Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2024 Dec 16;308(8):2160–2172. doi: 10.1002/ar.25612

Reassessment of Xenodens calminechari with a discussion of tooth morphology in mosasaurs

Henry S Sharpe 1,✉, Mark J Powers 1, Michael W Caldwell 1
PMCID: PMC12239698  PMID: 39682068

Abstract

Xenodens calminechari is a mosasaurid taxon named by Longrich et al. (2021) based on the holotype MHNM.KH.331, a left maxilla with several teeth. This holotype was obtained nonscientifically (without technical supervision) from an area in Morocco that yields many manipulated or forged specimens. Examination of Longrich et al. (2021) reveals four tooth crowns occupy what appear to be two alveoli in MHNM.KH.331, and there is potential adhesive connecting the tooth crowns to the maxilla on their lateral sides. We argue that the articulated tooth crowns of this taxon were artificially placed in the maxilla, rendering at least two apomorphies of this taxon the product of forgery. Longrich et al. (2021)'s claims of fused tooth ‘roots' in MNHM.KH.331 are instead calcified periodontal ligament and alveolar bone that have ankylosed, resembling the typical mosasaurid condition. Differing tooth crown morphology does not preclude the referral of the teeth of this specimen to a younger ontogenetic stage of another mosasaur (possibly Carinodens) because many extant lizard species show drastic ontogenetic changes in the dentition. We argue that Xenodens calminechari represents a nomen dubium. This specimen constitutes a confluence of two persistent problems in vertebrate paleontology: material sourced from commercial excavations that has not been adequately tested for forgery, and taxa named from tooth‐based holotypes that ignore the potential for intraspecific dental variation and interspecific convergence in dental characters, as are common in squamates. We suggest that Longrich et al. CT scan MHNM.KH.331, and we supply CT examples for identifying artificially added tooth crowns to Moroccan mosasaur jaws. Finally, we provide recommendations for the designation of mosasaurid holotypes.

Keywords: dentition, mosasaur, paleontology, squamate, taxonomy

1. INTRODUCTION

The aims of both paleontological systematics and hypotheses built thereupon are complicated by holotypes that do not properly diagnose the taxa they represent. One way this can arise is from holotypes that do not present a complete enough suite of characters to differentiate between closely related species that share many overlapping morphologies, leading to subsequent issues in properly resolving taxonomy (e.g., Coombs Jr, 1988; Evans et al., 2017). Numerous fossil vertebrate species have been established based on holotype material consisting of tooth crowns (e.g., Arambourg, 1952; Cope, 1876; Leidy, 1856) or tooth crowns associated with a small amount of nontooth skeletal material that does not differ from other known species in morphology (e.g., Fox, 1969; Longrich et al., 2023). We term these type specimens, for which all or the vast majority of unique character combinations are limited to the dentition, “tooth‐based holotypes.”

Complications can also arise in holotypes from fabrication of diagnostic elements or combining multiple specimens into one. Such methods of forgery are sometimes performed by fossil collectors and dealers and are frequently recognized only after the publication of the specimen as though it were a legitimate biological individual (e.g., Rossi et al., 2024; Rowe et al., 2001). A minor but common modification among forged specimens is the addition of tooth crowns to tooth‐bearing elements; both commercial collectors/dealers and early museum paleontologists are responsible for this (e.g., UALVP 10, a Gorgosaurus skull collected by George F. Sternberg and J. A. Allan for the University of Alberta in 1921, had potentially‐unrelated tyrannosaurid teeth manually placed in its jaws to better complete the skull). It is therefore crucial for paleontologists describing specimens, either from museum collections or commercial acquisition, to carefully ascertain the origins of each constituent element.

Longrich et al. (2021) described MHNM.KH.331, an isolated mosasaurid left maxilla and teeth (Figure 1a,c) from the Sidi Chennane phosphate mine, Morocco, that likely originated from the upper Maastrichtian strata of Couche III. The specimen was not collected under scientific conditions (i.e., the recording of locality data such as GPS, elevation, and formal strata assessment) and was obtained by the authors after excavation and collection by mine workers. It is extremely common among Moroccan fossils found this way to be modified to increase monetary value. This can range from the addition of tooth crowns into unoccupied tooth positions, to the complete fabrication of a specimen. Longrich et al. (2021) described MHNM.KH.331 as the holotype of Xenodens calminechari, a small‐bodied mosasaur with an unusually shark‐like dentition compressed into a saw‐like tooth row. The dentition of Xenodens calminechari was described as “not just unique among mosasaurids, but among tetrapods” (Longrich et al., 2021, p. 10), a rather extraordinary claim that warrants closer examination given the high frequency of forged fossils from the type locality (e.g., Figures 1b,d,e and 4).

FIGURE 1.

FIGURE 1

Open in a new tab

(a) MHNM.KH.331, holotype of Xenodens calminechari in lateral view in matrix, photograph adapted from Longrich et al., 2021; (b) fake mosasaurid jaw (UALVP unlisted) from Morocco showing tooth crowns artificially attached to unrelated bone fragments, ‘in matrix’; (b) close‐up of tooth crowns in lateral view of MHNM.KH.331, photograph adapted from Longrich et al. (2021); (d) close‐up of (c) showing tooth crown‐‘jaw’ intersection; (e) close‐up of Halisaurus arambourgi UALVP 56123 showing forged tooth‐jaw intersection.

