Cenomanian theropod teeth from Algora (central Spain), new evidence for majungasaurine abelisaurids in Europe Authors: Mirella López-Miguel, Angélica Torices, Elisabete Malafaia, Francisco Ortega, Adán Pérez-García Published: 2026 ABSTRACT AbstractIn this work 40 isolated teeth of theropod dinosaurs from the Cenomanian (Upper Cretaceous) locality of Algora are described. The paleontological area of Algora (Guadalajara province, Spain) provides the highest concentration of macroremains of Cenomanian vertebrates in southwestern Europe. Isolated theropod teeth are the most abundant remains in Algora as far as this group is concerned. Isolated theropod teeth constitute important evidence to analyse theropod diversity, since they are quite common in the fossil record. There have been previous tentative assignations of these theropod teeth to different theropod groups such as Carcharodontosauridae and Abelisauridae. However, in this work we have conducted a more exhaustive and robust analysis with a larger sample size using morphological comparisons using multivariate and cladistic analyses. The results of the analyses carried out in this work support the identification of the isolated teeth of Algora as a theropod abelisaurid, most likely belonging to the subfamily Majungasaurinae. This would further corroborate the presence of this abelisaurid lineage in Europe during the Late Cretaceous, from the Cenomanian onwards.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-026-53271-3.Keywords: Theropod, Teeth, Cladistic, Morphometry, Upper Cretaceous, SpainSubject terms: Ecology, Ecology, Evolution, Zoology Introduction Information on the vertebrate faunas from the early and middle stages of the European Late Cretaceous, especially between the Cenomanian and the Coniacian, is extremely limited compared to that from both the Early Cretaceous and the Santonian-Maastrichtian interval1–5. The fossil record shows that the faunas of the Lower and uppermost Cretaceous differ markedly from one another2,6–11. The Algora site in Spain (Guadalajara Province, Spain,Fig. 1a), of uppermost middle or lowermost upper Cenomanian age, is considered the main concentration of Cenomanian vertebrate macroremains currently known in southwestern Europe3,12–14,15. Prior to the discovery of this paleontological site, information on the Spanish Cenomanian vertebrate faunas came almost exclusively from a single locality (i.e., a quarry in Oviedo, in the Asturias Province), which had yielded only a few discoveries, consisting of isolated and generally fragmentary microremains5. Considering the paleogeographical position of the Iberian Peninsula at that time, the Algora site provides valuable data for understanding the transition between Early and Late Cretaceous vertebrate faunas in Europe, evidencing how faunal exchanges between Gondwana and Laurasia developed2. (a). Geographical location of the Cenomanian (Late Cretaceous) palaeontological locality of Algora (Guadalajara, central Spain); (b). Isolated theropod tooth from the Algora site. c-g. Examples of different preservation states of isolated theropod teeth from Algora: (c), ALG 219, tooth fragment; (d), enamel wear in ALG 193; (e), eroded enamel and carinae without denticles in ALG 198; (f), ferruginous crust covering the labial face of ALG 203; (g), damaged enamel in ALG 213. In general, isolated teeth are the best preserved and most abundant theropod remains, due to the hardness of the enamel, which facilitates their preservation, and the continuous tooth replacement that occurred throughout the life of the individual16. Thanks to their abundance, theropod teeth provide important taphonomic, paleoenvironmental, and paleoecological information about the localities where they are found, as well as insights into the evolution of paleocommunities and the geographical and temporal distribution of theropod clades (e.g17–21. Furthermore, data on the size and morphology of the teeth and their denticles can be used to formulate hypotheses about individual body size, bite force, and feeding behavior22–24. Postcranial skeletal remains are less abundant than the teeth in the fossil record, as they are more fragile, especially in small theropods, and more susceptible to postmortem taphonomic processes that lead to rapid decomposition and fragmentation25. In this sense, isolated teeth are the most abundant remains of theropod dinosaurs at Algora2,26 (Fig. 1b), displaying different states of conservation (Fig. 1c-g). Over time, techniques