[Palaeontology, 2022, e12597] Feeding habits of the Middle Triassic pseudosuchian Batrachotomus kupferzellensis from Germany and palaeoecological implications for archosaurs by EUDALD MUJAL 1 , 2 , CHRISTIAN FOTH 3 , ERIN E. MAXWELL 1 , DIETER SEEGIS 1 and RAINER R. SCHOCH 1 , 4 1 Staatliches Museum f€ur Naturkunde Stuttgart, Rosenstein 1, 70191 Stuttgart, Germany; eudald.mujalgrane@smnsw-bw.de, erin.maxwell@smnsw-bw.de, dieter.seegis@smnsw-bw.de, rainer.schoch@smnsw-bw.de 2 Institut Catala de Paleontologia Miquel Crusafont, ICTA-ICP building, c/ de les columnes s/n, 08193 Cerdanyola del Valles, Catalonia, Spain 3 Department of Geosciences, University of Fribourg, Chemin du Musee 6, 1700 Fribourg, Switzerland; christian.foth@gmx.net 4 Fachgebiet Pal€aontologie, Institut f€ ur Biologie, Universit€at Hohenheim, Wollgrasweg 23, 70599 Stuttgart, Germany Typescript received 5 August 2021; accepted in revised form 23 December 2021 Abstract: Bite traces on fossil bones are key to deciphering bones, and their relative abundance. In addition, tooth macro- feeding ecology and trophic interactions of vertebrate past scopic wear and microanatomy are proven as independent ecosystems. However, similarities between traces produced by lines of evidence of feeding ecology. Comparing bite traces on different carnivorous taxa with similar dentitions, and mis- fossil and present-day bone assemblages, we observe that bone identifications due to equifinality, hinder confident identifica- modifications by the crocodylomorph lineage (from Triassic tions of the bite makers. Here, we correlate bite traces with pseudosuchian archosaurs to extinct and extant crocodylians) macroscopic wear and microanatomy of the teeth of the pseu- are strikingly similar, including taxa with and without zipho- dosuchian archosaur Batrachotomus kupferzellensis from the dont teeth. Such a set of features differs from bone modifica- Triassic Lower Keuper fossil lagerst€atten (southern Germany), tion assemblages produced by taxa with similar ziphodont untangling its feeding habits and shedding light on the bite teeth outside the pseudosuchian lineage, such as theropod traces generated by ziphodont teeth (teeth with serrated cari- dinosaurs and the Komodo monitor, suggesting phylogeny is a nae). Individually, bite traces reflect tooth morphology, better predictor of feeding ecology among saurian reptiles than whereas composite bite traces and their frequency are related tooth morphology. to feeding behaviour and explain tooth macroscopic wear and microanatomy. Therefore the identification of the bite maker Key words: bite traces (‘bite marks’), ziphodont teeth, is possible by analysing composite bite traces, their location on tooth macrowear, tooth histology, archosaur, Lower Keuper. D I R E C T evidence of feeding ecology and trophic interac- dentition evolved convergently in different amniote tions of fossil vertebrates mostly relies on the identifica- clades (Whitney et al. 2020) but produces similar bite tion of bite traces on bones (Fiorillo 1991; Jacobsen 1998; trace morphotypes (D’Amore & Blumenschine 2009). Pobiner 2008; Drumheller et al. 2020), stomach contents Taxa with ziphodont teeth include archosauromorphs and processed food in the form of coprolites and regurgi- (mostly purported terrestrial carnivores, but also phyto- talites (Qvarnstr€ om et al. 2019; Gordon et al. 2020). saurs and some marine crocodylomorphs), Varanus Regarding bite traces, however, the identification of the komodoensis (the extant Komodo monitor) and gorgo- producer (the bite maker) commonly remains impossible nopsian therapsids (D’Amore & Blumenschine 2009, (Hone & Chure 2018) and can usually only be confidently 2012; Pirrone et al. 2014; Brink et al. 2015; Hone & identified at a major clade level (Mikulas et al. 2006; Chure 2018; Whitney et al. 2020). Interpretations of Jacobsen & Bromley 2009; Pirrone et al. 2014; Augustin feeding ecology and trophic interactions based on bite et al. 2020; Drumheller et al. 2020; Drymala et al. 2021). traces of extinct taxa with ziphodont teeth are often lim- In addition, even if carcass utilization by carnivores is ited to non-avialan theropod dinosaurs (Fiorillo 1991; recognized, feeding habits may still remain uncertain Jacobsen 1998; Jacobsen & Bromley 2009; Drumheller (Hone & Rauhut 2010; Drumheller et al. 2014, 2020). et al. 2020), with additional information obtained from This is particularly the case for ziphodont teeth (teeth bite traces produced by V. komodoensis (D’Amore & with serrated carinae; Brink et al. 2015) because this Blumenschine 2009, 2012). © 2022 The Authors. doi: 10.1111/pala.12597 1 Palaeontology published by John Wiley & Sons Ltd on behalf of The Palaeontological Association. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. 