FIGURE 4.

FIGURE 4

Open in a new tab

CT scan example for recognizing falsely attached teeth in mosasaur fossils from Morocco. (a, b) Anterior portion of left dentary of Halisaurus arambourgi UALVP 56123 from type locality of Xenodens calminechari in (a) lateral and (b) dorsomedial views; (c) sagittal CT scan of UALVP 56123 showing mixture of untampered and falsely‐placed teeth; (d) coronal CT scan of UALVP 56123 6th dentary tooth showing true placement of this tooth in this alveolus; (e) coronal CT scan of UALVP 56123 5th dentary tooth showing false placement of a shed tooth over a potentially‐unrelated articulated mineralized periodontal ligament, as evident from the lack of root tissue intruding into the ligament; (f) coronal CT scan of UALVP 56123 4th dentary tooth showing false placement of broken tooth crown in empty alveolus filled with adhesive; (g) coronal CT scan of UALVP 56123 2nd dentary tooth showing false placement of broken tooth crown in partially‐empty alveolus (some nonresorbed ligament still present); (h) illustration of correct tooth implantation and attachment in coronal view, modified from (d), Caldwell (2007), and LeBlanc et al. (2017); (i) illustration of tooth crown falsely placed on toothless ligament with broken‐off crown; (j) illustration of tooth crown falsely placed on root and ligament, modified from (e); (k) illustration of tooth crown falsely placed on an empty alveolus, modified from (f).

Institutional abbreviations: FMNH, Field Museum of Natural History, Chicago, Illinois, USA; MHNM, Muséum d'Histoire naturelle de Marrakech, Université Cadi Ayyad, Morocco; MNHN, Muséum National d'Histoire Naturelle, Paris, France; OCP, Office Chérifien des Phosphates, Khouribga, Morocco; UALVP, University of Alberta Laboratory for Vertebrate Paleontology, Edmonton, Alberta, Canada.

2. METHODS

Longrich et al. (2021) included several, high‐detail photographs of MHNM.KH.331, inclusive of the dentition. We compared these photographs to mosasaurid and teiid jaws in the UALVP collections that preserve tooth implantation and replacement data. We also compared published photos of MHNM.KH.331 to a previously purchased fake mosasaurid jaw (UALVP unlisted) (Figure 1b,d) that was produced in, and exported from, Morocco; this specimen consists of several pieces of likely unrelated bone compiled together into a roughly dentary‐shaped element, to which several mosasaurine teeth have been manually adhered, providing one example of forgery from a similar locality. UALVP 56123 (cf. Halisaurus arambourgi) was purchased from Morocco and likely originated (as with MNHM.KH.331) from Couche III; subsequent CT scanning revealed that several tooth crowns had been manually placed in the specimen, providing another example of forgery from a similar locality. Scans for UALVP 56123 were performed at the University of Alberta in a Nikon XT H 225 ST Helical CT Scanner at 130kv with a resolution of 16 microns to examine tooth attachment and replacement in mosasauroids.

3. TOOTH ARRANGEMENT OF MHNM.KH.331

In mosasaurids, individual alveoli each bear a single tooth consisting of an enamel and dentine crown, a dentine root portion, and a large amount of calcified periodontal ligament attached to the cement on the root, and to the alveolar bone forming the socket (Caldwell, 2007; Caldwell et al., 2003; LeBlanc et al., 2017) (Figure 2d–e). This alveolar bone not only comprises the “interdental plates” between adjacent alveoli but also lines the remainder of the socket labially, lingually, and ventrally (Caldwell et al., 2003). As new teeth develop and move toward the tooth row, resorption pits are created in the mineralized periodontal ligament posteromedial to the current tooth crown (Figure 2d,e; Caldwell, 2007; Rieppel & Kearney, 2005), superficially resembling typical lizard tooth replacement where the resorption pit invades the actual dentine root of the tooth. The relationship between tooth crowns, alveolar tissue, and replacing teeth is 1:1:1, as can be neatly observed in extant teiids and all known mosasaurine, plioplatecarpine, and tylosaurine mosasaurs (Figure 2a–e). The exception is when a tooth is new and resorption of the ligament has not yet begun, during which time there is a conspicuous gap beneath said tooth in the series of resorption pits.

FIGURE 2.

FIGURE 2

Open in a new tab

Tooth replacement in teiid and mosasaurid lizards; fully formed and incipient resorption pits are marked by arrows. (a) Dental arcade in internal view of Tupinambis teguixin, FMNH 140193; (b) dental arcade in internal view of Dracaena sp., FMNH 207657; (c) close‐up of (b) showing multiple confluent, but still distinct, resorption pits; (d) mirrored left dentary in medial view of Globidens simplex, MHNM.KHG.221; (e) dentaries in dorsal view of Platecarpus tympaniticus, UALVP 55497.