for studying isolated teeth have been progressively developed and refined, with the application of methods such as cladistic and morphometric analyses, allowing for more accurate systematic classifications27–29. However, the assignment of these remains to lower taxonomic levels is often challenging, due to their high degree of homoplasy and the scarcity of detailed descriptions of theropod tooth crown morphology28. Initially identified as Carcharodontosauridae indet3, the theropod teeth from Algora were more recently reinterpreted as cf. Abelisauridae2, a diverse group of theropod dinosaurs mainly known from the Cretaceous of Gondwana30,31,32. In the present work, qualitative morphological characters and quantitative methods (cladistic and morphometrics analyses) are applied to test the phylogenetic affinities of the isolated theropod teeth from this Laurasian locality, considering both the few previously referenced teeth and a relatively large collection of unpublished specimens. Methodology Comparative methodology and terminology Denticles and crown morphology, crown ornamentation, and enamel surface texture of the isolated theropod teeth from Algora were examined using the digital microscope dino-lite Edge AM7915MZT. The dental nomenclature and protocol proposed by Hendrickx et al33 were followed to describe the specimens. The terminology for anatomical orientation proposed by Dodson & Smith34 and Hendrickx et al33 as well as the anatomical and morphometrics terms and abbreviations defined by Smith et al29 and Hendrickx et al33 (Fig. 1 Appendix I, 1) were used. Thirteen measurements of the dental crowns (CBL, CBW, CH, AL, MCL, MCW, CA, MA, MC, MB, DA, DC, DB) (Fig. 2 Appendix I,1.2) as proposed by Hendrickx et al33 were taken using the digital microscope dino-lite Edge AM7915MZT and the software associated to this microscope. Six of these measurements (CBL, CH, MA, MC, DC, DA) were then added to the comparative dataset. Labial view of theropod teeth from the Cenomanian fossil site of Algora (Guadalajara, central Spain). (a). ALG 193; (b). ALG 194; (c). ALG 195; (d). ALG 196; (e). ALG 197; (f). ALG 198; (g). ALG 199; (h). ALG 201; (i). ALG 202; (j). ALG 203; (k). ALG 204; (l). ALG 205; (m). ALG 206; (n). ALG 207; (o). ALG 208; (p). ALG 209; (q). ALG 210; (r). ALG 211; (s). ALG 212; (t). ALG 213; (u). ALG 214; (v). ALG 217; (w). ALG 218; (x). ALG 219; (y). ALG 221. Another important feature considered in the taxonomic classification of teeth is the cross-section outline of the crown. According to Hendrickx et al.33, the basal cross sections outline is significant for the taxonomic classification of isolated theropod teeth, as it provides important information about their position in the jaw and allow assign them to mesial or lateral dentition, or even to a specific theropod clade. Multivariate statistical and cladistic analyses In order to quantitatively validate the morphotypes, we performed six multivariate statistical analyses, including one principal component and five discriminant analyses, using SPSS 26 (IBM SPSS Statistics for Windows, v. 26.0, 2020). In the statistical analysis, a total of 238 teeth, represented by 775 from the database and 40 teeth from Algora, were analysed (Appendix I, 3). For the database, we selected mainly Upper Jurassic to Upper Cretaceous carcharodontosaurids such as Acrocanthosaurus and Carcharodontosaurus, abelisaurids such as Majungasaurus, Abelisaurus, Chenanisaurus, Arcovenator and some indeterminate abelisaurid from different fossil sites of the Iberian Peninsula, Allosaurus, Tyrannosaurids such as Daspletosaurus, Tyrannosaurus and Gorgosaurus, Megalosaurids such as Duriavenator, Megalosaurus, Dubreuillosaurus and Torvosaurus from Smith et al.29, Richter et al.35, Fanti et al.17, Tortosa et al.36, Hendrickx and Mateus27,37, Hendrickx et al28,38, Pérez-García et al9,13, Longrich et al.39,Young et al.40, Malafaia et al.41,42) and Isasmendi et al.43,44. Additionally, we include other theropod taxa found regularly for other Upper Cretaceous Iberian sites such as different dromaeosaurids, Troodon and Richardoestesia45–4720. The variables considered for the statistical analysis were: CBL, CBW, CH, DentD/mm and DentM/mm. Before the statistical analysis, the data were normalized using Log (x + 1), so that the data with zero values could be used in the analysis (Appendix I, 3). Also, the statistical analyses considered only those specimens that have measurements in all variables, so teeth that are broken