A new janassid (Chondrichthyes, Petalodontiformes) from the Late Mississippian of Alabama, USA

ABSTRACT Teeth from the late Mississippian Bangor Limestone, Monteagle Limestone, and Pride Mountain Formation (early to middle Chesterian = late Viséan–early Serpukhovian) of northern Alabama, USA, are designated as Cavusodus whitei, gen. et sp. nov. The species is similar to Cypripediodens cristatus Duffin and Ward, known only from the Eyam Limestone Formation (Mississippian, late Viséan) of Derbyshire, England. Cypripediodens cristatus, the only species of the genus, is distinguished by teeth having an elongate crown, an apical cusp having a nearly circular cross-section, a secondary cusp on the lingual face, and several cristae on the lingual heel. Cavusodus whitei teeth resemble those of Cypripediodens cristatus but differ in lacking the secondary cusp, in having a more labiolingually compressed apical cusp, in possessing more robust lingual cristae, and in possessing a longitudinal groove on the labial face. The genera comprising the Janassidae are reconsidered. It is proposed that only Janassa, Cholodus, and Cypripediodens be retained, in addition to the new genus. urn:lsid:zoobank.org:pub:DB68190C-E9E6-48C7-B5A7-8A72D174CBCE


The Petalodontiformes
The Petalodontiformes Patterson 1965 is an order of Palaeozoic chondrichthyan fishes, the relationships of which are poorly understood (Hansen 1985;Lund et al. 2014).They are known mainly from isolated teeth, but also from rare associated dentitions and a handful of more complete specimens.Species known from more than dentitions include Janassa bituminosa (von Schlotheim, 1820), Belantsea montana Lund 1989, Netsepoye hawesi, 1989, Siksika ottae, 1989, Obruchevodus griffithi;Grogan et al. 2014, Petalorhynchus beargulchensis Lund et al. 2014, and Fissodopsis robustus, 2014.The latter six species were described from the Late Mississippian Bear Gulch Limestone of Montana, USA.Of these seven species known from more than isolated remains, Janassa bituminosa differs from the other six in having a dorsoventrally flattened body.Hence, its teeth might have functioned differently than in the other taxa.Due to the scarcity of articulated specimens, the relationships of the Petalodontiformes with respect to other chondrichthyans are unclear.
Petalodontiform teeth are characterised by (1) labiolingually compressed crowns (labially convex and lingually concave), (2) medially acuminate or rounded crown apices, (3) imbricated basal ridges, and (4) sigmoid lateral profiles (Hansen 1985).The teeth are arranged in the elasmobranch pattern of mesiodistally separated files, new teeth within a file emerging lingually.Ginter et al. (2010) recognised four families within the Petalodontiformes -the Petalodontidae Newberry and Worthen 1866, the Pristodontidae Woodward 1889, the Janassidae Hansen 1985, andthe Belantseidae Lund 1989.The Janassidae, understood as a group that includes Janassa Münster 1839 but not Petalodus Owen 1840 in Owen (1840Owen ( -1845)), which is assigned to the Petalodontidae, is sometimes attributed to Jaekel (1898).This is incorrect.Jaekel (1898) defined three subfamilies of the family Petalodontidae -the Janassinae, the Polyrhizodontinae, and the Pristodontinae.Jaekel's family Petalodontidae corresponds roughly to the order Petalodontiformes as currently understood.The Janassinae Jaekel (1898) includes the two genera Petalodus and Janassa.Thus, the Janassinae do not correspond to the Janassidae as currently understood.Of the other subfamilies of the Janassidae defined by Jaekel (1898), the Polyrhizodontinae includes Polyrhizodus McCoy 1848 and Gyracanthus Agassiz 1837 in Agassiz (1837Agassiz ( -1843) ) (now considered to be an acanthodian, not a petalodont), and the Pristodontinae includes Pristodus Davis 1883.Hansen (1985, p. 528) proposed the family Janassidae, in which he included Janassa, Fissodus St. John and Worthen 1875 and possibly Peltodus Newberry and Worthen 1870 and Cholodus St. John and Worthen 1875, should they prove to be distinct.Ginter et al. (2010) included only Janassa and Fissodus in the Janassidae, regarding Cholodus as a junior synonym of Fissodus and Peltodus as a junior synonym of Janassa.Lund et al. (2014) carried out a cladistic analysis based on taxa having holomorphic data.The resulting phylogenetic tree supports the Petalodontiformes as a clade including Belantsea Lund 1989, Netsepoye Lund 1989, and Obruchevodus Grogan et al. 2014 The family-level division of the Petalodontiformes of Ginter et al. (2010) was modified by Lund et al. (2014), based on a cladistic analysis based only on dental characters.This cladistic analysis supports the previously recognised families Petalodontidae, Belantseidae, and Janassidae and two new families, the Petalorhynchidae Lund et al. 2014 andthe Obruchevodidae Lund et al. 2014.