2 PALAEONTOLOGY Here, based on the tetrapod collection from the Lower and brown colours, whereas those from Vellberg- Keuper (Erfurt Fm., Ladinian, Middle Triassic) Kupferzell Eschenau are mainly grey-greenish and occasionally and Vellberg-Eschenau fossil lagerst€atten, south-western brown. Nevertheless, the palaeoenvironments for both Germany (Urlichs 1982; Schoch & Seegis 2016; Schoch localities are similar, being interpreted as lacustrine sys- et al. 2018), we reconstruct an exceptional palaeoecologi- tems with marine influence and/or connection to marine cal scenario by combining multiple morphological and environments, as evidenced by the presence of stratigra- statistical analyses. We correlate the diverse bite trace phically close marine carbonate units, and the presence of assemblage with the macroscopic wear and microanatomy marine fossils, such as bivalves and sauropterygian reptiles of the teeth of the up to five-metre-long pseudosuchian within the UGM (Urlichs 1982; Nitsch 2015c; Hagdorn archosaur Batrachotomus kupferzellensis (Gower 1999; et al. 2015; Schoch & Seegis 2016; Mujal & Schoch 2020; Gower & Schoch 2009), the bite maker of most of the DS, EM, RRS, pers. obs.) traces studied herein. Since bite traces are confidently cor- related to tooth macroscopic wear and therefore to microanatomical adaptations in tooth structure, our work MATERIAL AND METHOD shows that these can be independent lines of evidence to decipher feeding behaviours. With the aim of deciphering Institutional abbreviations. MHI, Muschelkalkmuseum Hagdorn, the feeding ecology behind bite traces produced by zipho- Ingelfingen, Germany; SMNS, Staatliches Museum f€ ur Natur- dont teeth through morphological and statistical analyses, kunde Stuttgart, Stuttgart, Germany. we show a linkage between bite makers’ habits and phy- logeny among archosaurs and, more broadly, saurian reptiles. Fossil sampling and analyses All prepared fossil material from the Kupferzell and GEOLOGICAL SETTING Vellberg-Eschenau fossil lagerst€atten (south-western Ger- many) housed in the SMNS and MHI collections was The fossil tetrapod bones studied in this work come from examined using a magnifying glass, a dissecting micro- two different fossil lagerst€atten, named Kupferzell and scope, and lighting in different orientations, to identify Vellberg-Eschenau (Fig. 1). These rich fossiliferous bite traces (e.g. Blumenschine et al. 1996; D’Amore & deposits correspond to the Untere Graue Mergel (UGM; Blumenschine 2009; Drumheller & Brochu 2014; Drum- lower Grey Marls), a unit within the upper part of the heller et al. 2020). Bite traces were systematically classified Lower Keuper facies from south-western Germany (Let- according to their morphology (ichnotaxonomy) follow- tenkeuper, Erfurt Formation), of Ladinian (late Middle ing Mikulas et al. (2006), Jacobsen & Bromley (2009) and Triassic) age (Urlichs 1982; Schoch & Seegis 2016) Pirrone et al. (2014). In addition, equivalences of the (Fig. 1A). In south-western Germany, the Lower Keuper identified ichnotaxa with the taphonomic terms coined by mainly consists of an alternating carbonate–siliciclastic Binford (1981) and extensively used in the literature (e.g. succession 20–25 m thick deposited in an epicontinental Pobiner 2008; D’Amore & Blumenschine 2009; Njau & platform overlying the marine carbonate Muschelkalk Gilbert 2016; Drumheller et al. 2020) are provided. Asso- facies (Brunner & Bruder 1981). This system is part of ciations between bite traces and their orientation and the Central European Basin, and during the Lower Keu- location on bones were also surveyed. All of this informa- per deposition it was occasionally flooded by marine tion can be found in Data S1 (and Mujal et al. 2022). In transgressions from the Tethys sea, whereas the main sed- taphonomic studies, bite/tooth traces are often named imentary inputs came from the northern palaeogeo- bite/tooth ‘marks’ but, since they are biogenic structures graphic highs (Etzold & Schweizer 2005; Nitsch 2015a, (i.e. ichnofossils or trace fossils), the term ‘trace’ is pre- 2015b). The UGM is an overall succession of mudstones ferred, with use of the term ‘mark’ restricted to non- and marlstones, with different layers and bonebeds well- biogenic structures (Bertling et al. 2006; Mikulas et al. characterized at both the Kupferzell and Vellberg- 2006; Jacobsen & Bromley 2009; Vallon et al. 2015). Eschenau localities (Urlichs 1982; Nitsch 2015c; Hagdorn Bite traces produced by the dragging of denticles on et al. 2015; Schoch & Seegis 2016). This unit is embedded the carinae of teeth (ichnotaxon Knethichnus parallelum) between two dolostone banks, which may also contain were measured on digital photos using ImageJ (v.1.52a) fossils (Hagdorn et al. 2015; Schoch & Seegis 2016; Mujal following D’Amore & Blumenschine (2012). Such mea- & Schoch 2020). surements account for the maximum number of denticles The two