Introduction

Pachycormiformes is a diverse clade of Mesozoic (Early Jurassic – Late Cretaceous) stem-teleost actinopterygians which are well known for their morphological and morphometric disparity. Certain derived members of the subfamily Asthenocorminae Cooper et al., 2022 achieved gigantic body sizes (≤ 16 m TL), were edentulous, and evolved convergent ‘suspension-feeding’ ecologies in a parallel trajectory to extant whale sharks (Rhincodon), basking sharks (Cetorhinus), and certain cetaceans (Cooper et al., 2022; Dobson et al., 2019; Friedman, 2011; Friedman et al., 2010; Liston, 2007, 2008, 2010; Schumacher et al., 2016). The largest of these asthenocormines, Leedsichthys problematicus Woodward, 1889, from the Callovian – Oxfordian of Europe and South America is considered to be the largest and heaviest bony fish to have ever lived (Martill et al., 1999; Liston, 2007, 2008, 2010, 2013; Gouiric-Cavalli et al., 2019; Johanson et al. 2022). Pachycormiformes is considered by most authors to belong to the stem-group teleosts within Teleosteomorpha Arratia, 1999, but outside of crown-group Teleostei (e.g., Cooper et al., 2022; Gouiric-Cavalli & Arratia, 2022; Maxwell et al., 2020). However, the placement of Pachycormiformes within Teleosteomorpha changes depending on the outgroup when tested with a phylogenetic analysis, meaning that Pachycormiformes likely holds an important (albeit unresolved) position in the poorly understood Neopterygii – Teleostei transition (Arratia, 1999, 2001; Arratia & Schultze, 2013; Cooper, 2023; Cooper & Maxwell, 2022; Cooper et al., 2022; Gouiric-Cavalli & Arratia, 2022). Pachycormiformes is therefore an important and novel clade to test many major evolutionary hypotheses concerning the origins, interrelationships, and ecological radiation of early teleosts during the Mesozoic.

The Upper Jurassic Kimmeridge Clay Formation of Dorset has yielded a high number of pachycormiform remains ranging from isolated teeth to near complete and well-articulated skeletons. However, the vast majority of these materials in museum collections remain unidentified and undescribed. Pachycormiforms have only briefly been described from the formation at the generic level (Hypsocormus sp., Orthocormus sp., Leedsichthys sp., ?Asthenocormus sp.) (Martill & Brito, 2020; Woodward, 1895) and therefore the true taxonomic diversity and trophic disparity of the clade is likely severely underestimated in this important marine palaeofauna. Recent examination of the fossil fishes in the Etches Collection (Kimmeridge Bay, Dorset) has revealed a plethora of important new taxonomic and palaeobiogeographic data on the actinopterygian fauna in the Kimmeridge Clay Formation. One such observation is the new occurrence of a large-bodied asthenocormine pachycormiform, ascribed here to Asthenocormus cf. titanius (Wagner, 1863), the description of which is the subject of the present paper. The specimen in question, an incomplete caudal fin with associated elements of the caudal endoskeleton, was previously described by Martill and Brito (2020) as an unpaired fin of the supposedly ‘oldest true acipenseriform sturgeon’ (Acipenseridae). The authors drew this conclusion based on two large scutes associated with the fin, the morphology of which being somewhat reminiscent to those in extant acipenseriform sturgeons. Indeed, this is not the first time pachycormiform remains were misidentified as acipenseroid, with Woodward (1889) initially interpreting the first remains of Leedsichthys from the Oxford Clay Formation as belonging to a gigantic sturgeon. The interpretations by these recent authors had novel and revolutionary implications for the origins of Acipenseridae, as previously true sturgeons were only known from the Cretaceous onwards (Martill, 2023), with the off-shore setting of the Kimmeridge Clay Formation implying a possible marine origin for a clade which is today almost exclusively freshwater (Grande & Bemis, 1991, 1996; Hilton & Grande, 2023; Martill, 2023). Reinterpretation of the specimen as a pachycormiform therefore has broader evolutionary implications.

At present, three species of Middle-Upper Jurassic asthenocormines are described in the literature: Martillichthys renwickae Liston, 2008 (Callovian-Oxfordian, Peterborough), Leedsichthys problematicus Woodward, 1889 (Callovian—?Kimmeridgian, England, France, Germany, Chile), and Asthenocormus titanius (Wagner, 1863) (Kimmeridgian – Tithonian, Germany). The caudal fins of these taxa are highly derived and all show reduced ossification in the caudal endoskeleton. The caudal fin is well known from several examples in Asthenocormus titanius, but is poorly know in Leedsichthys problematicus whereby only the fin rays have been described based predominantly on a single large incomplete caudal fin excavated by Alfred Leeds in the 19th Century (Liston & Noè, 2004; Liston, 2008; Johanson et al. 2022). Currently nothing is known of the caudal endoskeleton including the pre-caudal scutes or hypural plate(s) in Leedsichthys. The caudal fin of Martillichthys renwickae (NHMUK PV P.61563) is much more complete, including the dorsal pre-caudal scute, and was recently re-described in Cooper et al., (2022: supplement). Incompleteness of Leedsichthys and Martillichthys predominantly due to their weakly ossified skeletons and large sizes is a major challenge for comparative anatomy and phylogeny in pachycormiforms. The comparatively excellent preservation of Asthenocormus titanius from Bavaria therefore offers something of a ‘Rosetta Stone’ to decipher and reconstruct the anatomy of other poorly known asthenocormines (Dobson et al., 2019).

Asthenocormus titanius (Wagner, 1863) is a large (2–3 m TL) edentulous asthenocormine pachycormiform, characterised by an elongate and narrow skull, an absence of ossified suborbital and infraorbital bones, premaxilla and dermosphenotic (ossified in basal asthenocormines), and an elongate body without pelvic fins or mineralised scales (Lambers, 1992). The caudal fin of this species is diagnostic and highly derived in contrast to more basal pachycormiforms (e.g. Euthynotus spp.) (Arratia, 2008; Arratia & Lambers, 1996; Arratia & Schultze, 2013; Wenz, 1967). The caudal fin is deeply forked, with a high aspect ratio and each lobe formed of a roughly equal number of basal fulcra and principal lepidotrichia, whereas fringing fulcra are absent on all fins. The paired pre-caudal scutes in this taxon are highly diagnostic, being massive and plate-like; characterized by wide lateral processes and a prominent anterior notch (Cooper et al., 2022; Gouiric-Cavalli et al., 2019; Lambers, 1992).

The type and only species, Asthenocormus titanius, is an iconic taxon of the famous Upper Jurassic (Kimmeridgian – Tithonian) Solnhofen-type lithographic plattenkalk limestones of southern Germany, with specimens recorded from the lagoonal basins of Solnhofen, Eichstätt, Mörnsheim, Blumenberg, Langenaltheim, and Painten (Bavaria) (Frickhinger, 1994; Lambers, 1992; Mainwaring, 1978; Quenstedt, 1852; Vetter, 1881; Wagner, 1863; this paper). Isolated fragments from enormous individuals indicate that Asthenocormus titanius is likely the largest vertebrate in the assemblage (see discussion). Asthenocormus has a complicated taxonomic history. Initially, Quenstedt (1852) first described fragments of this fish as Caturus sp. while Wagner (1863) described the holotype specimen as Eugnathus titanius. Later, Vetter (1881) described a large, complete example of this species, naming it Agassizia titania and documented soft tissue preservation and small prey fishes inside of the gut. Woodward (1895) recognised that the genus name Agassizia was preoccupied by an echinoderm and proposed the replacement name Asthenocormus. Eastman (1914) described a second species, ‘Asthenocormus’ retrodorsalis, however the species was later reassigned to a new genus, Pseudoasthenocormus retrodorsalis (Eastman, 1914) by Lambers (1992). Ps. retrodorsalis belongs to the pachycormiform subfamily Hypsocorminae based on the presence of pelvic fins in this species: note that the absence of pelvic fins is an apomorphy of Asthenocorminae (Cooper, 2023; Cooper et al., 2022). Unfortunately, most of the historic specimens of Asthenocormus, including the holotype, were destroyed during WWII. A specimen at the JM-E was later selected as the neotype, and 5 surviving referred specimens were also identified (Lambers, 1992). A newly prepared example of this fish (DMA-AS-1) from the Painten plattenkalk (late Kimmeridgian) housed in the Dinosaurier Museum Altmühltal has since been uncovered and is exceptionally preserved, with further additional specimens since identified in museum collections (see results below).