and, therefore, not complete, are not used in the analyses. Cladistic analyses were also performed using the data matrix of Hendrickx et al.48 focused on the dentition of non-avian theropods.The data matrix consists of 148 non-ordered and similarly weighted discrete dental characters scored in 124 saurischians (i.e., 122 non-avian theropods and five early branching saurischians) including the isolated teeth of Algora, (Appendix I, 2.1.1). The cladistic analyses were performed using TNT 1.649 following the methodology of Hendrickx et al28,48,50 using a constrained topological tree of non-avian theropod relationships and the teeth from Algora as floating taxa to allow their placement in any position. As in the methodology of Hendrickx et al.48, only operational taxonomic units (OTUs) for which more than 50% of crown-based characters (i.e., characters on the mesial and lateral dentition) could be scored were kept in the analysis. As Hendrickx et al.48 propose, the dataset was divided into mesial and lateral dentitions so that only mesial and lateral based characters were kept in the data matrices. Only specimen ALG-193 and ALG-221 were preliminarily identified as mesial teeth and scored in the mesial-crown based data matrix accordingly. Both of data matrix from Hendrickx et al.48 were phylogenetically bracketed between the early branching dinosaur Daemonosaurus chauliodus51 as recovered by Cau52, and the avialan theropod Archaeopteryx lithographica53. Following the methodology previously mentioned by Hendrickx et al.48, the cladistic analyses were performed using two constrained trees representing alternative hypotheses of theropod relationships. The first topology hypothesizes herrerasaurids as early branching saurischians, elaphrosaurines as early branching ceratosaurs, allosauroids among avetheropods, megaraptorans within tyrannosauroids, and scansoriopterygids among oviraptorosaurs. It is based on the results of Moro et al.54 for non-theropod saurischians, Ezcurra et al.55 for non-averostran theropods, Hendrickx et al, first analysis) for Ceratosauria, Kellermannetal56, split dataset) for non-maniraptoriform tetanurans, Zheng et al.57 for derived Tyrannosauroidea, Cuesta et al. (2022) for Ornithomimosauria, Kubo et al.58 for Alvarezsauria, Kobayashi et al.59 for Therizinosauria, Hao et al.60 for non-scansoriopterygid oviraptorosaurs, Brum et al.61 for non-eudromaeosaur Dromaeosauridae, Czepiński62 for Eudromaeosauria, and Kubota et al.63 for Troodontidae. The second tree topology proposed by Hendrickx et al.48 postulates a phylogenetic position of elaphrosaurines within noasaurids, megalosauroids among carnosaurians, megaraptorans within allosauroids, compsognathids as the earliest branching coelurosaurs, therizinosaurs and alvarezsaurs as sister taxa, and Ornitholestes as the earliest branching oviraptorosaur. This topology was mainly based on the results of the cladistic analyses performed by Pol et al.64 for Ceratosauria, Pradelli et al.65 for Carnosauria, Zheng et al.57 for non-tyrannosaurid Tyrannosauroidea, Dalman et al.66 for Tyrannosauridae, and Chapelle et al.67 for Ornitholestes (the rest of the neocoelurosaur classification being the same as in the first topology). The search strategy followed the methodology outlined by Hendrickx et al48, using a combination of tree-search algorithms, Wagner trees, TBR branch swapping, sectorial searches, Ratchet (perturbation phase stopped after 20 substitutions), and Tree Fusing (5 rounds), until 100 hits of the same minimum tree length were achieved, and finally subjected the best trees to a final round of TBR branch swapping. The resulting command in TNT was “xmult = hits 100 rss fuse 5 ratchet 20” followed by “bb”. Using “sect:slack 50”to increase the slack for sector in case the message “out of memory re-packing data for sectorial searches” appeared during the analysis. Finally, the synapomorphies, autapomorphies, and other diagnostic characteristics defining each taxon were identified. Systematic paleontology Dinosauria68 Saurischia69 Theropoda70 Ceratosauria71 Abelisauridae72 Majungasaurinae36 Majungasaurinae indet. Material Forty isolated theropod teeth from the Spanish Cenomanian site of Algora, housed in the Museo de Paleontología de Castilla-La Mancha (Cuenca, Spain), were selected for this study. They correspond to the specimens: ALG 193 (Fig. 2a), ALG 194 (Fig. 2 b), ALG 195 (Fig. 2c), ALG 196 (Fig. 2d), ALG 197 (Fig. 2e), ALG 198 (Fig. 2f), ALG 199 (Fig. 2g), ALG 201 (Fig. 