The Janassidae
The Janassidae includes Janassa, which is known from holomorphic specimens of the type species J. bituminosa, and other genera known only from teeth.The teeth of Janassa are strongly sigmoid, have an extended, ridged, lingual heel, and have a base oriented at nearly a 90° angle with respect to the lingual heel (Figure 1a-b, d).As can be seen in Figure 1c, the coronal apices tilt orally (occlusally) and probably carried out a grasping or cutting function.The ridged, lingual heels might have served a crushing function (Hancock and Atthey 1869;Hancock and Howse 1870;Jaekel 1899;Eastman 1903;Woodward 1920).
The replacement pattern of the teeth of J. bituminosa is unusual.Only the youngest teeth in a file have the lingual faces exposed and are functional in processing food (Figure 1c).Older teeth are retained but are covered by the younger teeth and form a supporting structure for the youngest, functional teeth.Figure 1e illustrates this tooth replacement pattern.A similar dental pattern, with older teeth in a file retained in a non-functional, nonocclusal position (Brandt 2009, fig. 9) is inferred for Janassa korni (Weigelt, 1930).
The dentition of J. bituminosa is weakly heterodont, as the symphyseal teeth are symmetric, nondenticulate, and large, whilst the most distal teeth are asymmetric, denticulate, and small (Hancock and Howse 1870, pl. 2, fig. 2;Brandt 1996, figs. 13b, c;Lund et al. 2014Lund et al. , p. 1020)).The genus needs revision, as it has become something of a `wastebasket' taxon for species having teeth that have extended, ridged lingual heels (Lund et al. 2014(Lund et al. , p. 1024)).Some of the nominal species of Janassa may need to be referred to other genera or even to other families.For example, J. korni has a dentition that differs from that of J. bituminosa in having denticulate crowns and a higher degree of heterodonty and might deserve generic distinction (Lund et al. 2014).A revision of Janassa is beyond the scope of the present work.
Peltodus (type species P. unguiformis Newberry and Worthen 1870) is known only from isolated teeth.The teeth resemble those of Janassa in having a blunt-tipped, thin crown which is convex labially and concave lingually.However, they lack an extended, ridged lingual heel, having only an `imbricated belt' (St. John and Worthen 1875, p. 410), and the angle between the base and the lingual heel is not close to 90°.Overall, the teeth of Peltodus differ so greatly from those of Janassa that there is no sufficient evidence to place them in the same family, much less to synonymise the genus with Janassa, as has been proposed (Woodward 1889).
Fissodus is known only from isolated teeth that resemble those of Peltodus but differ in having crowns that are cleft into two or three parts.St. John and Worthen (1875) described two species of Fissodus, F. bifidus and F. tricuspidatus, but did not designate a type species.The earliest designation of F. bifidus as the type species of Fissodus apparently was by Hay (1902).F. bifidus has a bifid crown, cleft into two acutely pointed lobes.As in the case of Peltodus, there is no justification for retaining Fissodus in the Janassidae since the morphology of the isolated teeth differs so much from that of Janassa teeth.The type species, F. bifidus, has an imbricated lingual band `composed of three or four or five narrow imbrications' (St. John and Worthen 1875, p. 414) that is much wider mesiodistally than high basoapically, in sharp contrast to J. bituminosa (Figure 1).F. pattoni Etheridge 1877, lacks basal imbricating ridges entirely and thus should not be referred to Fissodus because of the difference from the type species.Bifid lower symphyseal teeth of several petalodont species known from the Bear Gulch Limestone, if found isolated, would likely have been assigned to Fissodus, even though they represent several distinct genera and probably more than one family (Lund et al. 2014).Even if Fissodus, based on the type species, is a valid genus, the nominal species now assigned to that genus likely do not form a monophyletic group.None of the bifid teeth from the Bear Gulch Limestone have the same shape or proportions as those of the type species of Fissodus (Lund et al. 2014).
Cholodus St. John and Worthen 1875 (type species C. inaequalis St. John and Worthen 1875) was based on incomplete teeth representing only the apical portions of the crowns.These are bifid or trifid, as for Fissodus.Woodward (1889) pointed out similarities between Cholodus and Fissodus but did not formally synonymise the two genera.Others (Eastman 1903;Obruchev 1964;Zangerl 1981;Ginter et al. 2010) have asserted that Cholodus is a junior synonym of Fissodus.However, based on more complete teeth referred to C. inaequalis, Cholodus is distinct from Fissodus and is quite close to Janassa.One of these teeth, ClMNH 9078, has been figured by Hansen (1996, fig. 21-4.7, 21-4.8)reproduced here as Figure 2.This tooth has a ridged, lingual heel that is higher basoapically than wide mesodistally, a strongly sigmoid lateral profile, and an angle between the lingual heel and base of approximately 90°.In these respects, the tooth is like those of J. bituminosa and divergent from those of F. bifidus.Hence, Cholodus should be removed from synonymy with Fissodus and retained in the Janassidae.ClMNH 9078 has an acuminate crown that is not cleft.Based on new material, Hansen (1985, p. 530) believes that such non-cleft teeth belong to the same species as the bifid or trifid teeth, called Fissodus inaequalis in Hansen (1985) and Cholodus inaequalis in Hansen (1996).
Cypripediodens Duffin and Ward 2017 was assigned to the Janassidae by those authors.Terminology for parts of the teeth of Cypripediodens as used here, differing slightly from Duffin and Ward (2017), is shown in Figure 3a-b.The holotype of C. cristatus is shown in Figure 3c-d.Cypripediodens resembles Janassa in its strongly sigmoid lateral profile.However, it differs in other ways.A pointed apical cusp is present, called the `labial cusp' by Duffin and Ward (2017), and this cusp is broader in cross-section than the blunt-tipped, flattened, `petal-like' apical cusp of Janassa.The ridged, lingual heel, called `cristae' by Duffin and Ward (2017) is much shorter than in Janassa.A secondary cusp, called the `lingual cusp' by Duffin and Ward (2017) is present between the apical cusp and the lingual heel.The lingual face extends from the apical cusp to the basal end of the lingual heel.The labial face extends from the apical cusp to the crown-base junction.
In summary, the Janassidae are here presumed to comprise only three genera: Janassa, Cholodus, and Cypripediodens.