A minimum of 13 alveoli are allegedly present in MHNM.KH.331 (Figure 3a; Longrich et al., 2021; Figure 4b). Four tooth‐crown‐bearing teeth are seemingly preserved in articulation, superficially appearing to occupy the 9th–12th alveoli following Longrich et al. (2021); the rest of the teeth are preserved as either empty alveoli or ligament masses with in situ roots from which the crown has been broken off (Figure 3A). Longrich et al. (2021, p. 5) claim “Teeth are closely packed, with a slight gap between the anterior carina of one tooth and the posterior carina of the other.” The alveoli lacking attached tooth crowns of MHNM.KH.331 (alveoli 1–8) do not show this tight packing (Figure 3a) and are similarly spaced compared to those of other mosasaurs (Figure 2d,e), and each bears a single pulp cavity and resorption pit when a ligament and root are preserved (Figure 3a). Only the preserved tooth crowns show this “closely packed” arrangement (Figure 3a); to remedy this disjunct, Longrich et al. (2021) reconstructed a drastic change in tooth crown size between the anterior teeth and teeth 9–12. However, the four “articulated” tooth crowns only show two resorption pits proximally (Figure 3c), strongly suggesting that only two teeth should be present here. This spacing of only two teeth as indicated by resorption pitting matches the tooth spacing anterior to this point in the maxilla. This disjunct does not result from the joining of adjacent resorption pits, as the occurrence of this in extant lizards results in anteroposteriorly long pits that span multiple tooth positions and still retain internal evidence of individual tooth divisions (Figure 2c). Longrich et al. (2021: 2, 10)'s proposed tooth arrangement of “teeth closely packed to form a saw‐like cutting edge”, “not just unique among mosasaurids, but among tetrapods” is thus incorrectly diagnosed.

FIGURE 3.

FIGURE 3

Open in a new tab

Left maxilla of Xenodens calminechari, MHNM.KH.331: All photographs adapted from Longrich et al. (2021). (a) Maxilla in ventral view, showing irregular proposed tooth spacing in posterior tooth row; (b) close‐up of (a) showing nonfusion of tooth roots due to separation of alveoli (marked by asterisks) by alveolar bone; (c) maxilla in medial view showing a disjunct arrangement of tooth crowns and resorption pits (which indicate alveolar spacing); (d) close‐up of tooth crowns in medial view showing the irregular junction of tooth crowns and basal tissue; (e) close‐up of tooth crowns in lateral view showing potential adhesive connecting tooth crowns to maxilla.

The problematic arrangement of four tooth crowns in two alveoli in MHNM.KH.331 must result from one of two sources: in vivo dental pathology, or post‐excavation modification of the specimen. One specimen of Carinodens belgicus (OCP DEK/GE 455) apparently shows the same condition for the 9th and 10th tooth crowns (Schulp et al., 2009, Figures 1d, 2d, and 3d). This condition does not affect other alveoli in described Carinodens tooth‐bearing elements (Schulp et al., 2009). Of note, OCP DEK/GE 455 was a nonscientifically collected specimen (Schulp et al., 2009) and may have been modified in the manner described below.

In Longrich et al. (2021) (Figure 5), the attachment of the tooth crowns is photographed in detail. In lateral view, the tooth crowns appear to be joined to the maxilla by a gummy, paste‐like material that is smeared to the ventrolateral surface of the maxilla (Figure 3e). We interpret this material as a likely adhesive meant to manually attach the tooth crowns to the maxilla. Similar material is visible in both faked Moroccan mosasaur jaws and real jaws to which potentially unrelated teeth have been manually added (Figure 1b,d,e). This material is visible in Longrich et al. (2021)'s photograph of MHNM.KH.331 in its supposedly original matrix. A frequent modification of commercially‐collected mosasaurid fossils from Morocco is the attachment of tooth crowns to elements that may or may not be tooth‐bearing or belong to the same species (Figure 1b,d,e). This can be done to fossils both free of (Figure 1e) and embedded in (Figure 1d) “matrix,” so photographic evidence of MHNM.KH.331 in matrix (Figure 1a) is not proof of the absence of this kind of modification. In medial view, this material is not present, and material from mineralized periodontal ligament medially overlays the tooth crowns (Figure 3d), unlike typical mosasaur tooth attachment in which the periodontal tissue is positioned distinctly below the tooth enamel. The second preserved tooth crown arises directly ventral to a resorption pit, unlike the typical mosasaurid condition (Caldwell, 2007; Rieppel & Kearney, 2005), and the tooth crown anterior to it is correctly located anterolateral to the respective resorption pit. We interpret this as an indication of the tooth crowns likely being manually placed on the articulated mineralized periodontal ligament and glued on using adhesive applied only to the lateral tooth root–crown junction. The first and third preserved tooth crowns are located in correct positions, and so may represent true associations of tooth crowns to this maxilla, but possible adhesive material and medial overlap of mineralized periodontal ligament over these crowns still indicate possible forgery. The apparent oblique orientation of the tooth crowns relative to the jaw bone, another diagnostic character for Xenodens calminechari, could therefore result from the angle at which the teeth have been glued to the jaw.

FIGURE 5.