2h), ALG 202 (Fig. 2i), ALG 203 (Fig. 2j), ALG 204 (Fig. 2k), ALG 205 (Fig. 2l), ALG 206 (Fig. 2m), ALG 207 (Fig. 2n), ALG 208 (Fig. 2o), ALG 209 (Fig. 2p), ALG 210 (Fig. 2q), ALG 211 (Fig. 2r), ALG 212 (Fig. 2s), ALG 213 (Fig. 2t), ALG 214 (Fig. 2u), ALG 217 (Fig. 2v), ALG 218 (Fig. 2w), ALG 219 (Fig. 2x), ALG 220, ALG 221 (Fig. 2y), ALG 222 (Fig. 3a), ALG 223 (Fig. 3b), ALG 224 (Fig. 3c), ALG 225 (Fig. 3d), ALG 226, ALG 227 (Fig. 3e), ALG 228 (Fig. 3f), ALG 229 (Fig. 3g), ALG 230 (Fig. 3h), ALG 231 (Fig. 3i), ALG 232 (Fig. 3j), ALG 233 (Fig. 3k), ALG 234 (Fig. 3l), ALG 235 (Fig. 3m). Labial view of theropod teeth from the Cenomanian site of Algora (Guadalajara, central Spain). (a). ALG 222; (b). ALG 223; (c). ALG 224; (d). ALG 225; (e). ALG 227; (f). ALG 228; (g). ALG 229; (h). ALG 230; (i). ALG 231; (j). ALG 232; (k). ALG 233; (l). ALG 234; (m). ALG 235. Description These specimens correspond to ziphodont teeth that are labiolingually compressed and blade-shaped, with serrated carinae on both margins, a convex mesial margin, and a straight or weakly convex distal margin in labial and lingual views. Their cross-section is lanceolate (sensu33. Only ALG 193 and ALG 221 have a more triangular shape, with convex mesial and distal margins, indicating an anterior position. Their cross-section type is salinon-shaped (sensu33. Regarding crown height, the teeth range from 6 to 29 mm (Appendix I, 1.2, Table I). Almost all of the specimens have a CBR between 0.34 and 0.60, characterizing them as labiolingually compressed, except for ALG 193 and ALG 221, which present a CBR of 0.69 and 0.64, respectively (Appendix I, 1.2, table I) and therefore are considered mesial teeth (sensu29,33. All teeth have an apex that is apically oriented and slightly displaced distally. The CHR values range from 0.57 to 2.47, indicating that these are apicobasally elongate teeth. Both labial and lingual surfaces are mesiodistally convex, except for ALG 193 and ALG 221, which present a strongly convex labial surface and a flattened lingual surface. Denticles are present on both the mesial and distal carinae. Their shape varies from subquadrangular to rectangular, being more mesiodistally elongated than apicobasally. The outlines of external margins are straight or slightly convex. The number of denticles per 5 mm ranges from 13 to 16, with a variation of 2 or 3 denticles observed at the apical, basal, or central parts of the crown (Appendix I, 1.2, Table I). Most Algora teeth do not exhibit interdenticular sulci; when present, these sulci are short and poorly developed, with ALG 207 being the only specimen showing a certain angle (Fig. 4b). The better preserved denticles are located at about two thirds of the crown height on both mesial and distal carinae, as many teeth have broken or eroded bases and apexes. Specimens ALG-198, ALG 204, ALG 206, ALG 224, and ALG 232 have eroded carinae and therefore lack denticles. The lateral teeth have an irregular and non-oriented enamel texture (sensu33,38 and tenuous marginal and transverse undulations that are only visible at certain angles (Fig. 4a). Most teeth have damaged enamel due to erosion, with dentine exposed in some parts as a result of enamel wear. Some crowns exhibit large ferruginous crusts that partially or completely cover one or both surfaces of the tooth. Details of some features of theropod tooth crowns from the Cenomanian site of Algora showing: (a), marginal undulation (mun) in ALG 193; (b), interdenticular sulci (ids) in ALG 207; (c), wear facet (wfa) produced by tooth-to-tooth contact in ALG 199. ALG 199, ALG 221, and ALG 227 show wear facets on the apical part of the tooth, which are interpreted as resulting from tooth-to-tooth contact73 (Fig. 4c). Algora teeth appear as fully or partially complete isolated crowns, lacking roots. ALG 194, ALG 197, ALG 214, ALG 217, ALG 224, and ALG 235 are incomplete crowns, missing apexes and having damaged bases, which made obtaining accurate measurements difficult. Some teeth also show partially damaged or broken carinae, especially in the basal part, such as ALG 201, ALG 202, ALG 225, ALG 228, and ALG 233, ALG 211, ALG 217, ALG 218, ALG 219, ALG 220, and ALG 226 are fragments or remain attached to the matrix. Results The results of the Principal Component Analysis show that the first two principal components explain 84.331% of the total variance. The variables with greater weight for the first component are those related to overall tooth size, while the variables with more weigh