Institutional abbreviations
ALMNH: Alabama Museum of Natural History, Tuscaloosa, Alabama, USA; ClMNH: Cleveland Museum of Natural History, Cleveland, Ohio, USA; OGM: Orton Geological Museum, The Ohio State University, Columbus, Ohio, USA.

Miscellaneous abbreviations and acronyms
FAD: First Appearance Datum; GSSP: Global Boundary Stratotype Section and Point; Ma: 10 6 years ago.

General remarks
The locality where specimen ALMNH Paleo:20409 was found is designated as Locality 1.When the present study was nearly completed, it was discovered that other relevant specimens were already reposited at the ALMNH.Specimens ALMNH Paleo: 9780 and 9776 were found at locations designated as Localities 2 and 3, respectively.All three localities are in northern Alabama and are Chesterian (late Mississippian) in age (Figure 4).Information for Localities 2 and 3 was taken from the ALMNH records, and no field investigations were carried out.

Palaeogeographic setting
Figure 5 shows the palaeogeography of North America during the late Mississippian.Locality 1 is marked by a red dot.It lies within a region known as the Black Warrior Basin (Pashin and Gastaldo 2009) and is about equally remote from the Illinois Basin and from the Appalachian Basin.

Locality 1
Locality 1 is a limestone bed of the Bangor Limestone, near the shore of Little Bear Creek Reservoir, Franklin County, Alabama, USA.The holotype of Arcuodus multicuspidatus Itano and Lambert 2018 (ALMNH Paleo:9392) is from the same locality and horizon.The precise location is on file at the ALMNH.