FIGURE 5

Open in a new tab

(a, b) Tooth crown outlines of MHNM.KH.331 in dorsal (top) and lateral (bottom) views, adapted from Longrich et al. (2021); (c–f) tooth crowns outlines of Carinodens in dorsal (top) and lateral (bottom) views, adapted from Bardet et al. (2008); (c) OCP.DEK/GE 445; (d) MNHN 6314; (e) MNHN 6340; (f) OCP.DEK/GE 446; (g) ontogenetic changes in the tooth crown outlines of Varanus niloticus in left mediolateral view, adapted from D'Amore (2015); (h–j) ontogenetic changes in the tooth crown outlines of iguanian and teiid lizards in right medial view, adapted from Estes and Williams (1984) (root‐crown junctions suggested by dotted lines, as these are not clearly demarcated in the original figures). In each case, tooth crowns are oriented downward, arrows indicate the direction of ontogeny when present, and (g–j) teeth are sampled from presumed homologous tooth positions based on alveolar count from the posterior margin of the tooth‐bearing element.

Given the presence of four tooth crowns on two alveoli, the apparent signs of manual adhesion of tooth crowns to this specimen, and the high frequency of forged specimens originating from the type locality, we believe that the closely packed tooth arrangement and obliquely oriented tooth crowns of MHNM.KH.331 most likely result from forgery. One method that could test this hypothesis is computed tomography (CT) scanning of MHNM.KH.331. True complete and articulated teeth versus artificially placed teeth show distinctly different CT results, as can be visualized by CT scans of UALVP 56123 (Figure 4). When a tooth is correctly implanted in the tooth‐bearing element, the tooth root and pulp cavity extend into the mineralized periodontal ligament internally, which in turn radiates outward to contact the alveolar bone lying in the tooth‐bearing element (Figure 4d,h). When crowns have been placed artificially, the root is missing (shed or broken teeth being placed; Figure 4e–g), the root does not extend into the ligament at all (Figure 4e,i,j), or the ligament is partially or completely absent beneath the tooth (Figure 4f,g,k). We suspect that CT scanning of MHNM.KH.331 will reveal a similar situation to Figure 4e,i,j, in which the tooth crowns have been manually placed on unrelated roots. In addition, CT scanning may reveal mid‐formation replacement teeth that could elucidate a proper tooth morphology of this specimen.

4. TOOTH MORPHOLOGY OF MHNM.KH.331

Longrich et al. (2021, p. 4) stated that: “Tooth roots are mediolaterally compressed and expanded anteroposteriorly so that adjacent roots contact each other and fuse to create a wall of bone supporting the teeth. This configuration is unique among squamates, with the exception of Carinodens, which shares these fused tooth bases (Schulp et al., 2009).” This description of tooth attachment by Longrich et al. (2021) is incorrect and demonstrates a lack of current understanding of the literature regarding mosasaurid dental histology. First, the anteroposteriorly expanded ‘tooth roots’ as identified by Longrich et al. (2021) are not roots but rather are mineralized periodontal ligaments (see Caldwell et al., 2003; LeBlanc et al., 2017). The actual dentine root of a mosasaur tooth is never expanded, and confusing the true root with the periodontal ligament mass is anatomically and histologically erroneous. Secondly, alveolar bone can clearly be seen separating the anterior alveoli between the shed teeth (Figure 3b). The periodontal ligament masses therefore do not form a single fused mass; this superficially appears to be the case in the posterior teeth of MHNM.KH.331 due to poor resolution between the alveolar bone and calcified/mineralized periodontal ligament. In mosasaur dental attachment the final stage of ankylosis can increase the mass of the calcified periodontal ligament and thus blur the boundary between the calcified ligament and the exposed margin of the alveolus (i.e., the interdental plate) formed by the alveolar bone (Caldwell, 2007; Caldwell et al., 2003; LeBlanc et al., 2017; Luan et al., 2009). Similar poor resolution between the tooth bases and the alveolar bone (resulting occasionally in superficially “fused” tooth bases) is also observable in several living teiid lizards (Figure 2a–c); therefore, even this condition cannot be described as unique among squamates. An understanding of the correct anatomical and histological terms and tissues associated with dental attachment in squamates specifically, and vertebrates generally, is not present anywhere in the language, assessment/analysis, and conclusions as presented in Longrich et al. (2021). This has no doubt contributed to the use of a probable dental forgery as the basis for diagnosing and describing a supposedly unique fossil mosasaur genus.

The tooth crowns of MHNM.KH.331 are different from those observed for most mosasaurs, and do vaguely resemble shark teeth in general shape, though the lack of serrations suggests a nonshark identity (L. Nelson, pers. comm.). The crowns closely resemble some of those referred to Carinodens in lateral view (Figure 5a–c), diverging in their smooth enamel and the extreme labiolingual compression of the crown in MHNM.KH.331 compared to Carinodens tooth crowns, which are more labiolingually swollen (Bardet et al., 2008; Dollo, 1913). In profile, they are also not unlike those of a number of mosasaur sister groups, e.g., Coniasaurus crassidens (Caldwell & Cooper, 1999). However, for the sake of clarity within mosasaurs, we will focus here on Carinodens. The teeth of MHNM.KH.331 also diverge from Carinodens in their size; they are roughly half the size of those described as Carinodens, originating from a smaller individual based on tooth‐bearing element size (Figure 5a–f; Bardet et al., 2008; Dollo, 1913; Longrich et al., 2021; Schulp et al., 2009). A series of tooth crowns identified as Carinodens from the same locality (Bardet et al., 2008) show an interesting trend. The smallest (Figure 5c) have a nearly identical lateral profile to those of MHNM.KH.331, but are larger, have wrinkled enamel, and are slightly more labiolingually swollen. Larger teeth (Figure 5d) are more labiolingually swollen but retain a similar lateral profile to those of MHNM.KH.331. Finally, the largest teeth (Figure 5e–f) are similar to the Carinodens belgicus type in all views. Bardet et al. (2008) identified these smaller, morphologically different teeth as anterior teeth; however, apparently articulated anterior teeth known from Carinodens are rounded and peglike in lateral view (Schulp et al., 2009, Figure 1c), resembling the anterior teeth of Globidens.