Remains of Asthenocormus titanius and Asthenocormus sp. have previously been recorded as being present in the Middle and Upper Jurassic of England (Lambers, 1992; Martill, 1991; Martill & Brito, 2020; Schaeffer & Patterson, 1984). However, a review of these records concludes that in all instances the materials in question had been misattributed. Specimens previously recorded as Asthenocormus sp. from the Oxford Clay Formation (Martill, 1991) were transferred to Martillichthys renwickae Liston, 2008 or Asthenocorminae gen. et sp. indet. (Cooper et al., 2022; Dobson et al., 2019; Liston, 2008). A supposedly ‘juvenile’ specimen of ?Asthenocormus sp. was recently described from the Kimmeridge Clay Formation of Dorset (Martill & Brito, 2020). Our re-examination of this specimen in the Etches Collection (see results) reveals the presence of pelvic fins and pectoral fringing fulcra: both shared characters of Hypsocorminae, but absent in Asthenocorminae (Cooper, 2023; Cooper et al., 2022). We therefore conclude that all previous records of Asthenocormus in the UK are invalid, with the new specimen from the Etches Collection described herein representing the first true occurrence of this genus outside of the South German plattenkalks.

Geological setting

The Kimmeridge Clay Formation of England’s Jurassic Coast crops out along the southern coastline of Dorset and is extensively well exposed in the cliffs and foreshore between St. Alban’s Head and Brady Bay (Fig. 1), with further exposures in Ringstead Bay, Osmington Mills and Portland Bill (Weymouth). The region holds significant cultural and scientific importance for its rich heritage of fossil collecting and palaeontological research. Of the numerous well-known fossil hunters to scour the Dorset foreshores over the last two centuries, few compare in regards to dedication and expertise to Dr Steve Etches (Museum of Jurassic Marine Life; The Etches Collection) who has meticulously collected and prepared fossils from the Kimmeridge Clay Formation of Dorset for over half a century. Dr Etches has unearthed an impressive, world-renowned collection of Kimmeridgian-Tithonian aged fossils, many of which are new to science. His collection has recently been made available for study and forms the primary source of data for this paper. Although often overshadowed by the better-studied marine and terrestrial reptiles, the actinopterygian fishes of the Kimmeridge Clay Formation comprise the vast majority of the vertebrate collection and are among some of the most scientifically informative specimens in the Etches Collection.

Fig. 1
figure 1

Regional map of the Isle of Purbeck (Dorset) showing outcrops of the Kimmeridge Clay Formation between Brady Bay and St. Alban’s Head. The collection horizon of MJML K1964 at Freshwater Steps is indicated with a Star. Scale bar equals 1 km

Full size image

The Upper Jurassic Kimmeridge Clay Formation of Dorset is regarded as a marine Konservat-Lagerstätte and has become well-known for its exceptionally preserved vertebrate and invertebrate faunas (Martill & Etches, 2021). Vertebrate fossils from this formation are often articulated and may preserve soft tissue, stomach contents, and display features of taphonomic interest such as predation traces (e.g. Jacobs & Martill, 2020). The section of the Kimmeridge Clay Formation exposed in Dorset comprises a rhythmic sequence of soft bituminous mudstones, calcareous mudstones, coccolithophore-dominated limestones, and black kerogen-rich laminated oil shales (Etches et al., 2009; Martill & Etches, 2021). The age of the formation ranges from the Pictonia baylei ammonite Zone of the Kimmeridgian, to the Virgatopavlovia fittoni ammonite Zone of the early Tithonian (Gallois, 2020) (Fig. 2). It is thus approximately coeval with the Bavarian plattenkalks of Painten (late Kimmeridgian), Solnhofen, Eichstätt and Langenaltheim (early Tithonian: Schweigert, 2007, 2015) and the Nusplingen plattenkalk of Baden-Württemberg (late Kimmeridgian: Dietl & Schweigert, 2001; Schweigert, 2007).

Fig. 2
figure 2

Stratigraphic log of the Kimmeridge Clay Formation of Dorset with the collection horizons of MJML K1964 (Asthenocormus cf. titanius) and MJML 1556 (Hypsocorminae gen. et sp. indeterminate) indicated. Lithologies, biozones and unit names are based on Gallois (2020)

Full size image

Actinopterygian fishes contribute the most diverse and morphologically disparate group of vertebrate fossils in the Kimmeridge Clay Formation, although the vast majority of these species remain undescribed, poorly diagnosed, misattributed, taxonomically ambiguous, or are new to science. Various actinopterygian groups are present in the formation including ginglymodians (Lepidotes sp.), several undescribed ganoid halecomorphs, caturids (Caturus spp., Strobilodus sp.), pycnodontiforms (Gyrodus spp.), stem-teleosts (aspidorhynchids, pachycormiforms, ichthyodectiforms), and various crown-group teleosts (Teleostei) (Martill & Brito, 2020; Woodward, 1895). Here, we reappraise certain pachycormiform and supposed acipenserid remains in the Etches Collection, and report on the first true occurrence of Asthenocormus cf. titanius from the UK.

Materials and methods

A unique specimen from The Etches Collection in the Museum of Jurassic Marine Life (MJML) in Kimmeridge Bay, Dorset, is reappraised here as belonging to the large ‘suspension-feeding’ pachycormiform fish Asthenocormus cf. titanius (Fig. 3). MJML K1964 is an incomplete epaxial (upper) caudal fin lobe with associated remains of the caudal endoskeleton and the paired dorsal and ventral pre-caudal scutes, the morphology of which is highly diagnostic of A. titanius (see below). Martill and Brito (2020) previously misattributed this specimen to Acipenseriformes, proposing it to be the oldest ‘true sturgeon’ (Acipenseridae) thereby suggesting a Late Jurassic origin for Acipenseroidei (Polyodontidae + Acipenseridae). The specimen derives from the hudlestoni Zone (early Tithonian) of the Kimmeridge Clay Formation and was collected from a cliff-fall by Dr Steve Etches at Freshwater Steps (approximately 3 km East of Kimmeridge Bay) (Figs. 1, 2). The slab containing the bones was carefully prepared from the underside by Dr Etches using a soft air-abrasive powder.

Fig. 3
figure 3

Asthenocormus cf. titanius, MJML K1964 (Etches Collection), incomplete caudal fin preserving the paired pre-caudal scutes and remnants of the caudal endoskeleton. A Overview of specimen. B Detail of lepidotrichia bifurcation (loaction indicated with and arrow in A). Arrows indicate zones of bifurcation. Scale bars equal 50 mm (A) and 20 mm (B)

Full size image

Additionally, we also reappraise an allegedly juvenile specimen of ?Asthenocormus sp. identified by Martill and Brito (2020) based on an incomplete “toothless” fish collected by Steve Etches (The Etches Collection, Museum of Jurassic Marine Life, Kimmeridge, Dorset) from the elegans Zone (earliest Tithonian) of the Kimmeridge Clay Formation at Hen Cliff on the eastern side of Kimmeridge Bay from the wavecut platform. Due to foreshore erosion, most of the bones were present only as impressions, and so the entire specimen was cast in plaster.

Anatomical nomenclature used here follows conventional homologies (e.g., Schultze, 2008), with terminologies for the pachycormiform caudal fin and fin ray/fulcra classification/patterns based on Patterson (1977), Mainwaring (1978), Arratia and Lambers (1996), Arratia (2008), Arratia and Schultze (2013), and Gouiric-Cavalli and Arratia (2022). The scute element preceding the basal fulcra in the caudal fins of pachycormiforms has previously been referred to as the ‘dorsal scute’ (Arratia, 2008; Arratia & Lambers, 1996; Arratia & Schultze, 2013; Cooper & Maxwell, 2022; Cooper et al., 2022; Gouiric-Cavalli & Arratia, 2022), however we find this terminology to be inappropriate as some pachycormiforms possess a separate dorsal and ventral scute associated with the caudal fin (see comparative anatomy). We propose a new term to encompass both the dorsal and ventral scutes: the ‘pre-caudal scutes’, including both the ‘pre-caudal dorsal scute’, and the ‘pre-caudal ventral scute’. Terminology for morphological landmarks in the pre-caudal (dorsal + ventral) scute follows Cooper et al., (2022: Fig. 7).