Age
The Bangor Limestone in northwest Alabama is approximately 130 to 180 m in thickness and is Chesterian (late Mississippian) in age (Szabo et al. 1988;Thomas 1972Thomas , 1979)).Figure 6 shows the correlation of the Chesterian Regional North American Stage with standard subdivisions of the Mississippian Subsystem.The Chesterian is divided into three substages, from oldest to youngest: Gasperian, Hombergian, and Elviran.
An attempt was made to further refine the age of Locality 1 by conodont and echinoderm biostratigraphy.The bed at Locality 1 has previously been sampled for conodonts (Itano and Lambert 2018).Conodont P1 elements identified from the bed were Cavusgnathus regularis, C. unicornis (α), C. unicornis (β), and C. naviculus.Of these taxa, C. naviculus has the most restricted range (Abplanalp et al. 2009).These conodont taxa can be correlated with the Mississippian Faunal Units defined by Lane and Brenckle (2005).The base of Faunal Unit 11 is defined by the FAD of Cavusgnathus naviculus; the base of Faunal Unit 12 is defined by the FAD of Adetognathus unicornis.Cavusgnathus naviculus ranges into Faunal Unit 12.Given the small sample size of recovered conodont elements, the absence of A. unicornis is not sufficient evidence to rule out Faunal Unit 12. Thus, the conodont  Echinoderm macrofossils, including crinoids and blastoids, also can be used for correlation within the Mississippian (Ettensohn et al. 2007(Ettensohn et al. , 2009)).Biostratigraphically useful echinoderms have not been found at Locality 1.However, blastoids, including abundant Pentremites pyriformis and uncommon P. sulcatus and P. godoni have been found at a site about 550 m distant and stratigraphically approximately 15 m above Locality 1 (2022 email from L. White to me; unreferenced).Pentremites sulcatus is found only in the Hombergian Substage of the Chesterian Stage (Ettensohn et al. 2007(Ettensohn et al. , 2009)).P. pyriformis and P. godoni have wider ranges.In summary, the combined conodont and echinoderm evidence places Locality 1 in the Hombergian North American Regional Substage and the early Serpukhovian Stage.

Locality 2
According to the ALMNH records, Locality 2, where ALMNH Paleo:9780 was found, is at a roadcut at mile 125.5 on US Highway 72, near Woodville, Jackson County, Alabama, and is in the Monteagle Limestone.Locality 2 may be close to the locality of the chondrichthyan fossils reported by Stahl and Cicimurri (2005), who describe their locality as `an outcropping of the Monteagle Limestone along US 72 near Woodville, Jackson County, Alabama.'

Age
The Monteagle Limestone near Woodville, Alabama is early Chesterian (Gasperian) in age according to Haywick et al. (2016) and Kopaska-Merkel et al. (2020).The early Chesterian is correlated with the late Viséan Stage.In this region, the Monteagle Limestone is overlain by the Hartselle Formation, which is overlain by the Bangor Limestone.23.5) and, for regional substages of the Chesterian Regional Stage, Ettensohn et al. (2007, fig. 1;2009, fig. 11.1).`Age' is the geochronologic term for the equivalent chronostratigraphic `stage.'Bases of the Pennsylvanian Subsystem, Viséan Stage, and the Tournaisian Stage are defined by GSSPs.No GSSP has been defined for the base of the Serpukhovian Stage.The age shown is that of the base of the Tarusian Regional Stage in the Moscow Basin (Aretz et al. 2020).Placement of the base of the Hombergian Regional Substage near the base of the Serpukhovian Stage is from Ettensohn et al. (2007Ettensohn et al. ( , 2009)).Homb.= Hombergian Regional Substage in North America.

Locality 3
According to the ALMNH records, Locality 3, where ALMNH Paleo:9776 was found, is the Asphalt Quarry, south of Margerum, Colbert County, Alabama, 34° 44' 7.76" N, 88° 3' 24.22" W, in the Pride Mountain Formation.Presumably, this locality is the same as the one called the Alabama Rock Asphalt Quarry by Pashin et al. (1993) and the Asphalt Rock Quarry by Hassler and Stock (1993)