localities are stratigraphically equivalent, per millimetre, allowing for the identification of the though some lithological differences exist (Fig. 1B). The potential bite maker through comparisons with the tooth UGM layers in Kupferzell mostly present green, yellow assemblage from the Lower Keuper (Schoch et al. 2018). MUJAL ET AL.: FEEDING ECOLOGY OF ARCHOSAURS 3 FIG. 1. Geological setting. A, general stratigraphic section and geological map of the Lower Keuper in south-western Germany (mod- ified from Schoch & Seegis 2016; Schoch et al. 2018). B, detailed stratigraphic sections of the Kupferzell and Vellberg-Eschenau fossil lagerst€atten (modified and updated from Hagdorn et al. 2015; Schoch & Seegis 2016). Since most of the bite traces were assigned to Batrachoto- the maxillae and mandibles. The premaxillae of Batracho- mus kupferzellensis, teeth of this pseudosuchian archosaur tomus held four teeth each, these with a characteristic were analysed as explained below. slender shape (Gower 1999). Teeth from the premaxilla Tooth crowns of Batrachotomus from Kupferzell were almost cylindrical, with a nearly round or slightly (N = 258) and Vellberg-Eschenau (N = 56), all from the oval cross-section, and an apical portion which is slightly UGM, were analysed accounting for: (1) basic morpho- (linguo-)distally curved; their basal mesial carina often logical classification by measuring (with a digital calliper) lacks denticles (Data S2). Non-premaxillary teeth gener- the apicobasal height of the crown (or crown height, ally show diverse crown morphologies (Gower 1999; CH), the mesiodistal depth and the labiolingual width of Schoch et al. 2018), with differences in: (1) the distal cur- the crown at its base (or crown base length, CBL, and vature (symmetry in labial/lingual view); (2) the symme- crown base width, CBW, respectively; for tooth parame- try of cross-section (more or less convexity of the labial ters, see: Smith et al. 2005; Hendrickx et al. 2015); (2) side, though all are oval in cross-section); (3) curvature identification of the position within the jaw, distinguish- towards the lingual side; and (4) relative height of the ing premaxillary from non-premaxillary teeth (also distin- crown with respect to the mesiodistal depth of the crown guishing teeth from the dentaries and maxillae, when base. Both mesial and distal carinae possess denticles. possible). In order to identify the position of teeth within However, due to the isolated nature of most of the teeth, the jaws, we examined the dentigerous elements of Batra- in most cases it was not possible to identify whether the chotomus still preserving attached teeth and also followed non-premaxillary teeth were from the maxillae (with 11 tooth descriptions from Gower (1999) and Schoch et al. alveoli each) or the mandibles (with probably 12 alveoli (2018). Generally, premaxillary teeth display distinctive each) or from which tooth position. Only teeth still features allowing them to be distinguished from those of attached to maxillae or mandibles, as well as some teeth 4 PALAEONTOLOGY with a recurved labial side of the crown and straight pos- prepared in longitudinal section capturing either the terior edge that are characteristically from the anterior mesial or distal carina. The crowns of two non- part of the maxilla (Gower 1999), can be more confi- premaxillary teeth (SMNS 97009/11, 97009/17), with five dently positioned. According to Gower (1999, p. 38), thin sections each, were prepared in cross-section at vary- teeth from the posterior part of the maxilla are more ing points along the apicobasal axis. Thin sections were strongly convex anteriorly and have shorter crowns than prepared following standard palaeohistological methodol- those from the anterior part; however, these features seem ogy, with thicknesses ranging from 60 to 75 µm, and to be present in some mandibular teeth. In addition, examined under a petrographic microscope (Leica Schoch et al. (2018, p. 621) noted differences in tooth DM750P). Photographs were taken using an attached length across maxillae and dentaries, with the longest ele- Leica ICC50 W camera, and processed using Leica ments in the mid-portion of maxillae and anterior por- LAS EZ (v.3.1.0) software. tion of dentaries and the smallest teeth in the anterior and posterior ends of maxillae. Nevertheless, the isolated nature of most of the teeth precludes their association to Statistical analysis individuals, meaning that they could belong to either juvenile or mature/adult animals because teeth could be To investigate potential statistical differences regarding proportionally smaller or larger, respectively. Of note, all bite trace patterns on different bitten taxa (Mastodon- preserved dentigerous elements show that tooth replace- saurus vs Plagiosuchus vs Nothosaurus vs Batrachotomus), ment in Batrachotomus