Institutional abbreviations: BPSG = Bayerische Staatssammlung für Paläontologie und Geologie, Munich, Germany; DMA = Dinosaurier Museum Altmühltal, Bavaria, Germany; JM-E = Juramuseum Eichstätt, Germany; L = Helmet Leich Collection, Bochum, Germany (After Lambers, 1992); LF = Lauer Foundation for Paleontology, USA; MJML = Museum of Jurassic Marine Life, The Etches Collection, Kimmeridge Bay, Dorset, UK; NHMUK = The Natural History Museum, London, UK; SNSD = Senckenberg Naturhistorisches Sammlungen Dresden, Germany.

Results

Valid material of Asthenocormus in the Kimmeridge Clay Formation

Systematic palaeontology

Infraclass Actinopterygii Cope, 1887


Supergroup Teleosteomorpha Arratia, 2001


Group Pachycormiformes Berg, 1937


Family Pachycormidae Woodward, 1895


Subfamily Asthenocorminae Cooper et al., 2022


Genus Asthenocormus Woodward, 1895


Type species. Asthenocormus titanius (Wagner, 1863)


Important diagnostic features. Asthenocormus titanius is a large streamlined asthenocormine characterised by an absence of teeth on the external jaws, loss of a mineralized premaxilla and dermosphenotic, elongate and shallow skull and body, scales absent, dorsal fin in advance of anal fin, pelvic fins absent, pre-caudal scutes massive with plate-like lateral processes, spiral valve greatly elongate with roughly 70 mucosal rotations.


Asthenocormus cf. titanius

Figures 3, 4

Fig. 4
figure 4

Asthenocormus cf. titanius, MJML K1964, A details of the caudal endoskeleton and pre-caudal scutes. B Line drawing. Abbreviations: ano.scu, anterior notch of the pre-caudal scute; ebfu, epaxial basal fulcra; ‘eu’ 1–4, ‘epural’-like element 1–4; hbfu, hypaxial basal fulcra; ltpr.scu, lateral process of the pre-caudal scute; scu.d, dorsal pre-caudal scute; scu.v, ventral pre-dorsal scute. Scale bar equals 20 mm

Full size image

1852 Caturus sp.; Quenstedt, p. 216.


1863 Eugnathus titanius Wagner, p. 670.


1881 Agassizia titania Vetter, p. 97; pl. 3.


1887Agassizia (Eugnathus) titania Wagner; Zittel, p. 226.


1893Agassizia titania Vetter; Reis, p. 518.


1895 Asthenocormus titanius (Wagner); Woodward, p. 380.


1914 Asthenocormus titanius (Wagner); Eastman, p. 416.


1919Agassizia titania Wagner; Neumayer, p. 10; pl.2, fig. 3.


1961Eugnathus titanius Wagner; Kuhn, p. 35 (name only).


1961Asthenocormus titanius (Wagner); Kuhn, p. 36 (name only).


1961Agassizia titania Wagner; Kuhn, p. 37 (name only)


1987Asthenocormus titanius (Wagner, 1863); Viohl, p. 40, 60; fig. 4.


1992 Asthenocormus titanius (Wagner, 1863); Lambers, p. 207; pl.1–4; figs 1–4.


1994 Asthenocormus titanius (Wagner) 1863; Frickhinger, p. 200, fig. 425.


1994 unnamed bony ganoid; Frickhinger, p. 220, fig. 468A.


2008 Asthenocormus titanius (Wagner, 1863); Liston, p. 182, fig. 5.


2013 Asthenocormus titanius (Wagner, 1863); Arratia and Schultze; p. 89, 104; fig. 14.


2020 Asthenocormus titanius (Wagner, 1863); Maxwell et al., table 1 (name only).


2020 undescribed Acipenseridae; Martill and Brito, p. 36, fig. 3.2.


2022 Asthenocormus titanius (Wagner, 1863); Cooper et al., p. 5–37, fig. 7j.


2023 Asthenocormus titanius (Wagner, 1863); Cooper, p. 20–27, 254, 266; figs. 1, 2.


Material. Incomplete caudal fin preserving predominantly the epaxial lobe with associated remains of the caudal endoskeleton and paired pre-caudal scutes (MJML K1964), preserved on a triangular slab of compact mudstone (Figs. 3, 4).


Locality and Horizon. hudlestoni Zone, Kimmeridge Clay Formation, early Tithonian, Upper Jurassic; coastal exposure west of the White Stone Band at Freshwater Steps, Dorset (Steve Etches, pers. comms.).


Description. An articulated, albeit incomplete caudal fin preserving the epaxial basal fulcra, the epaxial principal lepidotrichia, fragments of the hypaxial basal fulcra, four ‘epurals’ from the caudal endoskeleton, and the paired dorsal and ventral pre-caudal scutes associated with the caudal peduncle. The fin has been prepared from the underside within its original matrix (260 × 170 mm), thereby retaining the original association of the bones (Fig. 3).


Caudal fin. The epaxial lobe is broken distally, hence the boundary separating the epaxial basal fulcra and the epaxial principal lepidotrichia (and therefore the exact number of each) cannot be determined. The lobe has a width of approximately 120 mm and comprises 44 elements, at least 20 of which are basal fulcra. Complete epaxial fulcra are narrow, very slightly anteroventrally compressed, tall, lack any ornamentation, and each terminates at a thin lanceolate distal point. Anterior-most basal fulcra are noticeably shorter and are more widely bifurcated at their bases, compared to the taller distal basal fulcra in which the bifurcated base becomes much shorter and narrower. The first 13 basal epaxial fulcra articulate dorsally with the ‘epurals’ (Fig. 4). The proximal ends of 12 rays from the hypaxial lobe (lower lobe) are displaced anteroventral to the ‘epurals’, and based on their similar morphologies likely correspond the hypaxial basal fulcra. The proximal ends of both the epaxial and hypaxial basal fulcra are weakly striated with slightly scalloped ventral margins (Fig. 4).

The epaxial lepidotrichia are entirely unsegmented and only distally bifurcate free from joints as per the Pachycormiformes condition (Liston et al., 2019; Mainwaring, 1978; Wenz, 1967). Morphologically, each lepidotrichium is robust, stiff, sub-oval in cross section, and weakly striated on the external surface. Completely preserved lepidotrichia close to the posterior midpoint of the fin frequently bifurcate, forming a wide elaborate fan (Fig. 3B). Complete rays in this region bifurcate a minimum of 6 times, with the first bifurcation occurring close to the proximal origin of the bone. In more anteriorly placed lepidotrichia, the first bifurcation does not appear until close to the distal extremity as observed in complete examples of Asthenocormus titanius from the Upper Jurassic of Bavaria (Fig. 5). Similar to MJML K1964, all of the epaxial lepidotrichia (principal rays) are situated posterior to the ‘epurals’ in Asthenocormus titanius.

