus is symptomatic of the many uncertainties regarding the phylogeny of the tryblidiids. The diverse assemblage of organisms preserved at the single locality represents one of the most remarkable faunas of this age worldwide.


Introduction
The Llangynog Inlier (Cope 1982) is an inlier of Neoproterozoic, Cambrian, Tremadoc and Early Arenig (Tremadocian and early Floian) rocks that lies some 10 km SW of the town of Carmarthen, South Wales (Fig. 1).The stratigraphy of the Early Arenig Series was established in the area to the east of the river Tywi (Towy) by Fortey and Owens (1978) and the same succession was recognized to the west of the river by Cope (1979), in the area later to be known as the Llangynog Inlier (Cope 1982).The lower of the two Early Arenig formations in the area, the Ogof Hên Formation, comprises two members: a lower Allt Cystanog Member and an upper Bolahaul Member.The latter comprises bioturbated interbedded shales and silty mudstones and has yielded an extraordinarily diverse fauna from one small quarry lying to the SSE of Dan-lan-y-Castell farm (NGR SN 3640 1639) and it has been awarded SSSI status and is under the stewardship of the Countryside Council for Wales (Fig. 1).The quarry was described under the name of the farm by Rushton et al. (2000) but now lies in the grounds of a house named Llyn-y-Llawr.
The fauna of the quarry includes the trilobite Merlinia murchisoniae (Murchison 1839), but the absence of M. selwynii (Salter, in Murchison 1859) indicates the lower part of the M. selwynii Biozone of the Moridunian Stage of the Early Arenig Series (Fortey and Owens 1978).However, it is other fossils from this quarry which make it one of the most remarkable faunas of this age worldwide.These fossils include the most diverse bivalve fauna of Early Ordovician age yet known, consisting of 20 species belonging to 18 genera (Cope 1996).Other notable finds include the earliest parablastoids (Paul and Cope 1982) and their later recognised holdfasts (Cope and Donovan 2005); the earliest calcified red alga (Riding et al. 1998); then the oldest bryozoan known (Taylor and Cope 1987) and the earliest gorgoniids (Cope 2005).
The fauna constitutes the earliest example of the 'Modern fauna' of Sepkoski (1981) as it is dominated by molluscs which make of 71.5% of the fauna collected.In addition to the bivalves (40.7% of the total fauna), other 1 3 molluscs are abundant and the quarry has yielded a varied nautiloid fauna (25.6%) described by Evans (2005).Gastropods are also common (4.5%) but, in contrast to the bivalves, are generally poorly preserved; they and one species of tergomyan, the multi-whorled Carcassonnella cf.vizcainoi Horný and Peel 1996, have recently been described by Ebbestad and Cope (2021).The only undescribed members of the molluscan fauna are the remainder of the tergomyans, which although only constituting some 0.7% of the fauna are taxonomically diverse and are the subject of this paper.
In contrast to the abundant molluscan fauna, the brachiopod fauna, frequently the dominant phylum in rocks of this age, comprises a mere 18% of the total fauna.Of eleven taxa of brachiopods over three quarters belong to only two species, while three genera are represented by single specimens.The fauna was described by Cocks and Popov (2019).
The fossils of the quarry are usually preserved as limonite coated moulds.The gastropods and tergomyans, unlike the rest of the fauna, are often distorted and preserved as variably compressed internal moulds, possibly due to lack of a strongly calcified shell.Ornamentation is often seen on external moulds, and fine preservation of structures like muscle scars on the internal moulds suggests infilling of fine mud and clay minerals.

Systematic palaeontology
Class Tergomya Horný, 1965b(nom. transl. Peel 1991a, ex Subclass Tergomya Horný, 1965b) Remarks.Peel (1991aPeel ( , emended 1991b) defined the Class Tergomya as containing generally bilaterally symmetrical untorted univalved molluscs in which the calcareous shell is usually exogastrically planispirally coiled through about half a whorl.The shell is often cap or spoon-shaped with an anterior apex, which may vary in its location from subcentral to the anterior extremity.Muscle scars are grouped into a horseshoe-shape on the dorsal surface; the apex lies outside of this ring in some forms and may overhang the anterior shell margin; in other forms, the apex is sub-central in position.Peel (1991b) recognised three orders, reflecting three different morphological groups, the Tryblidiida Lemche 1957 typically with a pilinaform shell, the Cyrtonellida Horný, 1963 with a planispirally coiled shell, and the Hypseloconida Peel 1991b with a tall cone-shaped shell.The Tergomya is equivalent to the sub-class Tergomya of Horný (1965a, b) and is also very similar to the Tryblidiida of Wingstrand (1985); Tryblidia is broadly used interchangeably with Monoplacophora in neontological usage (e.g., Lindberg 2009;Kocot et al. 2020;Ponder et al. 2020).For further details on the development of the tergomyan concept, see Peel (1991b), Wahlman (1992), Webers et al. (1992), Geyer (1994, andYochelson andWebers (2006).
