Systematic analysis
Theropoda Marsh, 1881
TetanuraeGauthier, 1986
MegaraptoraBenson, Carrano, and Brusatte, 2010
Megaraptoridae Novas, Agnolín, Ezcurra, Porfiri, and Canale, 2013
Gen. and sp. indet.
Material. MCF-PVPH-920 and MCF-PVPH-930; Morphotype 1 (Figure 2).
State of preservation and general morphology.The recovered teeth are well preserved and are represented by shed crowns without their roots. Specimen MCF-PVPH-920 lacks most of its enamel. The denticles are not worn and show their original shape.
Crown overall morphology.The teeth are ziphodont type with a labiolingually compressed crown, curved distally, strongly convex mesial margin, and a concave distal margin in lateral view. Labiolingual compression of the crown close to the cervix (CBR) is equal to 0.53 in MCF-PVPH-930 and 0.79 in MCF-PVPH-920, whereas at mid-crown (MCR) is equal to 0.53 and 0.86 in MCF-PVPH-930 and MCF-PVPH-920, respectively. The baso-apical elongation of the crown ratio (CHR) varies between 1.48 to 2.21. The mesial border shows a unserrated carina extending three-quarters apicobasally without reaching the cervix. This carina is curved lingually towards the base. On the distal view, the distal margin possesses a well-developed serrated distal carina, straight or very slightly bowed, and terminates beneath the cervix. The distal carina bears around 14/15 denticles per 5 mm at mid-crown and 19 denticles per 5 mm close to the cervix, whereas around 12/13 denticles per 5 mm are observed close to the apex. The shape of denticles on the distal carina is symmetrically convex to slightly asymmetrical with a parabolic margin. They are horizontal and subrectangular in shape, i.e., longer mesiodistally than apicobasal, and perpendicular to the distal margin. The interdenticular space is broad, whereas the interdenticular sulci, both at the mid- and the base crown, are long and well-developed. Lingual and labial surfaces exhibit a shallow median depression with a triangular shape in their basal half, with the lingual depression being more clearly defined and flanked by subtle ridges. Due to these depressions, the cross-section takes on a figure-eight shape near the level of the cervix. The crown shows arcuate transverse undulations. A braided enamel texture is observed. Marginal undulations, flutes, longitudinal grooves, or ridges (sensu[16]) are absent.
Megaraptoridae Novas, Agnolín, Ezcurra, Porfiri, and Canale, 2013
Gen. and sp. indet.
Material.MCF-PVPH-921, MCF-PVPH-922, MCF-PVPH-924, MCF-PVPH-925, MCF-PVPH-929, MCF-PVPH-930, MCF-PVPH-933, MCF-PVPH-934, MCF-PVPH-935, MCF-PVPH-937, MCF-PVPH-938, MCF-PVPH-939, MCF-PVPH-940, MCF-PVPH-941, MCF-PVPH-944, MCF-PVPH-945, MCF-PVPH-947, MCF-PVPH-948, and MCF-PVPH-949; Morphotype 2 (Figure 3).
State of preservation and general morphology.Two teeth MCF-PVPH-933 and MCF-PVPH-940 from the recovered sample are complete.Specimens MCF-PVPH-921/22, MCF-PVPH-933 and MCF-PVPH-929, MCF-PVPH-934/935, MCF-PVPH-937 and MCF-PVPH-933 and MCF-PVPH-944 are virtually complete, lacking some sectors of their crowns. The rest of teeth are represented by fragments of the crown where only one portion of its central zone is preserved.
Crown overall morphology.Specimens asigned to this morphotype have a crown labiolingually compressed and distally recurved, with a strongly convex mesial margin, and a concave distal margin in lateral view (zhiphodont type). Both the labiolingual compression of the crown close to the cervix (CBR) and at mid- (MCR) varies in a range from 0.51 to 0.63. The baso-apical elongation of the crown ratio (CHR) varies between 1.63 to 1.68. Unlike Morphotype 1, this one lacks a mesial carina on the mesial border. The serrated distal carina is well-developed, straight or very slightly bowed, and terminates beneath the cervix. The density of denticles close to the cervix (15 to 22.5 per 5 mm; DB) is greater than the density of denticles at mid-crown (12 to 18 per 5 mm; DC). The density of denticles at the apex is equal than that at mid-crown. The shape of denticles on the distal carina is asymmetrically convex with a parabolic margin, and are longer mesiodistally than apicobasally, and perpendicular to the distal margin. The interdenticular space is narrow, whereas interdenticular sulci are not present as in the morphotype previously described. Like Morphotype 1, the lingual and labial surfaces exhibit a shallow median depression with a triangular shape in their basal half, with the lingual depression being more clearly defined and flanked by subtle ridges. Due to these depressions, the cross-section takes on a eight-shape near the level of the cervix. The crown shows well-marked arcuate transverse and marginal undulations. A braided enamel texture is observed.