Fig. 5
figure 5

Asthenocormus titanius, complete example from the Kimmeridgian Painten Plattenkalk, Bavaria. Specimen in the Albersdörfer Collection, Dinosaurier Museum Altmühltal (DMA-AS-1). A Complete articulated skeleton in left lateral view measuring 1700 mm TL. B caudal fin and caudal peduncle, an arrow indicates the location of the articulated pre-caudal scutes. C Photo of the articulated dorsal and ventral pre-caudal scutes showing their anatomical arrangement at the bases of the paired basal (epaxial + hypaxial) fulcra. Abbreviations: anfn, anal fin; cufn, caudal fin; ‘eu’1–4, ‘epural’-like element 1–4; hapu, haemal arches of preural vertebrae; hp, hypural plate; nupu, neural arches of the preural vertebrae; pcfn.l, left pectoral fin; pcfn.r, right pectoral fin; scu.d, dorsal pre-caudal scute; scu.v, ventral pre-caudal scute. Scale bars equal 100 mm (A), 50 mm (B) and 20 mm (C)

Full size image

The caudal endoskeleton is weakly ossified in derived asthenocormines, with MJML K1964 only preserving the articulated ‘epurals’, but not the ‘uroneurals’ or the hypural plate, which are well developed and highly modified in Asthenocormus titanius (Lambers, 1992) and other Upper Jurassic asthenocormines (Cooper & Maxwell, 2022; Liston, 2008). The ‘epurals’ (see Arratia & Lambers (1996) and Arratia (2008) for issues regarding homology) are short and robust; each element is slightly taller than its predecessor. The first ‘epural’ is amorphously rectangular whilst ‘epurals’ 2–4 are more club-like with a slightly concave anterior waist, gently widened convex dorsal margins, and robust angular ventral margins. The number and morphology of these elements is consistent with the ‘epurals’ in referred Asthenocormus titanius specimens from southern Germany (Arratia & Schultze, 2013; Lambers, 1992; pers. oberv. SC). The ‘epurals’ in both MJML K1964 and Asthenocormus titanius only articulate distally with the anterior-most epaxial basal fulcra.


Pre-caudal scute. Associated anterior to the caudal fin are the paired dorsal and ventral pre-caudal scutes, the morphology of which closely matches that of Asthenocormus titanius, but not the Callovian English asthenocormine Martillichthys renwickae (see discussion). The dorsal pre-caudal scute is complete and exposed on its dorsal surface, whereas the ventral pre-caudal scute is exposed in ventral view and is damaged along its posterior margin (Fig. 4). The scutes are thin and plate-like, symmetrical, and characterised by a pair of massive ‘wing-like’ lateral processes, a deep medial anterior notch, and paired medial keels on the ventral surface (Fig. 6D–E). The lateral processes are wider distally than they are proximally, the anterior face is marked by a deep concavity before expanding anteriorly each side of the anterior medial notch. Both the dorsal and ventral surfaces are moderately striated with fine longitudinal grooves radiating across the lateral processes from the midpoint of the bone. The external margins of the lateral processes are moderately scalloped or weakly serrated in areas. Posteriorly, the lateral processes taper to an acute medial point, forming a small posterior process. It is unknown if this feature is also present in the damaged ventral pre-caudal scute.

Fig. 6
figure 6

Disparity in pre-dorsal scute morphology between Jurassic asthenocormine pachycormiforms. A Saurostomus esocinus, dorsal view, Toarcian, Baden-Württemberg (SMNS 56982); B Germanostomus pectopteri (holotype), dorsal view, Toarcian, Baden-Württemberg (SMNS 15815); C Ohmdenia multidentata (holotype), dorsal view, Toarcian, Baden-Württemberg (GPIT-PV-31531); D Asthenocormus cf. titanius, dorsal view, Kimmeridge Bay, Dorset (MJML K1964); E Asthenocormus cf. titanius, ventral view, Kimmeridge Bay, Dorset (MJML K1964); (F) Pachycormus macropterus, dorsal view, Toarcian, Baden-Württemberg (BSPG 1940-I-6); G Martillichthys renwickae (holotype), dorsal view (?), Callovian, Peterborough (NHMUK PV P. 61,563); H Asthenocormus titanius, dorsal view, Tithonian, Bavaria (L.1309, from Lambers, 1992); I Asthenocormus titanius, ventral view, late Kimmeridgian, Bavaria (DMA-AS-1, Albersdörfer Collection, Dinosaurier Museum Altmühltal). Abbreviations: ano.scu, anterior notch of the pre-caudal scute; asp.scu, anteriorly protruding spine of the pre-caudal scute; ltpr.scu, lateral process of the pre-caudal scute; mdr.scu, medial dorsal ridge of the pre-caudal scute; mk.scu, medial keel of the pre-caudal scute. All scale bars equal 5 mm. Figure parts A–C, G–H are redrawn and modified from Cooper et al. (2022)

Full size image

Remarks. The presence of unsegmented caudal lepidotrichia which asymmetrically bifurcate free from joints is a synapomorphy of Pachycormiformes (e.g. Dobson et al., 2019; Lambers, 1992; Maxwell et al., 2020) thereby confidently identifying MJML K1964 as a pachycormiform. An absence of fringing fulcra on the caudal fin, apparent absence of body scales, lack of a scaly caudal apparatus (see Arratia & Schultze, 2013; Cooper & Maxwell, 2022) and the presence of a pair of enlarged pre-caudal scutes with massive lateral processes confidently identifies this specimen as a derived member of the subfamily Asthenocorminae (Cooper, 2023; Cooper et al., 2022). Consequently, MJML K1964 is referred with confidence to Asthenocormus based on the distinct morphology of the pre-caudal scutes, which are much closer in form to those of A. titanius, than to Martillichthys renwickae (see comparative anatomy) (Fig. 6). Due to incompleteness, the specimen does not preserve sufficient key characters to justify referral to A. titanius (see Lambers, 1992) and therefore we tentatively refer it to Asthenocormus cf. titanius. It is not unlikely that the Kimmeridge Clay specimen, as suggested by the variation in pre-caudal scute morphology and its geographic occurrence far from the type area, might indeed represent a new undiagnosed species of Asthenocormus. However, more complete specimens from the Kimmeridge Clay Formation will need to be found to test this hypothesis.

Specimen MJML K1964 was previously proposed by Martill and Brito (2020) to belong to a new, as yet undescribed acipenseriform sturgeon, with the pre-caudal scutes misidentified as the lateral scutes typical of extant members of Acipenseridae. We do not identify any synapomorphies of either Acipenseridae or Chondrostei in MJML K1964, and therefore confidently rule out this interpretation. The unsegmented bifurcation in the lepidotrichia is not present in Chondrostei; rather the caudal and unpaired fins readily segment both in the presence of, and external to fin ray bifurcation (e.g. Grande & Bemis, 1996; Hilton & Forey, 2009). Furthermore, the presence of ‘epurals’ articulating with basal fulcra confirms that this is a caudal fin, and not an unpaired fin as proposed by the previous authors. The caudal fins of acipenseriforms are highly diagnostic and primitive. They are characterized by an asymmetrical heterocercal fin with the notochord extending into the dorsal lobe, segmented lepidotrichia, enlarged scute-like epaxial fulcra (syn. ‘dorsal caudal fulcra’, e.g., Hilton & Forey, 2009), and lateral rhombic caudal scales on the epaxial lobe (Grande & Bemis, 1996; Hilton & Grande, 2023). Chondrostei additionally do not possess ‘epurals’ which are characteristic of Pachycormiformes and other early diverging clades of Teleosteomorpha (see Arratia & Lambers [1996] for comments on homology). The interpretation by Martill & Brito (2020) was novel for acipenseroid evolution as it suggested a potential Jurassic origin for the true sturgeons, which had only previously been recorded from the Early Cretaceous onwards (e.g., Hilton & Grande, 2023; Hilton et al., 2011; Martill, 2023). Our overturning of this hypothesis returns the proposed earlier emergence of Acipenseridae in the fossil record forward to the Early Cretaceous.

Comparative anatomy

‘Epurals’. The ‘epurals’ (in addition to the missing ‘uroneurals’ – see Arratia & Lambers, 1996; Arratia & Schultze, 2013 for comments on homologies) hold important taxonomic value in Pachycormiformes as they provide strong evidence for the clade’s inclusion within the teleost stem-group Teleosteomorpha (Arratia, 1999; Arratia & Lambers, 1996; Arratia & Schultze, 2013; Patterson, 1977). For derived asthenocormines, the morphology of the ‘epural’-like elements is known for Asthenocormus titanius (Arratia & Schultze, 2013; Lambers, 1992; pers. obser. SC) and Martillichthys renwickae (Cooper et al., 2022), but are not yet identified in Leedsichthys problematicus (e.g., Liston, 2007, 2008). In congruence with the Kimmeridge Clay asthenocormine, the presence of approximately 4 ‘epurals’ is a shared character between the Kimmeridgian - Tithonian A. titanius and the Callovian M. renwickae (Cooper et al., 2022), although the morphology of these elements varies between the two genera. In the older M. renwickae, the ‘epurals’ are rounded in cross section, are tall and narrow, and only weakly expand at the distal extremities. These elements are comparatively stouter and more robust in A. titanius, with a somewhat rectangular to amorphous cross section. Furthermore, ‘eu’-1 is noticeably shorter than proceeding ‘eu’ 2–4 in A. titanius and MJML K1964, but not in M. renwickae where all four ‘epurals’ are uniform in size (Cooper et al., 2022). Therefore, the morphology of the surviving caudal endoskeleton in MJML K1964 more closely matches Asthenocormus titanius than Martillichthys renwickae.