Longrich et al. (2021) differentiated Xenodens calminechari from Carinodens based on the shape and texture of the tooth crowns. Unlike other skeletal elements, vertebrate teeth are continuously generated anew throughout an individual's ontogeny, and so may present different morphologies depending on an individual's ontogenetic stage (Davit‐Béal et al., 2006). Many extant lizard species have similar molariform teeth to Carinodens and Globidens. Among these species, some have ontogenetically invariant molariform tooth morphologies (Estes & Williams, 1984), although other juveniles possess peglike, bladelike, or multicuspid teeth that are replaced by molariform teeth through ontogeny (Figure 5g–j; D'Amore, 2015; Estes & Williams, 1984; Mertens, 1942). These changes can happen gradually, as subsequent tooth generations acquire more swollen morphologies (Figure 5g). Additionally, textural striations on tooth crowns (such as those used by Longrich et al., 2021, to differentiate Carinodens spp. from X. calminechari) have also been observed to be present in some lizards with molariform teeth but not juveniles of the same species (Figure 5i). As tooth crown morphology is known to be frequently intraspecifically variant (correlating with ontogeny) in extant lizards and given the current absence of an ontogenetic series for X. calminechari showing stable dental morphology, the dental characters listed by Longrich et al. (2021) are not adequate to diagnose X. calminechari as a taxonomic unit.

5. TAXONOMY OF MHNM.KH.331

There is a series of intermediate morphologies between the tooth crowns of MHNM.KH.331 and Carinodens from the same locality, and both taxa share anteroposteriorly elongated resorption pits. MHNM.KH.331 is also smaller than the described Carinodens material; while Longrich et al. (2021) identify small body size as a diagnostic trait of this taxon, they provided no assessment of the ontogenetic stage of MHNM.KH.331. Juvenile vertebrates are ubiquitously smaller than their respective adults, so the use of body size as a taxonomic character in the absence of any ontogenetic statement by Longrich et al. (2021) reflects a poor understanding of vertebrate growth. There is also a lack of overlapping nondental material between MHNM.KH.331 and the described Carinodens material. Given these arguments, we assert that the potentially chimeric material of Xenodens calminechari cannot be adequately ruled out as an earlier ontogenetic stage than those currently known from Carinodens.

The original diagnosis for Xenodens calminechari contains 13 characters (Longrich et al., 2021, characters listed in Diagnosis, here numbered 1–13 based on original listed order) which we consider: ontogenetically variable in most, if not all known vertebrates (1), impossible to compare with Carinodens due to nonoverlapping material (2, 3, 4); ontogenetically variable in numerous living squamates with similar dental morphology (5, 6, 7, 8, 9); likely product of forgery (10, 11); product of misidentification (12); and identical to Carinodens spp. (13). The holotype material of X. calminechari is likely a forgery, the supposedly unique tooth attachment mode was misidentified, and the constituent elements are possibly juvenile and cannot be adequately distinguished from Carinodens spp.; we therefore argue that Xenodens calminechari should be considered a nomen dubium.

We therefore consider Xenodens calminechari to be exemplary of two current issues in mosasaurid paleontology. Firstly, mosasaurids described from commercially collected fossils, especially from Morocco, may be more vulnerable to forgery, in addition to the recognized loss of stratigraphic data for these specimens (LeBlanc et al., 2012; LeBlanc et al., 2019). Forgeries of this nature, as demonstrated here, have the potential to influence taxonomic characters and the naming of new species. Extreme care should be taken in the description and peer‐review of such material, especially if the relevant morphologies involve features known to be frequent subjects of fakery (e.g., tooth‐maxilla attachment).

Secondly, Xenodens calminechari was named based on extremely incomplete material (a partial maxilla and four tooth crowns; Longrich et al., 2021), as was Carinodens belgicus (a tooth crown; Woodward, 1891), Carinodens fraasi (a dentary with three tooth crowns; Dollo, 1913), and Carinodens minalmamar (two dentaries with two tooth crowns; Schulp et al., 2009). Carinodens palistinicus was described based on relatively complete cranial material and some postcranial material, though the available data from the cranium is limited due to the preservation of the specimen (Kaddumi, 2009). Recommendation 73A of the International Code of Zoological Nomenclature states: “An author who establishes a new nominal species‐group taxon should designate its holotype in a way that will facilitate its subsequent recognition” (ICZN, 1999). The maxilla of MHNM.KH.331 does not present any unique combination of characters among mosasaurs (Longrich et al., 2021) and is possibly unrelated to the attached tooth crowns, leaving only the tooth crowns as diagnostic features. Regardless of the true association of the maxilla to the tooth crowns, Longrich et al. (2021) identified the tooth crowns and their arrangement as the most diagnostic features of this taxon.