Theropoda Marsh, 1881
Ceratosauria Marsh, 1884
Abelisauroidea Bonaparte, 1991
Abelisauridae Bonaparte and Novas, 1985
Gen. and sp. indet.
Material.MCF-PVPH-923, MCF-PVPH-936 and MCF-PVPH-946; Morphotype 3 (Figure 4).
State of preservation and general morphology.The collected crowns are quite well-preserved with MCF-PVPH-923 being the most complete. The remaining two lack their most basal portion (MCF-PVPH-936) or most apical portion (MCF-PVPH-946). All specimens described here, as well as all specimens studied, are considered to be shed teeth.
Crown overall morphology.The crowns are strongly labiolingually compressed with an MCR equal to 0.52, and are slightly curved in an anteroposterior sense with a straight distal profile. Both lingual and labial surfaces are strongly convex. Both mesial and distal carinae are well-developed, having serrations centrally positioned along the margins. The mesial carina bears 16 denticles per 5 mm close to the cervix, 12/15 denticles per 5 mm at mid-crown, and 14/15 denticles per 5 mm in the apex of the crown. The distal carina possesses 12 denticles per 5 mm at mid-crown and 11/15 denticles per 5 mm close to the apex. The relationship between the number of mesial and distal denticles (DSDI) is 1. The denticles of the mesial carina are asymmetrical and inclined apically from the mesial margin, whereas the denticles of the distal carina also are asymmetrical but perpendicular to the distal margin. The interdenticular space is narrow in the mesial carina and broad in the distal one. The enamel texture is smooth and not oriented in any preferential direction. The cross-section is lenticular almost at the level of the cervix.
TheropodaMarsh, 1881
CoelurosauriaHuene, 1920
Alvarezsauria Bonaparte, 1991
Alvarezsauridae Bonaparte, 1991
Gen. and sp. indet.
Material. MCF-PVPH-926; Morphotype 4 (Figure 5).
State of preservation and general morphology.This specimen consists in a small crown that lacks the labial side and the most apical portion.
Crown overall morphology.Although the labial side is not preserved, it can be inferred that its labiolingual compression (CBR) is weak, and the crown is slightly curved in an anteroposterior direction. In lateral view, its mesial margin is slightly convex, and its distal margin is slightly concave. The mesial and distal carina are present and serrated, although somewhat worn in some sectors. The mesial carina bears approximately 45 denticles per 5 mm close to the cervix, whereas the distal carina, accounts approx 77 denticles per 5 mm in the apex of the crown. The denticles in the mesial carina are asymmetrical and inclined apically from the mesial margin, whereas the denticles of the distal carina are asymmetrical and perpendicular to the distal margin. At least seven fainted ridges are observed about the basal zone of the crown. It is possible to observe in one sector, a constriction between the crown and the root. The enamel texture is smooth and not oriented in any preferential direction.
TheropodaMarsh, 1881
CeratosauriaMarsh, 1884
Abelisauroidea Bonaparte, 1991
Abelisauridae Bonaparte and Novas, 1985
Gen. and sp. indet.
Material.MCF-PVPH-931 and MCF-PVPH-932; Morphotype 5 (Figure 6).
State of preservation and general morphology.Both specimens are well preserved, lacking some basal (MCF-PVPH-931) or apical (MCF-PVPH-932) sectors of the crown.