The descriptive terminology developed by Jacquet and Brock (2016) for helcionelloid molluscs is applied for the conical taxa described herein.
Order Tryblidiida Lemche 1957 Superfamily Tryblidioidea Pilsbry in Zittel-Eastman, 1899 Family Proplinidae Knight and Yochelson 1958 Genus Proplina Kobayashi 1933 Type species.By original designation of Kobayashi (1933) Remarks.Webers et al. (1992) defined members of this superfamily as having relatively flat shells in which the apex is at or slightly overhanging the anterior margin of the shell; symmetrically developed pedal retractor scars are arranged in a nearly circular or fusiform pattern that is open at the anterior end.The earlier subfamily rank of Proplininae was adjusted to family level by both Webers et al. (1992) and Wahlman (1992), the latter providing a diagnosis for the family, emphasizing the shape, overhanging apex and the six pairs of muscle scars.
The type species Proplina cornutaformis Walcott 1879, was well-described by Knight (1941), Yochelson (1958), andStinchcomb (1986) with further discussion in Yochelson and Webers (2006).It is an ovate, low patelliform taxon with the apex protruding and overhanging the anterior margin, it has six pairs of muscle scars and comarginal growth lines on the shell.Although Proplina is reported from Antarctica (Weber et al. 1992;Yochelson and Webers 2006), Korea (Kobayashi 1933), Siberia (Bjalyi 1974), and Greenland (Fortey & Peel 1990; as Archinacella(?) cf. A. elongata (Cullison)) the core distribution is mid to eastern North America during the late Cambrian and Early Ordovician (Kobayashi 1933;Yochelson 1958;Yochelson and Weber 2006;Ebbestad et al. 2013; see also Stinchcomb 1986;Stinchcomb and Angeli 2002).Because a number of tergomyan taxa from the late Cambrian-Early Ordovician are distinguished on the basis of general shell shape, which besides convergence can also vary even within a species and during ontogeny, distinguishing genera like Proplina from contemporary taxa attributed to, e.g., Archinacella or Tryblidium is fraught with difficulties.Moreover, many species have not been revised, since they were described in the 19th or early twentieth century (see discussion in Yochelson and Webers 2006).
If only Terreneuvian-Floian members are concerned, at least 25 species may be included in Proplina (see Kobayashi 1933;Cullison 1944;Lochman and Hu 1962;Bjalyi 1974;Stinchcomb 1986;Webers et al. 1992;Yochelson and Webers 2006 and references in these).Typically, the shell is low, cap-shaped, ovoid or with a more elliptical aperture.Species range in size from 10 to ~ 60 mm.The apex protrudes to various degree anteriorly, affecting the steepness and curvature of the sub-apical area and the convexity of the supra-apical surface.Besides the type species only Proplina sibelius Stinchcomb 1986 andP. arcua Stinchcomb 1986 from the Lower Ordovician of Missouri preserve muscle scars.
A number of species attributed to the genus display a more oblique cone-shaped shell with a tall and straight to weakly concave sub-apical area (e.g., Proplina suttoni Stinchcomb 1986, P. convexum (Berkey 1898), P. rutfordi Webers et al. 1992).Stinchcomb (1986) observed that the shell-shape of P. suttoni was intermediate between that of a patelliform taxon and the taller shell of Hypseloconus.He also observed taphonomic differences in the preservation of scars among specimens of the type species as described from Missouri.
Diagnosis.A species of Proplina with a low cap-shaped shell (H:L ~ 0.36) with the greatest width at mid-length, width 80% of the length.Blunt apex protrudes moderately anteriorly with angle of inclination 120 degrees.