Pre-caudal scute. The pre-caudal scute is a morphologically variable element across Pachycormiformes (Fig. 6) where it has previously been demonstrated to hold important taxonomic value (Cooper et al., 2022). Although all asthenocormine pachycormids share the presence of a modified pre-caudal dorsal scute, some taxa in this clade also possess a modified ventral pre-caudal scute (Pachycormus + Asthenocormus), whilst other closely related taxa do not (Saurostomus + Germanostomus) (Cooper et al., 2022). Assessing the phylogenetic implications of this character across Asthenocorminae is problematic due to the caudal fins for many members of the clade being incompletely preserved (Ohmdenia; Martillichthys; Leedsichthys; Bonnerichthys) or entirely unknown (Rhinconichthys spp.). Numerous well-articulated caudal fins are known for Saurostomus esocinus, none of which present any evidence for a ventral pre-caudal scute (Cooper & Maxwell, 2022). If the scute serves a hydrodynamic function, it is possible that the uniquely modified fringing fulcra on the ventral (hypaxial) caudal lobe in Saurostomus esocinus and Germanostomus pectopteri (Cooper et al., 2022) perhaps compensated a similar streamlined function thereby rendering a modified ventral scute unnecessary.

A small, scale-like dorsal scute is present in the basal pachycormiform Euthynotus incognitus (Arratia, 2008; Arratia & Lambers, 1996), indicating the presence of a pre-caudal scute to be an ancestral feature in the clade. A few hypsocormine pachycormiforms (Hypsocormus insignis, H. posterodorsalis, Orthocormus roeperi, O. cornutus?, Sauropsis longimanus) are reported to possess an elongate rod-like dorsal pre-caudal scute (Arratia & Lambers, 1996; Arratia & Schultze, 2013; Maxwell et al., 2020). However, this structure in Hypsocorminae does not appear to be homologous with the dorsal scute of E. incognitus and Asthenocorminae; rather this modified element in Hypsocorminae + Sauropsis is possibly a mineralized tendon (A. López-Arbarello, pers. comm.). The subdermal placement of this element is clearly observed in the holotype of Sauropsis longimanus (BSPG AS VII 1089) where it is preserved underneath the preserved integument (pers. observ. SC & EM).

The presence of massive, laterally expansive, wing-like processes on the scute of MJML K1964 closely allies its morphology to derived Upper Jurassic representatives of Asthenocorminae, notably A. titanius and M. renwickae. Lateral processes are absent on the dorsal pre-caudal scutes of all Early Jurassic representatives of Asthenocorminae: Pachycormus macropterus (Fig. 6F), Saurostomus esocinus (Fig. 6A), Germanostomus pectopteri (Fig. 6B), and Ohmdenia multidentata (Fig. 6C). An anterior bifurcation of the scute, forming an anterior medial notch which is present in the scutes of Asthenocormus and Martillichthys, is shared in Pachycormus and Saurostomus, but not in Germanostomus or Ohmdenia (Fig. 6). A dorsal or ventral scute has not yet been identified in Leedsichthys problematicus (sensu Liston & Noè, 2004; Liston, 2008) or any of the Cretaceous asthenocormines (Bonnerichthys gladius and Rhinconichthys spp.) (e.g. Friedman et al., 2010).

The dorsal scute of M. renwickae is extremely robust with thickened lateral processes, and the anterior notch is very poorly developed, existing only as a weak concavity. This strongly conflicts with the much thinner plate-like morphology with a noticeably more elongate anterior margin of the scute shared between A. titanius and MJML K1964. However, similar to MJML K1964 and some smaller examples of A. titanius (Fig. 6H), the posterior margin of the scute in M. renwickae is convex although it does not taper as tightly to a posterior apex, and the lateral processes narrow distally with smooth rounded edges and do not show any traces of ornament (Cooper et al., 2022).

Of the materials referred to Asthenocormus titanius preserving pre-caudal scutes (Table 2), the Painten specimen (DMA-AS-1) has dorsal and ventral pre-caudal scutes that more closely match those of MJML K1964. The lateral processes are wide and thin, slightly narrowed distally, and possess weakly scalloped and serrated margins (Fig. 6I). Unlike the smaller Solnhofen A. titanius specimens (e.g., L. 1309: Fig. 6H) (see Table 1), neither the larger Painten example of A. titanius nor MJML K1964 show an anterior distal curvature at the extremities of their lateral processes. MJML K1964 differs from the Painten A. titanius specimen in that the posterior margin of the scutes are more acutely pointed, the lateral margins are more sharply scalloped, and the external surface bears weak radiating striations (Figs. 4B; 6D-E). Therefore, although not a perfect match, the morphology of the pre-caudal scutes in MJML K1964 most closely match larger individuals of A. titanius than they do smaller individuals of this species or M. renwickae. Based on a pre-caudal scute width of 37 mm, MJML K1964 likely represents a slightly larger individual of Asthenocormus than the Painten specimen (see morphometric analysis). The smallest known complete specimen of A. titanius (LF 2140P = 360 mm TL) displays a similar pre-caudal scute to that of the ‘small’ Solnhofen specimens, both lacking marginal striations and instead possessing very smooth lateral margins. The noticeable increase in the scalloped margins and external radiating striations in larger Asthenocormus individuals (e.g. MJML K1964) may therefore be a signal of progressive ontogenetic development in this genus. Unfortunately, most specimens of A. titanius from Bavaria do not usefully preserve the pre-caudal scutes (Table 2) and it is undetermined whether the variation observed between the Dorset and Bavarian Asthenocormus specimens are expressions of ontogenetic variation, intraspecific variation, or speciation. It is based on this uncertainty that we refer MJML K1964 to Asthenocormus cf. titanius.

Table 1 Materials from the UK previously assigned to Asthenocormus
Full size table
Table 2 Documented specimens of Asthenocormus titanius or referred to A. titanius
Full size table

Morphometric analysis

To provide an estimate for the reconstructed total length of MJML K1964, a preliminary morphometric analysis was performed using the pre-caudal scute width (PCSW) as a proxy to estimated total length (TL). For complete and well-articulated Asthenocormus titanius specimens, the total length was divided by the pre-caudal scute width to produce an estimated scaling ratio. Of the surviving specimens of A. titanius (Table 2), only 3 are skeletally complete, well-articulated, usefully preserve the pre-caudal scute, and are accessible for study. The rest are either incomplete, do not preserve the pre-caudal scutes, or are held in private collections. A scaling factor of × 61.71(2160 mm TL /35 mm PCSW) is calculated for the neotype specimen (JM-E SOS 542a/b), with a value of × 73.91 calculated for the Painten specimen (700 mm TL/23 mm PCSW). The complete Lauer Foundation specimen represents a small juvenile (360 mm TL) and therefore is likely less reliable for use in morphometric scaling of adult specimens (e.g. MJML K1964). These ratios imply a reconstructed total length for MJML K1964 of between 2283 mm est. TL (37 mm PCSW × 61.71) – 2735 mm est. TL (37 mm PCSW × 73.91), with an average total length estimate of 2509 mm est. TL (av. 67.81 × 37 mm). Even at the lowest possible estimate, MJML K1964 has a greater estimated total length than the largest complete specimen of Asthenocormus titanius from Germany (Table 1). It should be noted however that fragmentary post-cranial remains of significantly larger Asthenocormus titanius individuals with total lengths likely far exceeding MJML K1964 are known from the Bavarian plattenkalks (Table 1).