Crown overall morphology.Both possess an important labiolingual compression close to the base of the crown (CBR = 0.44 to 0.52 and MCR = 0.45 to 0.46). Distally they are recurved with a strongly convex mesial margin and a straight distal margin in lateral view. The mesial and distal margins possess denticulatedcarinae. In the mesial carina, the density of denticles close to the cervix (27.5 per 5 mm; MB) is greater than the density of denticles at mid-crown (20 to 25 per 5 mm; MC). The density of denticles at the apex is equal to the density at mid-crown (20 to 25 per 5 mm; MA). In the distal carina, the density of denticles close to the cervix (12.5 per 5 mm; DB) is lower than the density of denticles at mid-crown (15 to 22.5 per 5 mm; DC). The density of denticles in the apex is greater than that at mid-crown. The shape of denticles on both carinae is asymmetrically convex with a parabolic margin, and are longer mesiodistally than apicobasal, and inclined from the mesial and distal margin. The interdenticular space is narrow, whereas the interdenticular sulci are present and well-developed. Also, a concave surface adjacent to the distal carina is observed. The outline of the basal cross-section of the crown is lenticular to lanceolate. The texture of the enamel is smooth and not oriented in any preferential direction.
TheropodaMarsh, 1881
CeratosauriaMarsh, 1884
Abelisauroidea Bonaparte, 1991
Gen. and sp. indet.
Material.MCF-PVPH-942, Morphotype 6 (Figure 7).
State of preservation and general morphology.This specimen lacks of its most basal and apical portions. The denticles are worn, so it is difficult to visualize their original shape.
Crown overall morphology.The tooth is ziphodont type, slightly recurved distally with a convex mesial margin and a straight to slightly concave distal one. The MCR is 0.46, which implies a strong labiolingual compression at the level of the mid-crown. Both carinae are present, being denticulated, and centrally positioned along the mesial and distal margins. It is not possible to know whether the distal carina ends just above or below the cervix, although the mesial carina ends well above the cervix. The density of denticles is 2.4 per mm in both carinae. Adjacent to the distal carina, on the labial and lingual sides, there is a long apicobasally extended concave surface. Besides, the outline of the basal cross-section of the crown is lenticular to lanceolate in shape, and the texture of the enamel is smooth without any preferential direction.
Cladistic analysis
The cladistic analysis performed from the dentition-based data matrix using a constrained tree topology recovered four most parsimonious trees (MPTs; CI = 0.194; RI = 0.460; L = 1346 steps). Morphotypes 1 and 2 are found within Megaraptoridae (Fig. 8). The latter is recovered as the sister taxon of Murusraptor, whereas morphotype 1 was either found in a small polytomy as the sister taxon of Orkoraptor or Megaraptor (Fig. 8). Morphotypes 3, 5, and 6 are recovered within a small subclade as the sister taxa of Allosauroidea (Fig. 8). Morphotype 4 is recovered as a early-diverging member of Neocoelurosauria (Fig. 8).
The cladistic analysis performed with no constraint found more than a hundred most parsimonious trees (CI = 0.239; RI = 0.587; L = 1090 steps). In the strict consensus tree, Morphotype 2 is recovered as the sister taxon of Murusraptor, whereas Morphotype 1 is recovered within a large polytomy together with some megaraptorids, maniraptoriforms, tyrannosauroids, and allosauroids (Fig. 9). This is because Morphotype 1 is recovered as the sister taxon of Orkoraptor, which in turn is recovered as a later-diverging tyrannosaurid. Morphotypes 3, 5, and 6 are recovered within Abelisauridae in a polytomy with the brachyrostran forms (Fig. 9). Finally,Morphotype 4 is recovered in a small polytomy formed by some ornithomimosaurians, oviraptorosaurians, therizinosaurians, Mononykus, Shuvuuia, and Chilesaurus (Fig. 9).
Regarding the cladistic analysis conducted on the crown-based data matrix with no constraints, we found a hundred MPTs (CI = 0.240, RI = 0.630, L = 667 steps), and the strict consensus tree recovered a topology very similar to the previous analysis (Fig. 10).