occurred at alternating tooth posi- body regions (skull vs teeth vs free vertebrae vs sacrum vs tions (Gower 1999). ribs vs pectoral girdle vs pelvic girdle vs limbs) and bone Tooth macroscopic wear (macrowear) was classified in regions (proximal end vs distal end vs side (widest area) two ways (Table 1): qualitatively, distinguishing wear, of the shaft vs edge (narrowest area) of the shaft vs bone notches, spalling and breakage, as well as noting the areas edge vs bone centre), we converted the data collected in of the tooth crown with greater amounts of wear (lingual the bite trace database (Data S1) into three separate data- or labial, and mesial or distal); and (semi-)quantitatively, sets, scoring the absence or presence of each bite trace scoring the degree of tooth wear as none (0), low (1), morphotype for all bones within each group. A small high (2) or extreme (3). All tooth measurements and fraction of bones were not included in the analyses macrowear are integrated in Data S2 (see also Mujal et al. (Data S1) because the elements could not be identified 2022). Based on the UGM tooth sample, we compared due to their fragmentary nature or preservation. These macrowear between premaxillary and non-premaxillary datasets were analysed using a permutational multivariate teeth in a box-plot by averaging the amount of macro- analysis of variance (PERMANOVA). This method estimates wear (from none to extreme) of all teeth. In addition, we the potential overlap between two or more groups by compared regional differences in macrowear (using aver- testing the significance of their distribution on the basis ages) along the mesial and distal carinae and the tip in a of permutation (9999 iterations) and the Euclidean dis- box-plot. The number of occurrences (counts) of abra- tance as distance measures. In contrast to parametric sion (wear and notches) for the tips and each portion of tests, PERMANOVA does not require normal distribution of both mesial and distal carinae were counted and dis- the data (Anderson 2001; Hammer & Harper 2006). The played as percentages for each region of the tooth crown, spatial relationship between groups is expressed by an also differentiating between premaxillary and non- F-value and a Bonferroni-corrected p-value. Following premaxillary teeth. Wills et al. (1994), the three datasets were further trans- formed into Euclidean distance matrices and subjected to a principal coordinate analysis (PCOA). This method Tooth histology reduces multivariate data down to a new set of indepen- dent variables (principal coordinates) that are linear com- To complement the interpretations based on the macro- binations of the original set with zero covariance scopic wear of Batrachotomus teeth (see Discussion, (Hammer & Harper 2006), providing a ‘biting space’. below), we analysed their microanatomy, also comparing Afterwards, we ran a broken-stick method for each PCOA them to other taxa with ziphodont teeth (Brink et al. to obtain the number of principal coordinates (PCOs) 2015; Wang et al. 2015; Whitney et al. 2020). Histological that contain the relevant amount of total variation thin sections were prepared from isolated Batrachotomus (De Vita 1979; Jackson 1993). For these PCOs we applied teeth originating from the UGM and showing different a linear discriminant analysis (LDA), which reduces the degrees of macroscopic tooth wear. Four teeth (premaxil- number of PCOs to a smaller set of dimensions by maxi- lary: SMNS 97009/8; non-premaxillary: SMNS 97009/9, mizing the separation between the given groups using the 97009/10, 97009/16), with one thin section each, were Mahalanobis distance. This distance measure is estimated MUJAL ET AL.: FEEDING ECOLOGY OF ARCHOSAURS 5 TABLE 1. Types and amount of macroscopic wear as classified in this work. Macroscopic Description Amount of macroscopic wear wear types Low (1) High (2) Extreme (3) Wear Abrasion Denticles still recognizable, Denticles abraded to their Carina completely Notches Localized abrasion only partially abraded base removed, abrasion extending beneath the interdental sulcus Spalling Breakage of relatively small Small flake/s detached, Large flakes, covering a Large flakes, covering a and thin conchoidal/oval- original tooth large portion of the large portion of the shaped flakes morphology still easily tooth, but morphology tooth, morphology not recognizable still recognizable recognizable Breakage Loss of tooth portions As for spalling, but broken portions lack conchoidal/ oval fractures Lack of wear is scored as none (0). from the pooled within-group covariance matrix, result- RESULTS ing in a linear discriminant classifier and an estimated group assignment for each species. These results were Description of bite trace morphotypes on Lower Keuper cross-validated using Jackknife resampling (Hammer & bones Harper 2006; Hammer 