Teeth as the exclusive or primary character‐bearing element in a holotype (e.g., Cope, 1876; Leidy, 1856) often lead to taxonomic issues later as tooth morphology is very often identical among closely related species (e.g., Coombs Jr, 1988; Evans et al., 2017). Strong et al. (2020) recently demonstrated this concept of identical dentition among nonsister mosasaurid species, citing Massare (1987)'s argument for the likely homoplastic nature of mosasaurid tooth evolution due to convergent feeding strategies leading to false phylogenetic signals. It is probable that both convergent feeding strategies and close relationships could lead to virtually indistinguishable dentitions in distinct mosasaurid species: teeth, like any other skeletal element, are under no obligation to evolve one morphotype in total coincidence with only a single species. Extreme changes through ontogeny in squamate tooth morphology (D'Amore, 2015; Estes & Williams, 1984) also preclude these elements from adequately diagnosing species on their own. Finally, many squamates exhibit some degree of heterodonty, rendering a small sample of teeth incapable of capturing the entire range of dental variation within even a single individual, let alone an entire species.

Tooth‐based holotypes could very well include other species within their hypodigms, fail to capture intracranial variation, and/or split individuals of different ontogenetic states from a species established using one ontogenetic stage (Figure 6). Tooth‐based holotypes therefore do not facilitate subsequent recognition of the taxon they represent (and thus do not adhere to ICZN Recommendation 73A; ICZN, 1999). Although a larger‐scale analysis of the failure of many paleontological species to meet this recommendation, and thus their poor taxonomic utility, is outside the bounds of this contribution, we argue that most of the existing species of Carinodens, Xenodens calminechari, and other tooth‐based holotypes among Mosasauridae (e.g., Arambourg, 1952; Longrich et al., 2023) are known from nondiagnostic types.

FIGURE 6.

FIGURE 6

Open in a new tab

Issues with tooth‐based holotypes in vertebrate paleontology, using three hypothetical species of mosasaurids as an example. Holotype 1 was established using an anterior adult tooth of species 2, and holotype 2 was established using a juvenile posterior tooth from species 1. Neither holotype adequately represents the anatomy of its target taxon.

6. CONCLUSIONS

Based on this case study, we recommend the following:

  1. Xenodens calminechari be considered a nomen dubium.
    1. It is potentially chimeric, with tooth crowns attached to a possibly unrelated maxilla. This issue could be resolved by CT scanning MHNM.KH.331 to determine if any tooth crowns actually belong to the maxilla.
    2. Its diagnostic material (tooth crowns) does not allow for adequate anatomical comparison with other Mosasauridae.
    3. There is a strong likelihood that most diagnostic features for this taxon are the result of immaturity and forgery.

  2. Mosasaurid species named from commercially collected fossils should be treated with extreme caution and precautions (e.g., CT scanning) should be taken to recognize forgeries prior to publication.

  3. Mosasaurid species should be named from holotypes that include enough material to properly compare between mosasaurid species (e.g., complete or near‐complete skulls and representative postcrania).
    1. Tooth morphology can be both identical between closely related species, and extremely variable throughout the ontogeny of an individual as teeth replace constantly; tooth morphology as the only or primary available diagnostic element for a holotype is not adequate for establishing a morphological hypodigm.
    2. Existing mosasaurid taxa represented by tooth‐based holotypes should be considered nomina dubia until better diagnostic material is described.

AUTHOR CONTRIBUTIONS

Henry S. Sharpe: Conceptualization; investigation; writing – original draft; methodology; visualization; formal analysis. Mark J. Powers: Resources; writing – review and editing; conceptualization; methodology; software; formal analysis; investigation; data curation. Michael W. Caldwell: Project administration; supervision; resources; writing – review and editing; methodology.

Supporting information

Figure S1. Sagittal CT section of Halisaurus arambourgi dentary UALVP 56123.

AR-308-2160-s004.png (628.2KB, png)

Figure S2. Coronal CT section of Halisaurus arambourgi dentary UALVP 56123 at second alveolus.

AR-308-2160-s005.png (517.2KB, png)

Figure S3. Coronal CT section of Halisaurus arambourgi dentary UALVP 56123 at fourth alveolus.

AR-308-2160-s003.png (550.9KB, png)

Figure S4. Coronal CT section of Halisaurus arambourgi dentary UALVP 56123 at fifth alveolus.

AR-308-2160-s001.png (547.8KB, png)

Figure S5. Coronal CT section of Halisaurus arambourgi dentary UALVP 56123 at sixth alveolus.