for the second component are those associated with denticle density. As shown in Figure 5 Algora teeth overlap mainly with abelisaurid teeth and slightly with those from Allosaurus. They slightly overlap with those of Majungasaurus, Arcovenator and cf. Arcovenator indeterminate abelisaurids from Tunisia, the Iberian Peninsula and the Kem Kem beds of Morocco. Allosaurus teeth primarily overlap with those of Majungasaurus and the abelisaurids. Carcharodontosaurid teeth are also grouped together and partially overlapping with abelisaurid teeth. Graphical results of the discriminant analysis (DFA) with five variables (CBL, CBW, CH, MC and DC) based on the taxa-level dataset with 775 tooth crowns belonging to twenty-three taxa of Upper Jurassic to Upper Cretaceous theropods and the teeth from Algora (Guadalajara, central Spain). Credits of the silhouettes in Appendix I, 4. It is interesting to observe that Algora teeth overlap with abelisaurid teeth from a wide range of geographic locations and stratigraphic contexts. These similarities include specimens from the Lower Cretaceous of Tunisia and the Cenomanian of the Kem Kem bed in Morocco. In addition, the Algora teeth show morphological overlap with Late Cretaceous abelisaurid material from western Europe, including Arcovenator from southern France and from the Laño locality (Condado de Treviño, Spain), as well as indeterminate Abelisauridae from the South Pyrenean Basin and from the Poyos site in Guadalajara (Spain). These morphological affinities also extend to Majungasaurus, an Upper Cretaceous abelisaurid from Madagascar (Fig. 5). In the first discriminant analysis (Appendix I, 3, table II), we considered eight family levels, Allosauridae, Carcharodontosauridae, Abelisauridae, Dromeosauridae, Troodontidae, Richardoestesia, Tyrannosauridae and Megalosauridae. Algora teeth were left without assignation so the analysis could place them. The overall classification accuracy was 92.6% 14 Algora teeth were placed in the Abelisauridae family, 2 in Allosauridae and one in Carcharodontosauridae In the second analysis (Appendix I, 3, table III), the Algora teeth were left unassigned, and the analysis assigned 11 of them (of the 17 considered in the analyses) to Abelisauridae indet. (Abelisauridae indet. teeth from Early Cretaceous of Tunisia, from the Cenomanian of the Kem Kem beds, as well as Late Cretaceous abelisaurid teeth from France and the Iberian Peninsula). Four of the Algora teeth were assigned to Allosaurus, one to Arcovenator and one to Acrocanthosaurus. The overall classification accuracy in this case was 84.7%. In the third discriminant analysis (Appendix I, 3, table IV), the Algora teeth were considered as belonging to Majungasaurus. The overall classification accuracy was 83.5% at taxa level. All but six teeth were correctly classified within Majungasaurus. Two teeth were assigned to Allosaurus and four to Abelisauridae. The fourth linear discriminant analysis classified the Algora teeth within the family Abelisauridae. These results were highly consistent with the initial family-level analysis, yielding an overall classification accuracy of 91.8% (Appendix I.3, Table V). In the fifth linear discriminant analysis, the Algora specimens were treated as a distinct group. The overall classification accuracy at the taxon level was 83.9%. Of the 17 Algora teeth included in this analysis, nine were correctly classified within their own designated group, whereas four were assigned to Allosaurus and four to Abelisauridae (Appendix I.3, Table VI). These results strongly suggest that the Algora teeth can tentatively be assigned to Abelisauridae. In figure 5 Algora teeth clearly group with all the abelisaurid teeth and especially overlap with abelisaurid teeth from the Upper Cretaceous of the Iberian Peninsula. The Linear Discriminant Analyses, both at suprageneric and generic levels, also place Algora teeth mainly within Abelisauridae taxa suggesting that this assignation could be plausible. Cladistic analysis based on the lateral dentition data-matrix using the first constrained tree topology yielded 1 most parsimonious tree (MPT), with a length of 581 steps (Fig. 6a). This analysis classified the lateral teeth Algora teeth within Abelisauridae, as a sister taxon to Majungasaurus. The result of the same cladistic analysis using the second tree topology was one most parsimonious tree with a length of 588 steps and classified the teeth in the same position (Fig. 6b). Results of the cladistic