Disputed remains of Asthenocormus from the Kimmeridge Clay Formation

Systematic palaeontology

Subfamily Hypsocorminae (Vetter, 1881) sensu Cooper et al., 2022


Hypsocorminae gen. et sp. incertae sedis


Figure 7

Fig. 7
figure 7

Hypsocorminae gen. et sp. indet., MJML K1556, previously misattributed to Asthenocormus. A Specimen preserved predominantly as a plaster of Paris cast, in ventral view, missing only the post-anal fin region. The fish’s outline has been painted. B Line drawing showing detail of the ‘special fringing fulcra’ (see Gouiric-Cavalli & Arratia, 2022); the absence of these structures is an apomorphy of Asthenocorminae with the exception of Pachycormus. C Details of the paired pelvic fins – an apomorphy of non-asthenocormine pachycormiforms. Abbreviations: anfn, anal fin; fr.fu, ‘special’ fringing fulcra; pcfn.l, left pectoral fin; pc.lep, pectoral lepidotrichia; pvfn, pelvic fin (l = left; r = right). Scale bar equals 50 mm (A), 10 mm (C) and 5 mm (B)

Full size image

2019 – Kimmeridgian juvenile, Liston et al., 2019 p. 6, table 1.


2020. – ?Asthenocormus sp., Martill & Brito, p. 36, fig. 3.2.


Material. Incomplete medium-sized fish in ventral view, pectoral fins splayed laterally, missing the post-anal region (MJML K1556), prepared mostly as a plaster cast (Fig. 7A).


Locality and Horizon. elegans Zone, Kimmeridge Clay Formation, earliest Tithonian (Gallois, 2020), Upper Jurassic; eastern wave-cut platform of Kimmeridge Bay, Kimmeridge, Dorset.


Description. The fish is extremely weathered and hence is predominantly preserved as an external plaster cast measuring 380 × 215 mm. The skeleton is preserved in ventral view with the paired fins splayed and the anal fin poorly preserved. The specimen has a pre-anal length of 320 mm, with the skull region measuring 90 mm in length. The bladed pectoral fins measure 95 × 25 mm. The fish cast has been painted with a light pinkish-brown wash to highlight the original outline, whilst areas of original bone and integument are preserved as irregular black patches across the specimen. The apparently elongate shape of the skull is a combined artefact of taphonomy, weathering, and application of the paint wash which has inaccurately distorted the aspect ratio of the cranium.

The pectoral fins are gladiform and wide with the posterior fillet (Liston et al., 2019) either poorly developed or incomplete. Light microscopy reveals a single row of well-developed fringing fulcra preserved as impressions along the distal leading edge of the pectoral fins (Fig. 6B). This pattern is consistent with the pectoral fin morphology of Hypsocorminae, Sauropsis longimanus, and Pachycormus macropterus (Cooper, 2023; Wenz, 1967).

The pelvic fins are placed midway between the pectoral and anal fin, are each damaged distally, and comprise of ≥ 20 delicate lepidotrichia (Fig. 6C). It is unclear if the pelvic fins hold fringing fulcra. The anal fin is poorly preserved with only a few fin ray fragments remaining. There is no trace of the pre-anal scutes, squamation or intestinal tract.


Remarks. Martill & Brito (2020) attributed this specimen to ?Asthenocormus sp. based on an apparent absence of teeth, an elongate skull, and large scythe-shaped pectoral fins with unsegmented rays. Based on the small size of the specimen, the authors suggested it may represent a juvenile individual. Our discovery of paired pelvic fins in this specimen demonstrates non-Asthenocorminae affinities, as all asthenocormine pachycormids (including A. titanius) lack pelvic fins. Indeed, absence of pelvic fins is the shared apomorphy for Asthenocorminae (Cooper et al., 2022). The loss of pelvic fins is a basal condition in asthenocormines, the disappearance of which appears to pre-date the clade’s first occurrence in the fossil record during the Toarcian stage of the Early Jurassic (Cooper & Maxwell, 2022). Pelvic fins are always present in basal pachycormids (Euthynotus spp.) and the subfamily Hypsocorminae (Vetter, 1881, sensu Cooper et al., 2022), which includes the genera Hypsocormus and Orthocormus: both of which are represented in the Kimmeridge Clay Formation (Martill & Brito, 2020). The apparently “elongate” shape of the skull and absence of teeth are almost certainly artefacts of poor preservation and the fish being compacted in a dorsoventral orientation. Moreover, the morphology of the pectoral fin is closer to that of Hypsocormus spp. than to Asthenocormus titanius, notably in the presence of fringing fulcra. Pectoral fringing fulcra (syn. ‘special fulcra’ – Gouiric-Cavalli & Arratia, 2022) are only observed in the basal asthenocormine Pachycormus macropterus (Liston et al., 2019; Mainwaring, 1978; Wenz, 1967), but are absent in all other members of Asthenocorminae (Cooper et al., 2022). Fringing fulcra are, however, present in most hypsocormine pachycormids including Hypsocormus spp. (Arratia & Schultze, 2013; Cooper, 2023; Gouiric-Cavalli & Arratia, 2022; Lambers, 1992; Liston et al., 2019; Maxwell et al., 2020) and Sauropsis spp. (Maxwell et al., 2020; pers. observ. SC & EM). Specimen MJML K1556 is too poorly preserved to sufficiently identify synapomorphies of Hypsocormus, although the smaller size of the pelvic fins in relation to the pectoral fins of MJML K1556 supports referral to Hypsocormus sp. as opposed to Orthocormus sp. (sensu Maxwell et al., 2020). The Upper Jurassic hypsocormines Simocormus macrolepidotus Maxwell et al., 2020, Sauropsis depressus Eastman, 1914, and Pseudoasthenocormus retrodorsalis (Eastman, 1914) are ruled out as potential candidates based on body size, body elongation, and paired fin morphologies.

Discussion

Paleobiogeographic distribution of Asthenocormus

Asthenocormus has a brief stratigraphic range of late Kimmeridgian – early Tithonian. All confirmed records of this genus occur in Western Europe, in the northwestern Tethys. Outside of the Bavarian plattenkalks (incl. Solnhofen, Eichstätt, Painten), A. titanius has previously been reported based on tentative records from the Upper Jurassic of England (Oxford Clay Formation; Kimmeridge Clay Formation) and possibly Antarctica (Gouiric-Cavalli et al., 2019).

From the Callovian-aged Peterborough Member of the Oxford Clay Formation of Peterborough, Schaeffer and Patterson (1984) recorded a large, near-complete toothless fish as Asthenocormus sp. collected previously from the Dogsthorpe Pit by one of the present authors (D.M.M). This specimen was later described by Martill (1991), who mentioned two additional referred specimens: an unprepared cranium in a nodule, and a partial pectoral fin, both also referred to Asthenocormus sp. based on size, being much smaller than the constituent elements in the larger-bodied, although similarly edentulous pachycormiform Leedsichthys problematicus. It was based on this interpretation that Lambers (1992) proposed Asthenocormus to have a vicariant distribution between the English and southern German regions of the Tethys. Liston (2008) identified these materials as representing a unique genus and species distinct from both A. titanius and L. problematicus, erecting the binomial Martillichthys renwickae Liston, 2008 to accommodate these specimens. Subsequent re-description of the holotype and phylogenetic analysis supports M. renwickae as a valid taxon topologically distinct from both A. titanius and L. problematicus (Cooper & Maxwell, 2022; Cooper et al., 2022; Dobson et al., 2019; Friedman, 2011; Friedman et al., 2010; Schumacher et al., 2016). At present there is no evidence to support the presence of the genus Asthenocormus in the Oxford Clay Formation. The specimen reported as ‘?Asthenocormus sp.’ from the Upper Jurassic Kimmeridge Clay Formation (MJML K1556) is here found to also not belong to the genus Asthenocormus but rather is an indeterminate hypsocormine pachycormiform. No other horizons in the UK have reported remains of Asthenocormus, with the newly identified specimen MJML K1964 from the Tithonian of Dorset representing the only true occurrence of this historic German genus from the UK. A palaeobiogeographic distribution between southern England and southern Germany is therefore here reaffirmed for Asthenocormus, based on entirely new evidence.