Age
The Pride Mountain Formation in northwest Alabama is roughly equivalent in age to the Monteagle Limestone in northeast Alabama according to Haywick et al. (2016) and Kopaska-Merkel et al. (2020).Thus, it is thought to be early Chesterian (Gasperian) in age and correlated with the late Viséan Stage.However, recovery of the conodont Cavusgnathus naviculus (Hinde) from one of the limestone units of the Pride Mountain Formation at this locality (Hassler and Stock 1993) indicates that at least part of the Pride Mountain Formation at Locality 3 is in either Faunal Unit 11 or 12 of Lane and Brenckle (2005), which is Hombergian (middle Chesterian).Since the stratigraphic position at which ALMNH Paleo:9776 was found is unknown, its age is uncertain within the interval from early to middle Chesterian (late Viséan to early Serpukhovian).In this region, the Pride Mountain Formation is underlain by the Tuscumbia Limestone and is overlain by the Hartselle Formation, which is overlain by the Bangor Limestone.

Materials and methods
All material collected from Locality 1, including several janassid teeth not figured herein, are reposited at the ALMNH.Matrix was removed mechanically and by immersion in 5% acetic acid.No information is available on the preparation of material from Localities 2 and 3.
The 3D model of ALMNH Paleo:20409 (Figure 7) was generated from 144 focus-stacked photographic images acquired from various viewpoints, by use of the computer programme Agisoft Metashape Professional Edition, version 1.84, build 14,856 (64 bit).The focus-stacked images were generated internally by an Olympus TG-4 camera.The full version of the model (1739040 faces, 87.0 MB) is available in STL (Standard Triangle Language) file format, on the Zenodo data repository, as supplementary material.The version displayed as Figure 7 has been reduced to 180000 faces.

Included species
Only the type species.

Derivation of name
From Latin cavus = hollow, for the labial concavity displayed by the teeth; ancient Greek ὀδούς = tooth.

Diagnosis
Teeth strongly sigmoid in lateral view.Orally directed apical cusp.Presumed near-symphyseal teeth are wedge-like in lateral view.Approximately 5-7 U-shaped cristae on the basal end of the lingual face.The more strongly curved cristae include a narrow, basally directed lobe.Apical cusp separated from the series of cristae by a central trough.Labial face displays a longitudinal concavity.Short base attached to labial face, directed oppositely to apical cusp.Presumed lateral teeth differ from the presumed near-symphyseal teeth in having an apical cusp that is thinner linguolabially, a shorter central trough, less strongly curved cristae, and a broader, shallower labial concavity,

Remark
With the assignment of Cavusodus to the Janassidae, the family comprises four genera: Janassa, Cholodus, Cypripediodens, and Cavusodus.However, revision of Janassa might increase this number.

Derivation of name
In honour of Mr. Larry White of Russellville, Alabama, USA, collector of several significant chondrichthyan fossils from the Bangor Limestone, including the holotype of Cavusodus whitei.

Type horizon
Hombergian North American Regional Substage = early Serpukhovian Stage.

Diagnosis
As for genus.

Description
The teeth are strongly sigmoid in lateral view.The crown, in lingual view, is elongate, with a length about twice the width.The crown of the holotype is 14 mm long; the maximum width is 8 mm; the height of the apical cusp is 4 mm.In lateral view, the apical cusp of the holotype has the shape of a blunt wedge, directed orally.Five to seven cristae are present at the basal end of the lingual face.The cristae are U-shaped, oriented with the concave opening directed towards the apical cusp.The cristae are not uniformly curved; on the holotype the more basally located ones feature a more or less sharply curved, basally directed lobe.On the holotype, the central trough between the apical cusp and the most-apical crista is oval-shaped and is longer than wide.The central trough is not uniformly concave but has a broad, longitudinal, orally directed bulge having a width of about one-third of the width of the crown.On the holotype, a deep, narrow groove having a width about one-third that of the tooth is present on the labial face.The base, which is preserved on the referred specimen, ALMNH Paleo:9780, but not on the holotype, is short and is similar in shape and orientation to the base of Cypripedioens cristatus.It projects basally and aborally from the basal end of the labial face.
ALMNH Paleo:9780 is similar in overall shape and dimensions to the holotype, ALMNH Paleo:20409, but differs in minor ways.The lingual heel, demarcated by the presence of the cristae, is broader mesodistally than the apical region of the crown, unlike in the holotype.ALMNH Paleo:9776 differs in several ways from the holotype.The crown is thinner labiolingually, the central trough is shorter, and the lingual cristae are less strongly curved.Also, the labial concavity is broader and shallower.