Description.Shell low (H:L ~ 0.36), convex in lateral and transverse profiles with greatest height at mid-length.Supra-apical surface slightly convex.Apex blunt, overhangs anterior margin and protrudes moderately anteriorly.Angle of inclination 120 degrees.Sub-apical margin markedly concave and short (adapically).Transverse profile low dome-shaped with gently convex sides and rounded dorsum.Shell ovoid in outline, width 80% of length.Aperture planar Discussion.Deformation has compressed the shells dorsoventrally, so that the profile (laterally, transversally) probably originally was more vaulted and the apex more strongly recurved.The Welsh species is longer and lower than the type species, and the aperture proportionally more elongate, whereas the apex protrudes further anteriorly in the type species.Thus, the sub-apical margin is less curved in the type species, compared to that of Proplina areniga.
Proplina arcua Stinchcomb 1986 from the late Cambrian (Terreneuvian) of Missouri, USA reaches the same size as P. areniga, but the American taxon is more elongate and with the apex in large specimen projecting further anteriorly relative to the apertural margin.P. loganensis Lochman and Hu 1962 from the Tremadocian of Montana, USA has a broadly similar shell shape, but is smaller, proportionally lower and with a more ovate aperture.
Proplina?obtusa sp.nov. Figure 2g-q Etymology.From Latin obtusa, blunt.Referring to the rounded profile of the shell in lateral view.
Diagnosis.A species of Proplina with a low shell (H:L 0.45), weakly ovoid, greatest width anterior to mid-length, width 70% of length, blunt apex at or slightly protruding across anterior margin.
Description.Shell low (H:L 0.45), convex in profile with greatest height slightly anterior to mid-length.Largest specimen is 20 mm long (measured along apertural plane), 14 mm wide and 9 mm high.Supra-apical surface evenly convex posteriorly to margin and anteriorly to blunt apex that is in line with or slightly overhangs anterior margin.Angle of inclination near 90°.Sub-apical surface gently concave and tall (adapically).Shell weakly ovoid in outline, greatest width anterior to mid-length, width being 70% of length.Aperture planar without thickening.First order ornamentation consist of widely spaced low-relief comarginal rugae developed at irregular intervals.Shell otherwise seemingly smooth, thin.
Remarks.Two specimens are similar in the transverse profile (Fig. 2i, p) and seem to preserve the original convexity of the shell.Two other specimens are dorso-laterally skewed (Fig. 2m, n, j, k), which distort the sense of the shell height.This species is proportionally similar to the type species in the apertural shape and its dimensions relative to the height.The transverse profile compares to that of the specimen of Proplina cornutaformis illustrated by Yochelson (1958, Fig. 12).The Welsh species differs from the type species mainly in the position of the apex which is weakly recurved and protrudes only slightly over the anterior margin.
A tentative placement with Proplina is suggested by comparison with P. grandis Cullison 1944 from the Floian of Missouri, USA, P. sibelius Stinchcomb 1986 from the Tremadocian, also of Missouri, and P. magna (Flower 1968) (described as a species of Archinacella) from the Tremadocian of New York, USA.In all these three species, the apex only slightly overhangs the anterior margin and the sub-apical area is slightly concave and steep as in P. obtusa.All three species are much larger than P. obtusa, but whereas the proportions of P. magna are more similar to those of P. obtusa the two first species have a more ovate aperture and differ in the transverse and lateral convexity.Although poorly understood, P. niobe Billings 1865 (including its possible junior synonym P. nycteis Billings 1865, see Yochelson and Webers 2006, p. 29) from the Floian of Canada, is also comparable with P. obtusa in shape, relative proportions and development of the sub-apical area.In the Canadian species, the apex overhangs the margin, the angle of inclination is greater, and the shell is proportionally slightly higher than the Welsh species.Stinchcomb (1986) remarked that the apex in large specimens of Proplina sibelius usually did not protrude strongly as it did in smaller specimens.He also observed that the position and protrusion of the apex in P. arcua, from the Tremadocian of Missouri also varies during ontogeny, but in this species protrudes mores strongly in older specimens.