Evidence for a potential bipolar dispersal of Asthenocormus into the southern hemisphere is tantalizing but poorly supported. Gouiric-Cavalli et al. (2019) referred two associated gill rakers, a fragmentary hyomandibula and an incomplete ceratohyal to aff. Asthenocormus sp. from the lower Kimmeridgian Longing Member of the Ameghingo Formation in the north-eastern region of the Antarctic Peninsula. Morphologically, the gill rakers reported by these authors do show strong affinities with Asthenocormus (Lambers, 1992) as opposed to Leedsichthys, Martillichthys or Rhinconichthys (gill rakers are unknown for Bonnerichthys gladius) (Gouiric-Cavalli et al., 2019; Liston, 2013; Schumacher et al., 2016). However, as demonstrated in Leedsichthys problematicus (Martill 1999; Liston, 2007, 2013) gill raker morphologies in asthenocormine pachycormiforms show a high level of intraspecific plasticity, and thus are probably not usefully reliable for taxon recognition. The ceratohyal figured by Gouiric-Cavalli et al. (2019) is incomplete but appears to show the typical rectangular morphology seen in all asthenocormines which preserve this bone (e.g., Liston et al. 2008; Schumacher et al., 2016; Dobson et al., 2019; Cooper & Maxwell, 2022). A lack of diagnostic landmarks and spatial distance from the European type area casts into doubt the generic affinities of the Antarctic Peninsula asthenocormine remains reported by Gouiric-Cavalli et al. (2019). No further records of the genus Asthenocormus are reported outside of European Laurasia.

Pachycormiformes does have a bipolar distribution and is predominantly restricted to high temperate latitudes throughout their entire evolutionary history, with only few rare remains found in the tropics of the proto-Caribbean and northern Chile (Cooper, 2023; Gouiric-Cavalli et al., 2019; Gregory, 1923). Pachycormiforms were therefore possibly temperate fishes, although a severe lack of suitable pelagic marine middle-Mesozoic fossil localities in the palaeotropics (Benson & Butler, 2011) likely obscuring their true palaeobiogeographic distribution. The remains from the southern hemisphere tentatively support a bipolar distribution for Asthenocormus, with this genus likely migrating into southern Gondwana via the Proto-Hispanic Corridor (Gouiric-Cavalli et al., 2019). A similar bipolar distribution is congruent with the established palaeobiogeographic distribution of the much larger asthenocormine pachycormid Leedsichthys problematicus, which has similarly been reported from both Europe (UK, France, Germany) (Woodward, 1889, 1895; Michelis et al., 1996; Liston 2004; 2008; 2010) and South America (Chile, ?Argentina) (Martill et al., 1999; Liston et al. 2010; 2013; Gouiric-Cavalli, 2017; Gouiric-Cavalli et al., 2019). Whereas Leedsichthys is known from the Tropic of Capricorn in northern Chile (Liston, 2010; Martill et al., 1999), fossil remains referable to Asthenocormus are not yet known from within the tropical belt.

Palaeoecology of Asthenocormus

Asthenocormus is currently found as a rare faunal component in both proximal carbonate lagoonal (Bavarian plattenkalks) and epicontinental siliciclastic shelf environments (Kimmeridge Clay Formation) indicating this genus likely had a pelagic ecology. The total body size range for Asthenocormus titanius has yet to be determined due to a low sample size and incompleteness of most specimens. The lost holotype was said to be “almost 8 feet (~ 2.44 m) in length, excluding the tail”, giving it an estimated length of approximately 2500 mm TL. No surviving specimen exceeds 2220 mm TL (Table 1), however a pair of gigantic pectoral fins likely belonging to Asthenocormus titanius from the early Tithonian of Wintershof near Eichstätt (BSPG 1951 XVI-1) attests to the enormous potential size of this fish, with the pectoral fins alone measuring 790 mm long (pers. obs. SC.). Furthermore, an isolated asthenocormine caudal fin ray fragment from the Mörnsheim Formation (lower Tithonian) at Mühlheim (SNSB-BSPG 2015/80) (Fig. 8) possibly belonging to Asthenocormus titanius represents a gigantic caudal fin with a depth likely exceeding 2000 mm. Despite the distinctive bifurcation pattern confirming pachycormiform affinities, isolated fin rays are not usefully diagnostic in pachycormids and thus the taxonomy of these fin fragments should be referred with caution. The Kimmeridge Clay Formation Asthenocormus cf. titanius specimen additionally represents a sizable individual (est.TL = 2509 mm) which, based on the width of the pre-caudal scutes, implies a total length in excess of the largest complete specimens from Bavaria (Table 2). It is therefore plausible to hypothesise that the more proximal settings of the Bavarian plattenkalk were predominantly occupied by younger individuals, whilst larger members of this genus eventually migrated into more off-shore environments with greater water depths. However, the presence of rare fragments belonging to enormous individuals of Asthenocormus from the Solnhofen plattenkalks do not conform to this hypothesis. A similar body-size distribution governed by water depth has previously been proposed for the Early Jurassic pachycormid Pachycormus macropterus (Cooper & Maxwell, 2023; Cooper, 2024). Changes in water depth preference with ontogeny in Pachycormus is governed by a dietary shift as this taxon matures (Cooper, 2024). The diet of Asthenocormus is known from two specimens, the small one described by Vetter (1881) and the slightly larger neotype (JM-E SOS 542a/b; Viohl, 1990). These preserve remains of small teleosts (5–10% predator TL; Leptolepides-like, SNSD BaJ2344; Tharsis, JM-E SOS 542a/b) inside of the gut, which suggests small fishes were a frequent dietary component for A. titanius. The diet is entirely unknown in both the Kimmeridge Clay specimen and in the largest individuals from the Bavarian plattenkalks. It is therefore unknown whether diet changed with ontogeny in Asthenocormus in very small juveniles (< 1 m in length), but appears to have remained constant in fishes above this size. Furthermore, the smallest complete A. titanius specimen (360 mm TL), likely representing a juvenile individual (LF 2140P), does not appear to preserve any discernible prey remnants in the gastrointestinal tract.

Fig. 8
figure 8

Asthenocorminae gen. et sp. indet., SNSB-BSPG 2015/80, gigantic section of a single bifurcated caudal fin ray with affinities to Asthenocormus from the Lower Tithonian Mörnsheim Formation at Mühlheim. The fin ray is preserved in several pieces and has been digitally reconstructed. Arrows indicate the zones of bifurcation. Scale bars equals 100 mm

Full size image

Asthenocormines were historically referred to as the “suspension-feeding” or “filter-feeding” pachycormiforms on the basis of their elongate, toothless jaws and modified gill rakers, being somewhat convergent on planktivorous chondrichthyans and cetaceans (e.g. Friedman et al., 2010; Lambers, 1992; Liston, 2008). However, no planktonic prey has yet been identified in association with any of these taxa. Those members of Asthenocorminae which do preserve gut contents demonstrate either a piscivorous (Asthenocormus titanius + Martillichthys renwickae: Vetter, 1881; Martill, 1988; Cooper et al., 2022), teuthophagous (e.g. Saurostomus esocinus + Ohmdenia multidentata: Cooper & Maxwell, 2022), or generalist macrophagous diets (e.g. Pachycormus macropterus: Přikryl et al., 2012; Cooper & Maxwell, 2023; Cooper, 2024; Weis et al., 2024). The elongate toothless jaws of Asthenocormus were therefore likely a predatory adaption, with the enlarged buccal gape supporting a ram feeding strategy for engulfing shoals of smaller fishes. A similar feeding strategy was also likely adopted by Martillichthys renwickae based on its similar body size, jaw shape, and fossilised gut contents.

The diet of Asthenocormus is therefore that of a ‘small nekton’ bulk feeder as the size of the prey fishes (Leptolepides =  ≤ 150 mm TL) are each minuscule in relation to the predator’s body size, potentially allowing for multiple prey fishes to be ingested in a single successful strike. This diet is therefore clearly differentiated from the more macrophagous raptorial ecologies of toothed asthenocormines (e.g. Saurostomus and Pachycormus) which likely captured and processed individual prey items that were much larger in proportion to their own body size compared to Asthenocormus (Cooper & Maxwell, 2022; 2023; Cooper, 2024; Weis et al., 2024).