2020). Based on the confusion matrix (Stehman 1997), we estimated the error of correct The presence of bite traces on Lower Keuper bones has identification. previously been mentioned (Wild 1978, 1979, 1980; Finally, given that the microanatomy of Batrachotomus Schoch & Seegis 2016) but never analysed in detail. A teeth is strikingly similar to that of other hypercarnivores total of 189 bones and teeth (178 from Kupferzell and 11 with ziphodont teeth (see Microanatomy of Batrachotomus from Vellberg-Eschenau), all from the Untere Graue Mer- teeth, below), we statistically compared the overall tooth gel (UGM; Wild 1980; Urlichs 1982; Schoch & Seegis morphology or morphometry of Batrachotomus kupferzel- 2016), yield at least one bite trace (Data S1). Bite traces lensis with that of other archosauromorphs by combining are preserved on bones of Mastodonsaurus (N = 141), our measurements with the data from Hoffman et al. Plagiosuchus (N = 5), Nothosaurus (N = 2), ?Chroniosu- (2019) on archosauromorph teeth and Hendrickx et al. chidae (N = 1), and Batrachotomus (N = 40). Since bite (2015) on theropod teeth. Because both datasets include traces are trace fossils (Bertling et al. 2006; Pirrone et al. different sets of parameters, we restricted the comparison 2014; Vallon et al. 2015), they are ichnotaxonomically to the crown height (CH), crown base length (CBL), crown classified. We identified five bite trace morphotypes base width (CBW), number of denticles on the mesial mid- referred to previously described ichnotaxa (Mikulas et al. crown (MC) and number of denticles on the distal mid- 2006; Jacobsen & Bromley 2009) (Figs 2–6; Data S1): crown (DC). The latter two parameters were measured for Nihilichnus nihilicus, round/oval punctures produced by a subset of 23 teeth from the Batrachotomus kupferzellensis tooth tips; Linichnus serratus, linear grooves with serrated sample (Data S2). In addition, we created three ratios margins produced by serrated carinae; Knethichnus paral- between CBL and CBW (CBR), CH and CBL (CHR), and lelum, regular and parallel grooves produced by denticles MC and DC (MDCR). Specimens with incomplete mea- of carinae; Brutalichnus (two morphotypes), large holes surements were excluded from the final dataset. All mea- reaching cancellous bone indicating strong bites; and surements and ratios were log-transformed and analysed Machichnus-like, multiple small punctures connected with with LDA (including Jackknife resampling) and PERMANOVA short grooves produced by relatively small teeth. (with 9999 iterations and Bonferroni-correction for p-values). In addition, we ran the LDA with all Batrachoto- Ichnotaxon Nihilichnus nihilicus Mikulas et al., 2006. Cir- mus teeth grouped as ‘unknown’ and classified them based cular to oval-shaped or fusiform trace that may perforate on the training set, including the remaining specimens the bone cortex (‘punctures’ of Binford 1981 and from the two original datasets. All statistical analyses and D’Amore & Blumenschine 2009), destroying it and pene- ordination methods were performed with the program trating the cancellous bone (Fig. 2A, F, J). Alternatively, PAST v.4.03 (Hammer et al. 2001). the tooth may just incise the bone (‘pits’ of Binford 1981 6 PALAEONTOLOGY and D’Amore & Blumenschine 2009), bending and press- cortical bone), as occurs with Nihilichnus; thus, even if all ing the bone cortex into the spongiosa or superficially of these linear grooves correspond to the same ichno- breaking it into flakes that remain in adherence with one taxon the markedly shallower traces, although classified as another and with the spongiosa; in some cases depressing Linichnus, are termed ‘scores’ (as in D’Amore & Blu- fractures are present due to the collapse of the bone sur- menschine 2009). The traces of L. serratus within the rounding the tooth impact area (Drumheller & Brochu Lower Keuper assemblage range from short (2–3 mm; 2016). Both circular and oval traces are considered to be ‘simple dental hacks’ of Jacobsen & Bromley 2009) to rel- N. nihilicus, independent of degree of penetration into atively long (c. 1 cm, and up to 2 cm) grooves that are the bone. Nonetheless, the notably shallower bite traces usually straight to slightly curved; they can display short are here termed ‘pits’ (as in Binford 1981 and D’Amore striations parallel to the main groove. This trace is & Blumenschine 2009) to distinguish them from the formed by the serrated cutting edge of the tooth crown more penetrating ones, as they may represent different (i.e. from teeth possessing denticles on the carinae) and/ behaviours of the bite maker (mostly linked to the or the dragging of the (serrated) tooth tip on the bone strength of the bite). Within the area contacted by the (D’Amore & Blumenschine 2009; Jacobsen & Bromley tooth, the cortical bone is usually concentrically fractured, 2009; Pirrone