analyses performed on the data matrix of Hendrickx et al. (2025) focused on the non-avian theropod dentition (148 dental characters scored in 127 saurischians) using a constrained tree and the dental material under study as floating OTUs. Results of the cladistic analysis showing the whole theropod trees are available in Appendix I, 2.1. a–b, Strict consensus trees resulting from the cladistic analysis performed on the data matrix of 56 lateral crown-based characters using the first a, and b, second tree topology and the floating taxon. c–d, Strict consensus trees from the cladistic analysis performed on the data matrix of 32 mesial crown-based characters using c, the first tree topology and d, the first tree topology and the floating taxon. The cladistic analysis based on the mesial dentition data matrix using the first tree topology yielded 4 most parsimonious trees (MPT), with a length of 338 steps and classified the mesial teeth from Algora within Abelisauridae, in a polytomy comprised of seven taxa (Fig. 6c). Nonetheless the cladistic analysis using the second tree topology yielded 5 most parsimonious trees (MPT) with a length of 339 steps, very poorly resolved, with the mesial teeth placed in a large polytomy comprising 21 taxa, some of which belong to Abelisauridae and Allosauridae (Fig. 6d). According to the cladistic analyses, the mesial teeth of Algora are defined by a single autapomorphy (Appendix I, 2.2): (Char. 24) Mesial teeth, denticles contiguous over tip or very close to the apex: present. The node that groups Majungasaurus and the teeth of Algora is defined by three synapomorphies (Appendix I, 2.2): (Char. 41) Lateral teeth, interdenticular sulci between apical denticles on the mesial carina: present, short and poorly developed; (Char. 46) Lateral teeth, large transverse undulations on the crown in some teeth pronounced and well visible with light; (Char.48) Lateral teeth, marginal undulations (i.e., short undulations adjacent to carinae) in some teeth: present and short, the mesiodistal elongation is less than four times the space separating each undulation. Discussion The isolated teeth of Algora have several characters typical of abelisaurid lateral teeth, such as a non-oriented enamel texture, distal carinae that are straight or slightly concave and centred on the distal margin, mesial carinae that are convex and extend to the cervix, labial and lingual surfaces of the crown that are symmetrically convex in mesiodistal view and presence of short interdenticular sulci adjacent to mesial and distal denticles (on those teeth with preserved denticles). The mesial dentition of abelisaurids is particularly diagnostic among theropods22,33. In these teeth, the mesial and distal carinae reach the cervix, and the basal cross-section can be J-shaped type, with a strongly convex labial surface and a sigmoid or biconcave lingual surface,salinon-shaped27,74,75 or D-shaped32,76. The only Algora teeth identified as mesial possess these characters and exhibit a salinon-shaped basal cross-section. Premaxilary teeth of Majungasaurus are labiolingually compressed, with a CBR ranging from 0.55 to 0.7. The mesial teeth of Algora have a CBR of 0.69 and 0.64. The lateral teeth of abelisaurids are ziphodont, labiolingually compressed, distally recurved, and have mesial and distal carinae with denticles. In most abelisaurids, the distal margin of the lateral crowns is very slightly convex, straight, or weakly concave22,28,33,44,42. The apex does not extend beyond the most basal point of the distal margin, a condition also observed in the teeth of Algora. Lingual or labial depressions at the basal part of the crown, or concave surfaces near the carinae, are absent in the lateral dentition of Abelisauridae. The labial and lingual surfaces are mesiodistally convex along the entire crown, with the point of maximum curvature located at mid-height. All these characters are present in the teeth of Algora. Majungasaurus has a CBR ranging between 0.48 and 0.64 The crowns of Algora have a broader CBR range, from 0.34 to 0.60. For lateral crowns, abelisaurids generally have a CHR of about 222. The teeth of Algora have a wider CHR range, between 1.26 and 2.47, whereas Majungasaurus has a CHR of about 1.75. The cross-sections of the lateral teeth in abelisaurids are lanceolate-shaped, a feature also observed in the teeth from Algora. The denticles are small and their size does not vary significantly between the mesial and distal carinae, so DSDI is near 