Diversity of Pachycormiformes in the Kimmeridge Clay Formation

Despite being represented by a large number of well-preserved and excellently prepared specimens available in the Etches Collection, pachycormiforms remain poorly identified and mostly undescribed in the Kimmeridge Clay Formation. The most comprehensive faunal list is provided by Martill & Brito (2020) who, in addition to misidentifying an example of ?Asthenocormus sp., list materials of Hypsocormus cf. tenuirostris Woodward, 1889 (= Orthocormus tenuirostris (Woodward, 1889)) and ?Orthocormus sp.. These authors speculated that Leedsichthys may had been present in the Kimmeridge Clay Formation of the UK based on reports of this fish from the Kimmeridgian of France (Liston, 2010). Woodward (1895) had previously noted supposed Leedsichthys remains from the Kimmeridge Clay Formation but did not specify or figure these materials. Following extensive searches of the Etches Collection and NHMUK, we are unable to identify any diagnosable remains which would suggest that Leedsichthys problematicus is present in the Kimmeridge Clay Formation of Dorset.

Based on our examination of materials housed in the Etches Collection, we identify the following pachycormiform taxa as being present in the Kimmeridge Clay Formation of Dorset: Hypsocormus cf. insignis, Hypsocormus sp., Orthocormus cf. cornutus, Orthocormus sp., Asthenocormus cf. titanius, and Hypsocorminae gen. et sp. indet. (Fig. 9). No materials of ‘Hypsocormus’ tenuirostris (= Orthocormus tenuirostris) were identified in our analysis, with the specimen identified by Martill and Brito (2020) reappraised as a misidentified example of Orthocormus sp.. Orthocormus tenuirostris is a large hypsocormine pachycormiform from the Callovian Oxford Clay Formation (Martill, 1991), with no confirmed records of this species existing outside of the type formation (see Cooper [2023] for review on stratigraphy and palaeobiogeography). Several complete and articulated skulls of Hypsocormus cf. insignis are preserved in the Etches Collection (Fig. 9A, B), although unfortunately none preserve a post-cranial skeleton which is essential to confidently differentiate between the two described species of this genus: H. insignis Wagner, 1863 and H. posterodorsalis Maxwell et al., 2020. Two morphologies of Orthocormus are recognised in the formation: one has a gracile projecting rostrodermethmoid consistent with Orthocormus cornutus Weitzel, 1930 from the Tithonian of Bavaria, whilst a second morphology has a much stouter rostrodermethmoid and teeth which is closer to the unnamed Orthocormus sp. from the Kimmeridgian of Nusplingen in Baden-Württemberg (see Maxwell et al., 2020) (Fig. 9C, D). A large articulated post-cranial skeleton of Orthocormus in the Etches Collection displays fin proportions which are more closely consistent with O. cornutus than O. teyleri Lambers, 1988, or O. roeperi Arratia & Schultze, 2013. Frustratingly, the skull is missing, thereby obscuring which Orthocormus cranial morphology in the Kimmeridge Clay Formation this specimen matches with. Several incomplete skulls and jaw fragments collected from the coccolithophore-bedded limestone band of the Kimmeridge Clay Formation (White Stone Band) appear to represent a new undescribed species of Hypsocorminae with close affinities to Orthocormus spp. (Fig. 9F). We also identified an enigmatic dentary from Kimmeridge Bay (MJML K1464) which is unusual for possessing Orthocormus-like teeth, but differs from all pachycormiforms by displaying an anteriorly inflated corpus and the presence of fine, weakly ornamented ganoine on the external surface (Fig. 9E). Ganoine is not present on the external jaws of any hypsocormine pachycormiform, suggesting this jaw either belongs to a unique species of Hypsocorminae, or is not a pachycormiform. Whilst we were unable to identify any specimens with possible affinities to Leedsichthys problematicus in the Kimmeridge Clay Formation, there exists in the Etches Collection numerous large isolated bones of as-yet unidentified large-bodied actinopterygians. It is difficult to refer any of these elements with confidence to Pachycormiformes let alone to Leedsichthys. At present, we consider only the pachycormiform genera Hypsocormus, Orthocormus, and Asthenocormus to be present in the formation, with the number of species being much higher, although as yet undetermined. A detailed taxonomic review of the hypsocormine pachycormid remains in the Kimmeridge Clay Formation is needed to better understand the hidden taxonomic diversity and palaeobiogeography of pachycormiforms in the Upper Jurassic.

Fig. 9
figure 9

Diversity of hypsocormine pachycormiforms in the Kimmeridge Clay Formation. A Hypsocormus cf. insignis Wagner, 1863, MJML K(A) 1005, complete skull in right lateral view. B Details of the rostrum in MJML K(A) 1005. C Orthocormus sp., MJML K1909, complete skull and pectoral girdle prepared in right lateral view. D Orthocormus cf. cornutus, MJML K2533, anterior region of the skull roof (dorsal view) and lower jaws (external view) showing details of the anteriorly projecting rostrodermethmoid. E Hypsocorminae gen. et sp. indeterminate, MJML K1464, isolated left dentary with an unusually inflated anterior portion and ornamented with thin ganoine. F Hypsocorminae gen. et sp. indeterminate (?Orthocormus sp. nov.), MJML K1050, articulated skull and right pectoral fin, missing the rostrodermethmoid, from the bedded coccolithophore limestone horizon. Abbreviations: ag, angular; chy, ceratohyal; cor, coronoid; d, dentary; hyo, hyomandibula; mx, maxilla; na, nasal; op, opercle; p, parietal; pcfn, pectoral fin (l, left; r, right); pmx, premaxilla; prp, propterygium; rdm, rostrodermethmoid; so2, suborbital 2; t, teeth. Scale bars equal 10 mm (B), 20 mm (D–F), and 50 mm (A, C)

Full size image

Asthenocormus cf. titanius is only known from the Kimmeridge Clay Formation based on the single incomplete caudal fin reported here. No other confirmed asthenocormine materials were identified in the Etches Collection, suggesting that, despite their large body sizes and preference towards open-water pelagic environments, the subfamily Asthenocorminae was a rare faunal component in the Kimmeridge Clay palaeoichthyofauna. The weakly articulated and reduced skeletal ossifications of asthenocormines undoubtedly contributes to their perceived scarcity in the fossil record, especially from the Middle Jurassic onwards, with articulated remains of this subfamily restricted to Konservat-Lagerstätte (Cooper, 2023; Dobson et al., 2019; Lambers, 1992; Schumacher et al., 2016). The excellent preservation quality in the Kimmeridge Clay Formation and extensive collecting suggest that taphonomy is not the primary cause for the rarity of Asthenocormus in the formation; rather this taxon was likely an occasional faunal component. The Etches Collection Asthenocormus, despite being only known by a single fragment collected from a cliff fall, is well preserved, with this individual possibly representing a migratory individual which coincidently died in the environment.

Conclusions

The large edentulous pachycormid fish Asthenocormus cf. titanius is described from the Kimmeridge Clay Formation of Dorset, representing the first true record of this historic genus to be documented from the UK. All previous records documenting the presence of Asthenocormus in the UK are proven to be outdated, with the materials in question since referred to different genera or inconsistent with the pachycormid subfamily Asthenocorminae. The presence of stiff, unsegmented lepidotrichia which distally bifurcate independent of joints, reduced skeletal ossification, presence of 4 ‘epurals’, and a pair of massive pre-caudal scutes with larger plate-like lateral processes, strongly supports referral of MJML K1964 to the genus Asthenocormus. Variation in the pre-caudal scute morphologies of the Kimmeridge Clay Formation and German plattenkalk specimens suggest that the Dorset specimen may represent a unique species of Asthenocormus; however, due to incompleteness we conservatively refer the Etches Collection specimen to Asthenocormus cf. titanius. The proposed presence of an acipenserid sturgeon in the Kimmeridge Clay Formation as reported by Martill & Brito (2020) is dismissed, with the origins of Acipenseridae reverted forward to the Cretaceous. The palaeobiogeographic distribution of Asthenocormus is also extended far outside of the Bavarian plattenkalks. The edentulous, elongate jaws in smaller-bodied toothless asthenocormines, including Asthenocormus, are proposed here to be a buccal adaption for pelagic ram feeding on smaller fishes [or small nekton], as indicated by fish-dominated gut contents in these toothless taxa.