Order Kirengellida Rozov 1975 Superfamily Hypseloconoidea Knight 1952 Family Hypseloconidae Knight 1952 Remarks.The two new genera presented here, Celtopileus and Costulaconus, are tentatively placed with the Hypseloconidae.The family was erected by Knight (1952) and again as new by Knight (1956) and Stinchcomb (1986) (see Bouchet et al. 2017, p. 128), but subsequently the authorship has been referred to Knight (1956) (see, for example, Knight et al. 1960;Horný 1963;Stinchcomb 1986;Geyer 1994;Yochelson and Webers 2006).Webers et al. (1992) pointed out that with the inclusion by Rozov (1975) of Hypseloconus in the Family Romaniellidae Rozov 1975, the latter would become a junior synonym of the Hypseloconidae.
The family was placed with the Order Tryblidoidea (= Tryblidiida in modern usage) by Horný (1963), around the time when the concept of the Class Tergomya Horný, 1965b was emerging.Hypseloconids generally include narrow, medium-to-high cyrtiform (i.e., slightly curved) conical shells with the apex close to the margin-the apex itself bends away from the margin as the sub-apical surface is weakly convex and the supra-apical surface is weakly concave.Because they deviate from this morphology, Peel (1991b) suggested that Gasconadeoconus and Proteroconus of Stinchcomb (1986Stinchcomb ( , 1996) ) were not hypseloconids.As the two Welsh forms compare with Gasconadeoconus, the same reasoning may apply, although with the current lack of taxonomic agreement, it is at the moment unclear if Gasconadeoconus and the Welsh forms are to be placed among the hypseloconid or kiringellid tergomyans.Discussing the possible monoplacophoran ancestry to cephalopods, Dzik (2010) argued that the serial arrangement of in Kirengella and Hypseloconus was similar to that of the possible brachiopod-related Angarella from the Lower Ordovician of Siberia, thus excluding them from the monoplacophorans.Vendrasco (2012) argued against this suggestion and pointed out that the muscle arrangements in Kirengella and Hypseloconus were also similar to those seen in monoplachophorans.
Etymology.From the Celtic tribes and the Welsh Celtic identity, combined with the Greek Pileus meaning cap.The gender is masculine.
Diagnosis.Large tergomyan with a pointed patelliform shell; apex of cone sub-central.Internal mould with seven pairs of shell muscles in adult stage, arranged in horseshoe-shape.Shell thick and ribbed, except in apical region, which is smooth.
Remarks.This genus and species cannot be readily related to hitherto described tergomyans.In the holotype both the sub-apical and supra-apical surfaces of the sub-central apex are distorted, but with the open-ended part of the horseshoe-shaped muscle scars indicating the anterior.Specimens NMW 2017.15G.12 (Fig. 3i-k) and NMW 2017.15G.13 (Fig. 3l, m) retain better the original shape.They show that the genus is not laterally compressed as in large hypseloconid genera-like Gasconadeoconus Stinchcomb 1986 from the Tremadocian of Missouri, USA, and that it attains a much more patelliform shell shape with an elliptical apertural outline that does not taper anteriorly, a sub-central apex and slightly convex sub-apical surface and a straight supraapical and lateral surfaces.Several specimens show the thick and ribbed shell with a smooth apical region (Fig. 3).
Celtopileus calvapex sp.nov. Figure 3a-n Etymology.From Latin calva, bald, and Latin apex, top, pertaining to the smooth apex.

Diagnosis. As for genus.
Description.The holotype is somewhat distorted, with the slight compression of the shell increasing the length of the mould, reducing its width and probably increasing its height.The distortion is not symmetrical so that there has been a slight twisting of the mould.The length is 37.8 mm, the width 24.0 mm and the height 26.0 mm.The apex of the cone is slightly excentrally placed and there is a horseshoeshaped ring consisting of seven pairs of muscle scars around it.Paratype NMW 2017.15G.14 also shows these muscle scars, but not as clearly (Fig. 3h).The apex lies in line with the two pairs of the muscle scars at the open end of the horseshoe-shape.The scars become slightly smaller away from the open end.The ventral ends of the muscle scars are digitate in the holotype, suggesting the possible attachment of the ctenidia.Slight abrasion of one side of the apex occurred during collection of the holotype.Paratype NMW 2017.15G.13 is an external mould, low to moderate in height, from which a latex cast has been prepared (Fig. 3ik).The external features thus revealed show that the area around the apex of the shell is essentially smooth and it was not until a height of some 7 mm had been attained that the shell developed ornament.The ornament consists of radial ribs that become stronger and more widely spaced towards the shell margins.There is no increase in rib numbers as growth proceeded.The ribs are also seen on the internal moulds, showing that they are corrugations of the shell.