et al. 2014), with the tooth moving parallel with radial fractures accommodating the displaced frag- to the longitudinal axis of the crown; in other words, the ments within the bone. In traces with an oval outline carina (usually the mesial but occasionally the distal; see (usually with pointed ends), the displaced portion of the Discussion, below) is the main portion of the tooth trac- bone is symmetrically broken and displays a more ing the bone. depressed, central, straight ridge dividing the bitten area in two; they correspond to ‘bisected pits’, produced by Ichnotaxon Knethichnus parallelum Jacobsen & Bromley, laterally compressed teeth (e.g. D’Amore & Blumenschine 2009. Traces consisting of sets of parallel striations result- 2009; Njau & Gilbert 2016; Drymala et al. 2021). If the ing from dragging of the denticles of teeth carinae on the (fractured) cortical bone is still attached in the bite bone surface (D’Amore & Blumenschine 2009, 2012; impact area, it displays a main straight fracture running Jacobsen & Bromley 2009; Pirrone et al. 2014) (Fig. 2A, along the long axis of the trace, with concentric fractures B, D–I). They are very characteristic traces exclusively outlining the bitten area. In extreme cases, the bite results produced by ziphodont teeth, equivalent to the ‘striated in the formation of a sharp, deep, hole that penetrates marks’ and ‘striated furrows’ of D’Amore & Blu- the bone (but without mixing cortical and cancellous menschine (2009). In the Lower Keuper assemblage, sets bone) and preserves the transverse outline of the biting of parallel striations can be several centimetres long (up tooth, of fusiform shape due to the tooth carinae to 3.5–4 cm; Fig. 2G). They are either rectilinear, gently (Fig. 2A), allowing its shape, and thus the bite maker, to arched or markedly curvilinear and/or sinuous (the latter be identified (Drumheller et al. 2014) (see Identification resembling the ‘hook marks’ produced by crocodylians, of the bite maker, below). This bite trace morphotype see Njau & Gilbert (2016), even though their tooth cari- would result from the tooth tip piercing the bone nae do not possess denticles); also, a single trace can be (Mikulas et al. 2006; D’Amore & Blumenschine 2009), rectilinear in one part and become strongly curved in the mostly with a vertical movement of the tooth towards the other part (as do some ‘striations with internal striae’ bone in the direction of the crown longitudinal axis. All produced by crocodylians; Njau & Gilbert 2016). They traces with this outline on the Lower Keuper bones fit are in groups of 3 to up to 13 parallel striations, though with teeth of Batrachotomus kupferzellensis. this number can vary also in a single trace, reflecting the changing number of denticles in contact with the bone as Ichnotaxon Linichnus serratus Jacobsen & Bromley, a result of the curvature of the tooth carinae and/or the 2009. Traces composed of linear grooves with characteris- bone. This morphotype results from the transverse to tic serrated margins and a U- or V-shaped transverse sec- oblique movement of the tooth along the bone (D’Amore tion on the cortical bone (Fig. 2A–C, E). They are & Blumenschine 2012), thus, the movement is perpendic- equivalent to the ‘furrows’, ‘scores’ and ‘edge marks’ of ular to oblique with respect to the longitudinal axis of D’Amore & Blumenschine (2009) (see also Bindorf 1981). the tooth crown (i.e. the movement is mostly perpendicu- The ‘dental hacks’ described by Jacobsen & Bromley lar to that producing Linichnus serratus). (2009) may also correspond to short Linichnus, and are likely to be equivalent to ‘edge marks’. Nomenclatural Ichnotaxon Brutalichnus Mikulas et al., 2006. Traces cor- differences rely on the extension of the bite trace (from a responding to relatively large holes due to the removal few mm to up to 2 cm long) and the incision into the of portions of both cortical and cancellous bone. This bone (i.e. just deforming or completely removing the ichnogenus is recorded by two clearly differentiated MUJAL ET AL.: FEEDING ECOLOGY OF ARCHOSAURS 7 A B C D E F G J H I K FIG. 2. Selected bite traces on Lower Keuper bones and interpretative outlines. A, punctures, grooves and drags on opposing surfaces (dorsal, left; ventral, right) of a right femur proximal end. B, aligned and N/Z-shaped traces on a right rib. C, V-shaped trace on a rib. D, pivoting drag trace within a rhomboid-shaped flake on a rib. E, associated grooves and drags on a right fibula. F, drag trace within a deep flake and puncture on a right radius. G, elongated drag trace on a left ischium. H, fan-shaped drag trace on a rib. I, curved and linear drag traces and deep trace on a right rib. J, V/triangular-shaped deep trace and punctures on an ischium. K, multiple small traces on a rib. A–B, F, I–J, Mastodonsaurus: SMNS 81169 (A), 97010/1 (B, I), 84256 (F), 84244 (J). C–E, G, K, Batrachotomus: SMNS 97013/1 (C), 97013/2 (D), 80277 (E), 80268 (G), 97014 (K). H, ?Chroniosuchidae: SMNS 97012. Abbreviations: B1/B2, Brutalichnus morphotypes 1/2; K, Knethichnus parallelum; L, Linichnus serratus; M, Machichnus-like; N, Nihilichnus nihilicus. All scale bars represent 10 mm. morphotypes, here named morphotypes 1 and 2. The morphotype is represented by deep and elongated traces first one consists of relatively large and deep round to (also entirely removing the cortical bone) widening oval-shaped holes removing the cortical bone entirely towards the edge of the bone, outlining a triangular- and reaching the spongiosa (Fig. 2I). The second or V-shaped trace (Fig. 2J). The margins are roughly 8 PALAEONTOLOGY FIG. 3. Distribution of bite traces on bones from the Lower Keuper 1. A, right humerus with punctures on both ends and grooves on the shaft. B, right radius with punctures and deep traces on both ends and abundant drag traces on edge of the shaft (drag trace and puncture on the top of the inset photograph in B are shown in Fig. 2F). C–E, femora severely bitten, displaying diverse ichnotaxa and composite traces (proximal end of E shown in Fig. 2A). F, right fibula with puncture on the distal end, and long and thin drag traces on mid-shaft; arrow with L+K indicates position of traces shown in Figure 2E (on the opposite side, not visible in this image). A–E, Mastodonsaurus: SMNS 84255 (A), 84256 (B), 81361 (C), 84253 (D), 81169 (E). F, Batrachotomus: SMNS 80277. White arrows indicate position of enlarged images. Abbreviations: B1/B2, Brutalichnus morphotypes 1/2; K, Knethichnus parallelum; L, Linichnus serratus; M, Machichnus-like; N, Nihilichnus nihilicus. All scale bars represent 10 mm. serrated; sometimes fracture lines which are concentric bones, this bite morphotype may also have resulted to the area of impact, and depressed fractures (Drum- from dismembering or ripping off portions of the heller & Brochu 2016) are present. In both morpho- carcass. types, cortical bone may be still embedded (mixed) with cancellous bone within the deep holes. To gener- Ichnotaxon Machichnus (Machichnus-like) Mikulas et al., ate the second morphotype, apart from the stark bone 2006. Traces characterized by relatively small (1–2 mm perforation, the tooth would also have been forcibly long) hacks (in the sense of Jacobsen & Bromley 2009) pulled back, ripping off the bone. This ichnotaxon found in abundance and aligned when present (only three probably resulted from a strong stroke of the bite- bones bear this type of bite trace; Data S1) (Fig. 2K). The maker on the bone, possibly for the purpose of feeding traces, which only affect the cortical bone, consist of small on the marrow (Mikulas et al. 2006), though consider- pits or punctures associated with a relatively short and ing the distribution of these traces in the Lower Keuper straight groove with a pointed end, which may MUJAL ET AL.: FEEDING ECOLOGY OF ARCHOSAURS 9 B E A D F C H G I J K FIG. 4. Distribution of bite traces on bones from the Lower Keuper 2. A, ilium severely traced. B, edge of the shaft of a rib with a deep diamond-shaped flake displaying drag traces as in the inset of A. C, ilium with a deep flake as in the insets of A and B. D, rib with shallow drag traces outlining the same tooth movement as the flake with drags in C. E, rib with a transitional trace from groove to drag, showing the removal of a flake of bone with denticles dragged within. F, ilium with a groove potentially in transition with dragged denticles. G, ilium proximal end with a V-shaped deep trace and puncture. H, ischium (opposing side of this bone is shown in Fig. 2J) with a V-shaped trace and four aligned punctures fitting with the outline of a medium-sized Batrachotomus premaxilla. I, right ischium (pair of that of Fig. 2G) with a V-shaped trace and a linear groove (inset). J–K, proximal (J) and distal (K) ends of a right pubis with parallel grooves. A–E, G–H, Mastodonsaurus: SMNS 81342 (A), 81104 (B), 81349 (C), 81109 (D), 97010/2 (E), 81355 (G), 84244 (H). F, Plagiosuchus: SMNS 92055. I–K, Batrachotomus: SMNS 80268 (I), 52970 (J–K). Transitions between ichnotaxa are also observed (thin arrows between letters). Abbreviations: B1/B2, Brutalichnus morphotypes 1/2; K, Knethichnus parallelum; L, Linichnus serratus; M, Machichnus-like; N, Nihilichnus nihilicus. Large white arrows indicate position of magnified images; straight grey arrows indicate bite traces; curved grey arrows indicate position of trace on opposite side of element. All scale bars represent 10 mm. occasionally result in a T-shaped trace. Some grooves dis- of the Lower Keuper with this morphology are referred to play serrated margins, indicating that they were produced as Machichnus-like, as the identification of this ichno- by (small) serrated teeth. In