Discriminant analysis
The Discriminant Function Analysis (hereafter referred to as DFA), carried out on the entire dataset, categorized shed-isolated teeth corresponding to Morphotype 1 as belonging to neovenatorids and tyrannosaurids. Meanwhile, Morphotype 2 was identified among megaraptorids (Fig. 11). Morphotypes 3, 5, and 6 were classified as abelisaurids and therizinosaurids in the clade-level analysis, with PC1 and PC2 accounting for 48.13% and 19.86% of the total variance, respectively (see Supplementary Information 3). At the taxon level, most teeth were found to be associated with each other. Only one tooth of Morphotype 1 (MCF-PVPH-920) was classified with Zhuchengtyrannus, while Morphotype 2 (MCF-PVPH-939) was associated with Australovenator. The same pattern emerged for Morphotype 3 (MCF-PVPH-923), closely related to Aucasaurus (PC1 and PC2 accounted for 41.28% and 20.69% of the total variance, respectively; see Supplementary Information 3). In both clade-level and taxon-level analyses, the reclassification rate (hereafter RR) was found to be low, at 58.84% and 57.73%, respectively. The reclassification rate slightly improved in the DFA performed with the dataset where absent denticles were coded as inapplicable, reaching 58.12% at the clade level and 56.5% at the taxon level. In these analyses, isolated teeth were consistently classified within the same groups (clade-level: PC1 and PC2 accounting for 47.01% and 18.9%; taxon-level: PC1 and PC2 accounting for 41.76% and 15.84%; see Supplementary Information 3).
The DFA carried out using the whole dataset of[22] based on first-hand measurements, classified the isolated teeth as carcharodontosaurids, abelisaurids, dromaeosaurids and troodontids (clade-level analysis; PC1 and PC2 account for 49.66% and 24.22% of the total variance, respectively; Supplementary information 3). At the taxon-level (PC1 41.24% and PC2 19.14%), the shed teeth were found closely related to members of carcharodontosaurids, abelisaurids, and megaraptorids. The RR is better at the taxon-level (58.69%) than at the clade-level (54.42%). The DFA made when the absence of denticles was coded as inapplicable in dataset resulted in the teeth being grouped with carcharodontosaurids, abelisaurids, dromaeosaurids, tyrannosaurids, and troodontids (clade-level analysis; PC1 and PC2 account for 40.21% and 27.66% of the total variance, respectively; Supplementary information 3). In turn, at taxon-level, these were found closely related to members of carcharodontosaurids, abelisaurids, megaraptorids, and megalosaurids (PC1 34.68% and PC2 21.34%). The RR is slightly higher in clade-level (55.03%) and slightly lower in the taxon-level (55.34%).
In the DFA performed on the datasets restricted to taxa with teeth larger than two centimeters (i.e., the whole dataset of Hendrickx’s first-hand measurements), the isolated teeth are classified as megaraptorids, abelisaurids, megalosaurids, non-megalosauranmegalosaurids, tyrannosaurids, and non-tyrannosauroidtyrannosaurids (clade level; PC1 40.34% and PC2 27.36%; Fig. 11). At taxon-level, the teeth are grouped with the megaraptorids, abelisaurids, non-abelisauroid abelisaurids, and tyrannosaurids clades (taxon-level; PC1 50% and PC2 17.29%). The RR is 56.47% at clade-level and 63.39% at taxon-level, a percent higher than anterior analyses. In the DFA performed when the absence of denticles is considered inapplicable in the dataset, the same results were recovered, although the RR is 57.81% at the clade-level and 63.17% at the taxon-level.
In the DFA conducted on the datasets restricted to Argentine taxa, the different morphotypes are classified as megaraptorids, abelisaurids, non-abelisauroid ceratosaurians, and carcharodontosaurids (clade level; PC1 48.63% and PC2 28.76%; Fig. 12). At taxon-level, the results of the DFA recovered the same results. The RR increased drastically, being 80.25% at the clade-level and 79.63% at the taxon-level.
Cluster analysis
The cluster analysis, conducted on datasets limited to taxa with teeth larger than two centimeters and employing the hierarchical clustering option, identified the isolated teeth as belonging to neovenatorids, carcharodontosaurids, megaraptorids, megalosaurids, abelisaurids, and tyrannosaurids (Fig. 12). Similarly, the cluster analysis carried out on the dataset where the absence of denticles is deemed inapplicable yielded a classification akin to the previous analysis (see Supplementary information 3).
The cluster analysis, employing the neighbor-joining option, identified the shed teeth as members of neovenatorids, carcharodontosaurids, megaraptorids, megalosaurids, abelisaurids, allosaurids, and tyrannosaurids. When utilizing the dataset where the absence of denticles is considered inapplicable, the results remained consistent, albeit with minor variations in the composition of its members (see Supplementary information 3).