Remarks
The three specimens referred to Cavusodus whitei differ from one another but are presumed to belong to the same species.In particular, ALMNH Paleo:9776 differs from ALMNH Paleo:20409 and 9780, which are similar to each other.ALMNH Paleo:9776 might have belonged to a more distal tooth file than the other two specimens.Alternatively, it might belong to a different species, but until more specimens are found, so that the extent of heterodonty in Cavusodus whitei can be determined, it is not possible to draw a conclusion.
The longitudinal concavity on the labial face, present on all specimens of Cavusodus whitei, might have allowed for close contact between one tooth and the retained, next-older tooth, the apical cusp on the older tooth fitting into the concavity of the younger tooth.The fact that the concavity consists of a longitudinal groove rather than a simple indentation might indicate that the younger tooth moved longitudinally relative to the older tooth before being fixed into position.No concavity is present on the labial face of Cypripediodens cristatus (2022 email from C. Duffin to me; unreferenced).

Discussion
Cypripediodens cristatus and Cavusodus whitei are similar to each other and distinct from all other janassids, especially in the more robust form of the teeth, as viewed in lateral profile.They might comprise a subclade within the Janassidae.However, they also differ in several ways from each other, which justifies their being placed in different genera.Cavusodus whitei lacks the secondary cusp that is present in Cypripediodens cristatus.Cypripediodens cristatus lacks the labial concavity that is present in Cavusodus whitei.The lingual cristae are more prominent on Cavusodus whitei than on Cypripediodens cristatus.
The type horizon of Cavusodus whitei is younger than the known range of Cypripediodens cristatus (early Serpukhovian vs. late Viséan).However, the specimens referred to Cavusodus whitei from the Monteagle Limestone (ALMNH Paleo:9780) and from the Pride Mountain Formation (ALMNH Paleo:9776) might be older than the holotype of Cavusodus whitei, so that the stratigraphic ranges of Cavusodus whitei and of Cypripediodens cristatus might overlap.The age determinations of ALMNH Paleo:9780 and 9776 are hampered by lack of conodont or other biostratigraphic information for the beds in which they were found.

Figure 1 .
Figure 1.Various views of teeth of Janassa bituminosa.(a) Lingual face, with apex of crown at top and base at bottom; (b) lateral view, showing sigmoid profile; (c) articulated dentition of lower jaw in oral (occlusal) view; (d) terms used for orientations of janassid teeth, indicated on a tooth of Janassa bituminosa.Jaekel (1899, pl.14).(e) Symphyseal files of teeth of Janassa bituminosa, upper jaw to the right, lower jaw to the left, showing tooth replacement pattern.Largest teeth in each file in either jaw are the youngest and only functional teeth.Smaller and older teeth form a supporting structure for the functional teeth.After Hancock and Howse (1870)., pl. 2, fig. 3.

Figure 4 .
Figure 4. (a) Outline map of the United States with the state of Alabama highlighted.(b) Map of northern Alabama with counties outlined.Localities 1, 2, and 3 marked with dots.Holotype of Cavusodus whitei was found at Locality 1 in Franklin County.

Figure 5 .
Figure 5. Late Mississippian palaeogeographic map overlaid on outline map of North America.Locality 1, where the holotype of Cavusodus whitei was found, is indicated by red dot.Eq = Paleoequator; BWB = Black Warrior Basin; IB = Illinois Basin; AB = Appalachian Basin.Map modified from R. Blakey, Colorado Plateau Geosystems, Inc.

Figure 6 .
Figure 6.Correlation chart of standard and North American regional chronostratigraphic subdivisions of the Mississippian Subsystem, modified fromAretz et al. (2020, fig.23.5) and, for regional substages of the Chesterian Regional Stage,Ettensohn et al. (2007, fig.1; 2009, fig.11.1).`Age' is the geochronologic term for the equivalent chronostratigraphic `stage.'Bases of the Pennsylvanian Subsystem, Viséan Stage, and the Tournaisian Stage are defined by GSSPs.No GSSP has been defined for the base of the Serpukhovian Stage.The age shown is that of the base of the Tarusian Regional Stage in the Moscow Basin(Aretz et al. 2020).Placement of the base of the Hombergian Regional Substage near the base of the Serpukhovian Stage is fromEttensohn et al. (2007Ettensohn et al. ( , 2009)).Homb.= Hombergian Regional Substage in North America.