AR-308-2160-s002.png (587.8KB, png)

ACKNOWLEDGMENTS

We thank Gerardo Álvarez Herrera and Takuya Konishi for providing valuable feedback and suggestions that improved the quality and clarity of this manuscript, and Leonardo Kerber for the handling of our manuscript; Corwin Sullivan for the use of his laboratory's photography equipment; Fatima Iftikhar for discussions on teiid tooth attachment; Luke Nelson for discussions on shark dentition; Duane Froese for access to their laboratory's CT scanner; Joel Pumple for facilitating the scans and helping with the mounting process; Jordan Harvey for helping with the mounting process and postscan processing. Figures 1A, 1C, and 3A‐E are adapted from Cretaceous Research, Vol 123, Nicholas R. Longrich, Nathalie Bardet, Anne S. Schulp, Nour‐Eddine Jalil, “Xenodens calminechari gen. et sp. nov., a bizarre mosasaurid (Mosasauridae, Squamata) with shark‐like cutting teeth from the upper Maastrichtian of Morocco, North Africa”, Figures 3‐5, Copyright 2021, used with permission from Elsevier. H.S.S. is funded by a CGS‐M scholarship, and M.W.C. is funded by a Discovery Grant, both from the National Sciences and Engineering Research Council of Canada.

Sharpe, H. S. , Powers, M. J. , & Caldwell, M. W. (2025). Reassessment of Xenodens calminechari with a discussion of tooth morphology in mosasaurs. The Anatomical Record, 308(8), 2160–2172. 10.1002/ar.25612