There are some five well-marked growth increments on this paratype, forming irregular comarginal rugae, some of which correspond to slight changes in the angle of the cone.One area of the cast shows closely spaced growth increments.Paratype NMW 2017.15G.15 (Fig. 3n) is a very small example of the species.It is 13.3 mm in length and appears to show only five pairs of muscle scars.The apex is smooth and sub-central, but seems to compare with the apex of this species.
Remarks.The broad, low-to-moderately high shell of Celtopileus with its sub-central apex and distinct ornamentation differs markedly from other large (> 20 mm in length) late Cambrian-Early Ordovician tergomyans.A comparable genus is Gasconadeoconus Stinchcomb 1986 from the Lower Ordovician Gasconade Formation of Missouri, USA.This genus differs from Celtopileus in having a more laterally compressed and proportionally taller and more erect shell with the apex closer to the margin of the shell.The ornamentation in G. ponderosa Stinchcomb Scenella sp.undet. of Rasetti (1954;see also Peel 1991b, p. 56 andDzik 2010, pp. 106, 108) from the early Wuliuan of Canada is quite similar to Celtopileus.This is a patelliform taxon with a lower shell (H:L ~ 0.3) than that of Celtopileus, an elliptical apertural outline and a subcentral apex, around which six pairs of muscle scars are present in a semi-horseshoe-shape. Just as in Celtopileus the apex lies in line with the larger two pairs in the more open-ended part of the series, in what Rasetti (1954) considered an anterior position.Peel (1991b) interpreted this as a helcionelloid mollusc rather than a tergomyan and, in line with the morphology of this group, argued that the shell was endogastrically coiled.The anterior interpreted by Rasetti (1954) was, therefore, considered to be the posterior by Peel and thus the absence of scars in this position taken to indicate the location of the mantle cavity.However, if the presence of a circum-pedal pallial cavity with the ctenidia in the pallial groove is inferred in Celtopileus, arguments whether a mantle cavity was anterior or posterior becomes void.Instead, by comparison with proplinids, the apex in such more conical tergomyans would be towards the anterior, and likewise, the open section of the muscle scars would tacitly indicate the anterior.Yochelson and Webers (2006, pp. 22, 23) acknowledged the arguments concerning an anterior-posterior orientation but avoided making such a distinction themselves, noting that in, e.g., Hypseloconus, "there is no obvious indication of orientation" (Yochelson and Webers 2006, p. 22).
This appears to be the first report of a tergomyan in which a juvenile form has been described showing fewer muscle scars than the adult.Monoplacophorans appear basal to the conchiferans (Kocot et al. 2020), and the last common ancestor to extant molluscs was presumed to have had seven or eight dorso-ventral muscles pairs (see also Wingstrand 1985;Wanninger and Wollesen 2019).In molluscan evolutionary history, a reduction of the dorso-ventral musculature is evident (Haszprunar and Wanninger 2000), and a possible increase from five to seven pairs in Celtopileus seems to be at odds with this.However, late Cambrian and Lower Ordovician kirengellid and hypseloconid tergomyans display a greater disparity in the dorso-ventral musculature which varies between five and possibly eight pairs (i.e., in the Ordovician and Silurian Archaeophiala Perner 1903; Knight 1952) (see Yochelson 1958;Knight and Yochelson 1960;Rozov 1975;Stinchcomb 1986;Yochelson and Webers 2006).A reduction of muscle pairs in some lineages may have taken place already early in the evolutionary history, or alternatively the ancestral state of this character involved fewer pairs than the eight seen in recent forms.Modifications of these structures may also have occurred many times and in different lineages, and the living monoplacophorans may represent the only surviving lineage (see discussion in Lindberg 2009, pp. 198, 199).However, the observation in Celtopileus is not conclusive as only one example is known and preservational variability, such as observed by Stinchcomb (1986) concerning muscle scars in Proplina cornutaformis, must be taken into account.The single specimen could potentially represent another taxon, but it seems to compare with the other material of this species, and is, therefore, included in the type series.