128. Although, in abelisaurids the number of denticles may vary along the carinae, it never exceeds 20 denticles per 5 mm22. In Majungasaurus, denticle density ranges between 7 and 12 denticles per 5 mm at mid-height of both mesial and distal carinae22. In the teeth from Algora, the denticle density is approximately 13–16 denticles per 5 mm on both carinae. The shape of the denticles may vary among abelisaurid theropods. In most taxa, the denticles margin are symmetrically convex. Denticles may be diagonally oriented and apically inclined, as in Majungasaurus, or positioned perpendicular to the margin of the crown22. The denticles in the teeth of Algora are asymmetrically convex and oriented perpendicular or inclined to the margin of the crown. The distal denticles are subquadrangular to subrectangular and mesiodistally elongated, a condition also present in some teeth of Majungasaurus22,33. The mesial denticles are similarly subrectangular to subquadrangular in shape. Wear facets are commonly found on the apical part of the crown in the teeth of abelisaurids. It is important not to confuse them with spalled surfaces, which are formed by tooth-to-food contact, whereas wear facets result from tooth-to-tooth contact and are restricted to the apical part of the labial or lingual surface of the crown73. Three of the teeth of Algora present wear facets (Fig. 6c). The superficial enamel texture in abelisaurid teeth, as examined by Hendrickx et al.28, is irregular and non-oriented. This type of texture is characteristic of abelisaurids and contrasts with the oriented and braided enamel texture observed in many other theropod taxa, such as Ceratosauridae, Megalosauroidea, Neovenatoridae and Carcharodontosauridae37, Hendrickx et al., 2015c77. Most of the teeth from Algora exhibit transverse undulations, a feature quite common in abelisaurids (e.g28,31,39,Isasmendi et al43,44). One of the considered mesial teeth from Algora (ALG-193) presents marginal undulations (Fig. 6a), which is often the case in mesial abelisaurid teeth48. A new theropod, Caletodraco cottardi, was recently described from Normandy, France78. An isolated tooth referred to this taxon has also been described. However, it is embedded in the matrix and incomplete, so its measurements could not be taken for inclusion in the statistical analyses. This isolated tooth is lateromedially compressed, with a straight distal margin and a convex mesial margin. Both carinae are serrated, with 14 denticles per 5 mm that are apically inclined and vary in size toward the apex. The tooth lacks interdenticular sulci, and the denticles are more widely separated than those of the Algora teeth. These differences suggest that the Algora specimens belong to a taxon distinct from Caletodraco. However, because the tooth from Normandy was found isolated their diagnosis is not considered sufficient. Principal Component and Discriminant Analyses show that the Algora teeth fall clearly within Abelisauridae, and, in the fifth linear discriminant analysis, they are also recovered mainly as a distinct group, suggesting they could represent their own taxon. Algora teeth overlap with the morphospace of other abelisaurid taxa, and even when not pre-assigned to any group, the analyses consistently classify the Algora teeth as belonging to Abelisauridae. These results align with previous studies in which discriminant analyses were used to classify seven of these teeth as cf. abelisaurids2, clearly distinguishing them from Carcharodontosauridae3. The identification of the isolated Algora teeth as abelisaurids is further supported by cladistic analyses, which for the first time confidently place them within Abelisauridae and suggest a close relationship to Majungasaurus. The skeletal remains of theropods from Algora contribute to the scarce Cenomanian record of Europe. Despite the archipelago condition that Europe would have during the Cenomanian, similar fauna of theropods dinosaurs, at a broad taxonomic level, can be found throughout these landmasses. This Cenomanian fauna appears to represent a continuation of Lower Cretaceous or even older assemblages in Europe7. These ancient European theropod faunas were the result of the evolution of isolated tetanuran members7. However, new taxa have been found that suggest connections within Laurasia, such as the presence of dromaeosaurids and trodoontids79,80. Another clade of theropod dinosaurs thought to have dispersed to Europe from Gondwanan latitudes during this time is the