REFERENCES

  1. Arambourg, C. (1952). Les vertébrés fossiles des gisements de phosphates (Maroc‐Algérie‐Tunisie). Notes et Mémoires du Service géologique du Maroc, 92, 1–372. [Google Scholar]
  2. Bardet, N. , Pereda Suberbiola, X. , Schulp, A. S. , & Bouya, B. (2008). New material of Carinodens (Squamata, Mosasauridae) from the Maastrichtian (late cretaceous) phosphates of Morocco. In Proceedings of the Second Mosasaur Meeting (pp. 29–36). Fort Hays State University. [Google Scholar]
  3. Caldwell, M. W. (2007). Ontogeny, anatomy and attachment of the dentition in mosasaurs (Mosasauridae: Squamata). Zoological Journal of the Linnean Society, 149(4), 687–700. [Google Scholar]
  4. Caldwell, M. W. , Budney, L. A. , & Lamoureux, D. O. (2003). Histology of tooth attachment tissues in the late cretaceous mosasaurid Platecarpus . Journal of Vertebrate Paleontology, 23(3), 622–630. [Google Scholar]
  5. Caldwell, M. W. , & Cooper, J. A. (1999). Redescription, palaeobiogeography and palaeoecology of Coniasaurus crassidens Owen, 1850 (Squamata) from the lower chalk (cretaceous; Cenomanian) of SE England. Zoological Journal of the Linnean Society, 127(4), 423–452. [Google Scholar]
  6. Coombs, W. P., Jr. (1988). The status of the dinosaurian genus Diclonius and the taxonomic utility of hadrosaurian teeth. Journal of Paleontology, 62(5), 812–817. [Google Scholar]
  7. Cope, E. D. (1876). Descriptions of some vertebrate remains from the fort union beds of Montana. Proceedings of the Academy of Natural Sciences of Philadelphia, 28, 248–261. [Google Scholar]
  8. D'Amore, D. C. (2015). Illustrating ontogenetic change in the dentition of the Nile monitor lizard, Varanus niloticus: A case study in the application of geometric morphometric methods for the quantification of shape–size heterodonty. Journal of Anatomy, 226(5), 403–419. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Davit‐Béal, T. , Allizard, F. , & Sire, J. Y. (2006). Morphological variations in a tooth family through ontogeny in Pleurodeles waltl (Lissamphibia, Caudata). Journal of Morphology, 267(9), 1048–1065. [DOI] [PubMed] [Google Scholar]
  10. Dollo, L. (1913). Globidens fraasi, mosasaurien mylodonte nouveau du Maestrichtien (Crétacé supérieur) du Limbourg, et l'Ethologie de la Nutrition chez les Mosasauriens. Archives de Biologie, 28, 609–626. [Google Scholar]
  11. Estes, R. , & Williams, E. E. (1984). Ontogenetic variation in the molariform teeth of lizards. Journal of Vertebrate Paleontology, 4(1), 96–107. [Google Scholar]
  12. Evans, D. C. , Cullen, T. M. , Larson, D. W. , & Rego, A. (2017). A new species of troodontid theropod (Dinosauria: Maniraptora) from the horseshoe canyon formation (Maastrichtian) of Alberta, Canada. Canadian Journal of Earth Sciences, 54(8), 813–826. [Google Scholar]
  13. Fox, R. C. (1969). Studies of late cretaceous vertebrates. III. A triconodont mammal from Alberta. Canadian Journal of Zoology, 47(6), 1253–1256. [Google Scholar]
  14. ICZN . (1999). International code of zoological nomenclature (4th ed., p. 306). The International Trust for Zoological Nomenclature. [Google Scholar]
  15. Kaddumi, H. F. (2009). The first and most complete Carinodens (Squamata: Mosasauridae) skeleton yet with a description of a new species from the Harrana Fauna. In Kaddumi H. F. (Ed.), Fossils of the Harrana Fauna and the adjacent areas (pp. 49–64). Publications of the Eternal River Museum of Natural History. [Google Scholar]
  16. LeBlanc, A. R. , Caldwell, M. W. , & Bardet, N. (2012). A new mosasaurine from the Maastrichtian (upper cretaceous) phosphates of Morocco and its implications for mosasaurine systematics. Journal of Vertebrate Paleontology, 32(1), 82–104. [Google Scholar]
  17. LeBlanc, A. R. , Lamoureux, D. O. , & Caldwell, M. W. (2017). Mosasaurs and snakes have a periodontal ligament: Timing and extent of calcification, not tissue complexity, determines tooth attachment mode in reptiles. Journal of Anatomy, 231(6), 869–885. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. LeBlanc, A. R. , Mohr, S. R. , & Caldwell, M. W. (2019). Insights into the anatomy and functional morphology of durophagous mosasaurines (Squamata: Mosasauridae) from a new species of Globidens from Morocco. Zoological Journal of the Linnean Society, 186(4), 1026–1052. [Google Scholar]
  19. Leidy, J. (1856). Notice of remains of extinct reptiles and fishes, discovered by F. V. Hayden in the bad lands of the Judith River, Nebraska territories. Proceedings of the Academy of Natural Sciences of Philadelphia, 8, 72–73. [Google Scholar]
  20. Longrich, N. R. , Bardet, N. , Schulp, A. S. , & Jalil, N. E. (2021). Xenodens calminechari gen. Et sp. nov., a bizarre mosasaurid (Mosasauridae, Squamata) with shark‐like cutting teeth from the upper Maastrichtian of Morocco, North Africa. Cretaceous Research, 123, 104764. [Google Scholar]
  21. Longrich, N. R. , Jalil, N. E. , Pereda‐Suberbiola, X. , & Bardet, N. (2023). Stelladens mysteriosus: A strange new Mosasaurid (Squamata) from the Maastrichtian (late cretaceous) of Morocco. Fossil Studies, 1(1), 2–14. [Google Scholar]
  22. Luan, X. , Walker, C. , Dangaria, S. , Ito, Y. , Druzinsky, R. , Jarosius, K. , Lesot, H. , & Rieppel, O. (2009). The mosasaur tooth attachment apparatus as paradigm for the evolution of the gnathostome periodontium. Evolution & Development, 11(3), 247–259. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Massare, J. A. (1987). Tooth morphology and prey preference of Mesozoic marine reptiles. Journal of Vertebrate Paleontology, 7(2), 121–137. [Google Scholar]
  24. Mertens, R. (1942). Die Familie der Warane (Varanidae). Zweiter Teil: Der Schädel. Abhandlungen der Senckenbergischen Naturforschenden Gesellschaft, 465, 117–234. [Google Scholar]
  25. Rieppel, O. , & Kearney, M. (2005). Tooth replacement in the late cretaceous mosasaur Clidastes . Journal of Herpetology, 39(4), 688–692. [Google Scholar]
  26. Rossi, V. , Bernardi, M. , Fornasiero, M. , Nestola, F. , Unitt, R. , Castelli, S. , & Kustatscher, E. (2024). Forged soft tissues revealed in the oldest fossil reptile from the early Permian of the Alps. Palaeontology, 67(1), e12690. [Google Scholar]
  27. Rowe, T. , Ketcham, R. A. , Denison, C. , Colbert, M. , Xu, X. , & Currie, P. J. (2001). The Archaeoraptor forgery. Nature, 410, 539–540. [DOI] [PubMed] [Google Scholar]
  28. Schulp, A. S. , Bardet, N. , & Bouya, B. (2009). A new species of the durophagous mosasaur Carinodens (Squamata, Mosasauridae) and additional material of Carinodens belgicus from the Maastrichtian phosphates of Morocco. Netherlands Journal of Geosciences, 88(3), 161–167. [Google Scholar]
  29. Strong, C. R. , Caldwell, M. W. , Konishi, T. , & Palci, A. (2020). A new species of longirostrine plioplatecarpine mosasaur (Squamata: Mosasauridae) from the late cretaceous of Morocco, with a re‐evaluation of the problematic taxon ‘Platecarpus’ ptychodon . Journal of Systematic Palaeontology, 18(21), 1769–1804. [Google Scholar]
  30. Woodward, A. S. (1891). Note on a tooth of an extinct alligator (Bottosaurus belgicus, sp. nov.) from the lower Danian of Ciply, Belgium. Geological Magazine, 8(3), 114–115. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Figure S1. Sagittal CT section of Halisaurus arambourgi dentary UALVP 56123.

AR-308-2160-s004.png (628.2KB, png)

Figure S2. Coronal CT section of Halisaurus arambourgi dentary UALVP 56123 at second alveolus.

AR-308-2160-s005.png (517.2KB, png)

Figure S3. Coronal CT section of Halisaurus arambourgi dentary UALVP 56123 at fourth alveolus.

AR-308-2160-s003.png (550.9KB, png)

Figure S4. Coronal CT section of Halisaurus arambourgi dentary UALVP 56123 at fifth alveolus.

AR-308-2160-s001.png (547.8KB, png)

Figure S5. Coronal CT section of Halisaurus arambourgi dentary UALVP 56123 at sixth alveolus.

AR-308-2160-s002.png (587.8KB, png)

Articles from Anatomical Record (Hoboken, N.j. : 2007) are provided here courtesy of Wiley

RESOURCES