Etymology.From the Latin costula, small rib and conus, cone.The gender is masculine.
Diagnosis.Acute conical tergomyan with apex anterior of centre and attaining large size.Ornament of large numbers of fine radial lirae.
Remarks.The large size, conical shell and radial ornamentation readily distinguish Costulaconus from Kirengella or Hypseloconus.For comparison with other genera, see discussion under the species.
Diagnosis.As for genus.
Description.The holotype is the largest example of the species; it is an internal mould with associated external mould.The preserved length is 43.6 mm but the anterior extremity of the shell is missing and it is estimated that the total length would have been around 48 mm.The maximum width (measured on the partial external mould of the holotype) is 35.0 mm at the posterior 1/3 of the shell.The height is 30.3 mm (H:L = 0.63, moderate shell height).Shell outline ovoid, with a narrower width of the supposed anterior end.Apex erect and blunt, placed about 1/3 away from the anterior margin.Apical, sub-apical and lateral surfaces appear straight.The external shell has abundant fine lirae that can be traced back very near to the apex of the shell; this is seen especially in paratype NMW 2017.15G.18 (Fig. 4h, j), which is presumed to be a juvenile example.The lirae are also seen on parts of the internal moulds though the detail is here subdued, showing that the shell ornament is primarily external.From the external moulds it is seen that the lirae have a characteristic form; in between each pair of lirae is a smaller lira with about half the elevation of the principal ones.This feature is illustrated in Fig. 4e, j and can be seen on all external moulds, but these finer lirae are not visible on any of the internal moulds.There are growth parallel grooves on the internal moulds that appear to represent marked growth annulations; they are coincident with post-mortem distortion of the shell and may reflect sites of thinner shell after thicker growth annulations.There is no trace of muscle scars on any specimen.
Remarks.Besides the lack of apparent muscle scars in the available material of this species the more ovoid apertural shape, the higher shell and distinct ornamentation renders it readily distinguishable from Celtopileus.An ornament consisting of a fine lira between pairs of coarser lirae is not observed in comparable hypseloconids.The small (L = 13 mm), tall (H:L = 0.96) and conical shell of Kornoutella bohemica (Horný, 1963) from the Lower Ordovician Sárka Formation, Czech Republic has fine radial ornamentation, but this differs in being discontinuous as it is interrupted by irregular comarginal intervals of growth (Horný 1963(Horný , 1997)).In the large Gasconadeoconus ponderosa Stinchcomb 1986 from the Lower Ordovician of Missouri the ornamentation is similar in that it consists of fine radiating lirae and from the published image interlirae may be present.Furthermore, the shell of this species is similar to Celtopileus in size and overall shape but is proportionally taller (H:L close to 1), more laterally compressed and with the apex closer to the margin.
Description.Shell conical, slightly higher than wide (height = 11.5 mm), apex incompletely preserved but overhanging margin, with sub-apical surface markedly recurved just below apex while becoming more straight abapically.Supra-apical surface evenly convex.Ornamentation consists of fine comarginal growth lines.Shape of aperture unknown but apertural margin slightly curved in lateral aspect.
Remarks.The two available specimens are completely flattened but preserve the conical shape and ornamentation.The placement within the largely late Cambrian Hypseloconus is tentative, as the shape of the shells differ.The Welsh form resembles the specimen described as H. recurves by Stinchcomb (1986, Fig. 4.10) from the Lower Ordovician of Missouri, which Yochelson and Webers (2006) excluded from that species.They further remarked that Lower Ordovician occurrences of Hypseloconus probably instead represent a closely allied genus (Yochelson and Webers 2006, p. 43).The small genus Proteroconus (Stinchcomb 1986), emended by Stinchcomb (1996), bears some resemblance to the Welsh specimens in the curvature of the apex, although it is much more strongly recurved than in Proteroconus.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material.If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.To view a copy of this licence, visit http:// creat iveco mmons.org/ licen ses/ by/4.0/.

Fig. 1
Fig. 1 Stratigraphy and maps.a Moridunian stratigraphy (after Fortey and Owens 1978).b simplified map of Wales.c SE part of Wales.d the area south of Carmarthen showing the location of the study area