abelisaurid group. Abelisaurid fossil remains have been found mostly in meridional continents that were part of the supercontinent Gondwana, with a more limited record in Europe. Until now the presence of abelisaurids in this continent was described in many localities from both isolated teeth (e.g27Malafaia et al., 20162,42–44; and other more complete skeletal remains. The fossil record of the Abelisauridae in Europe extends from the Albian to the Maastrichtian (Buffetaut78,81, including Genusaurus sisteronis82, from the Albian-Cenomanian,Caletodraco cottardi, from the Cenomanian78, and Arcovenator escotae36, from the Campanian. The identification of Catelodraco and Genusaurus as members of Furileusauria (Buffetaut78,81) and Arcovenator placed within Majungasaurinae (Tortosa et al., 2014) could suggest the presence of two distinct lineages of Abelisaurids in Europe although their coexistence is still uncertain81. Based on these findings, the identification of the isolated theropod teeth from Algora, using multivariate statistical analyses (PCA and discriminant analysis) and cladistic analyses, as belonging to a taxon closely related to the genus Majungasaurus is considered plausible. This finding is highly significant, as it confirms the presence of the clade Majungasaurinae during the Cenomanian and supports the hypothesis that at least two distinct abelisaurid lineages coexisted during this time83. Conclusions A sample of previously unpublished isolated theropod teeth from the Spanish Cenomanian site of Algora is identified here as belonging to Abelisauridae, based on the results of cladistic, principal components (PCA) and discriminant analyses. These teeth show a unique combination of features found exclusively on abelisaurid lateral and mesial teeth, including a straight distal carina, symmetrically convex labial and lingual surfaces, a mesial carina that is centred on the mesial margin of the crown and extends to the cervix, and an irregular, non-oriented enamel texture. Additional diagnostic traits include a salinon-shaped basal cross-section in mesial teeth, presence of interdenticular sulci, and enamel undulations. The identification of the isolated theropod teeth from Algora as belonging to a taxon closely related to Majungasaurus is therefore considered plausible based on the combined analytical results. This interpretation is consistent with a dispersal of this theropod lineage from Gondwana into Europe during the Cenomanian, or possibly earlier and supports the hypothesis of two different lineages of abelisaurids coexisting in Europe during the Late Cretaceous . Declaration Supplementary Information Supplementary Information. Abbreviations Apical length Crown angle Crown apex Crown base length Crown base width Crown height Disto-apical denticle density Disto-basal denticle density Disto-central denticle density Distal carina Number of distal denticles per millimetre Number of mesial denticles per millimetre Interdenticular sulci Mesio-apical denticle density Mesio-basal denticle density Mesio-central denticle density Mesial carina Mid-crown length Mid-crown width Marginal undulations Wear facet Author contributions Mirella López-Miguel: Writing – review & editing, Writing – original draft, Methodology, Investigation, Data curation, Formal analysis, Conceptualization. Angélica Torices: Writing –review & editing, Writing – original draft, Investigation, Funding acquisition, Formal analysis, Data curation. Elisabete Malafaia: Writing – review & editing, Methodology. Francisco Ortega: Writing – review & editing, Supervision, Funding acquisition. Adan Perez-Garcia: Writing –review & editing, Writing – original draft, Supervision, Funding acquisition, Conceptualization. Funding Angélica Torices and Mirella López Miguel are supported by Ministerio de Ciencia, Inovacion y universidades, Proyecto Consolidación Investigadora 2023 CNS2023-144950. This research has been partially funded by the Spanish Ministry of Science, Innovation and Universities through project PID2023-148083NB-I00, the Regional Government of Castile-La Mancha (SBPLY/24/180801/000043 and SBPLY/25/180801/000043) and CEECIND/01770/2018 of the Fundação para a Ciência e a Tecnologia (FCT) Data availability Data will be made available online and on request to Mirella López Miguel (mirelope@ucm.es) Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Information. Data Availability Statement Data will be made available online and on request to Mirella López Miguel (mirelope@ucm.es)