its diagnosis, this ichnotaxon taxon is tentative. was interpreted as a result of gnawing (Mikulas et al. 2006) but, considering the specimens reported herein, it would more broadly represent repeated strokes on the Associations of bite traces from the Lower Keuper bone. Interestingly, a ‘gnawing-like’ behaviour has recently been reported in a small-bodied theropod dino- All the previously described biting ichnotaxa are associ- saur (Brown et al. 2021); the bite traces reported herein ated, with traces from the same and from different (also associated with a small-bodied tetrapod, probably morphotypes, and show gradual transitions between an archosauriform; see Identification of the bite maker, types or are found in association with each other below) may represent a similar behaviour. In any case, (being produced by the same taxon and/or individual; considering the potential high disparity of behaviours Figs 2A–J, 3–5; Data S1), with the exception of the (and wide range of potential bite makers), the bite traces Machichnus-like traces (Figs 2K, 6). The analysed bones 10 PALAEONTOLOGY A G H B C I D E F J M K L N O S Q P R T usually preserve several bite traces, in contrast to typical The most common association is a transition from non-avialan theropod-traced bone assemblages (Fiorillo Linichnus to Knethichnus, the long axes of each trace mor- 1991; Jacobsen 1998; Pobiner 2008; Hone & Rauhut photype being perpendicular to oblique from each other 2010; Augustin et al. 2020; Drumheller et al. 2020). (Fig. 2B, E). Transitions between Linichnus and MUJAL ET AL.: FEEDING ECOLOGY OF ARCHOSAURS 11 F I G . 5 . Distribution of bite traces on bones from the Lower Keuper 3. Severely bitten bones corresponding to ribs (A–K, M–T) and a left cleithrum (L). A–D, F–J, L–M, O–Q, S–T, Mastodonsaurus: SMNS 97010/1 (A; traces on dorsal side of the shaft and end of rib shown in Fig. 2B, I), 80938 (B), 97010/3 (C), 84188 (D), 97010/4 (F), 81104 (G), 80915 (H), 80923 (I), 81139 (J), 81210 (L), 81123 (M), 81137 (O), 81136 (P), 97010/5 (Q), 97010/6 (S), 82018 (T). E, N, Batrachotomus: SMNS 97013/3 (E), 97013/4 (N)). K, ?Chronio- suchidae: SMNS 97012. R, Nothosaurus: SMNS 80266. Black arrows point to the proximal ends of bones. Transitions between ichno- taxa are also observed (thin arrows and hyphens between letters). Abbreviations: B1/B2, Brutalichnus morphotypes 1/2; K, Knethichnus parallelum; L, Linichnus serratus; M, Machichnus-like; N, Nihilichnus nihilicus. Large white arrows indicate position of magnified images; straight grey arrows indicate bite traces; curved grey arrows indicate position of trace on opposite side of element. All scale bars represent 10 mm. A C B D F E F I G . 6 . Bite traces not produced by Batrachotomus on Lower Keuper bones. Bitten bones of Mastodonsaurus (A–D) and Batrachoto- mus (E–F). A, palate (SMNS 80704) in ventral view with close up (dashed black square) of the paired Nihilichnus. B, mandible (SMNS 92128) with paired Nihilichnus similar to those in A. Dashed grey circles in A and B outline the approximate shape of the biting teeth (probably a pair of tusks of Mastodonsaurus). C, palate (SMNS 81310) with slightly sinuous scores (grooves) without serrated margins, differing from those produced by teeth with serrated carinae (i.e. generating the ichnotaxon Linichnus serratus). D, left cleithrum (SMNS 81210) with small cluster of Machichnus-like traces on the left part of the photograph, and Knethichnus parellelum pointed with an arrow and ‘K’ on the right (see Fig. 5L). E–F, rib (SMNS 97013/5) with abundant Machichnus-like traces parallel to each other and with grooves (some with serrated margins) perpendicular to the rib long axis. Scale bars represent: 100 mm (A); 10 mm (inset of A, B–F). Knethichnus (e.g. Figs 2B, E, 3E, 4E) indicate that the generated by the denticles). This composite trace is some- tooth tracing the bone changed its direction of move- what similar to the bone spalling inferred for ment. Usually, in this composite trace, the length of crocodylian-induced bite traces (Njau & Gilbert 2016). Linichnus is greater than the width of Knethichnus (the Several bones preserve an association of Linichnus– latter corresponding to the set of parallel striations Knethichnus–Linichnus, with two parallel Linichnus and 12 PALAEONTOLOGY MUJAL ET AL.: FEEDING ECOLOGY OF ARCHOSAURS 13 FIG. 7. PCOA and LDA results from the bite traces. Biting spaces are compared for each bitten taxon (top), each skeletal region (limbs, pectoral girdle, pelvic girdle, ribs, sacrum, vertebrae, teeth and skull) (middle), and each bone region (proximal and distal ends, side and edge of the shaft, bone edge, and bone centre) (bottom) for the first two ordinates. Percentage of total variation for each ordinate is given in parentheses. Biplots visualizing the factor loadings of the LDA are shown next to the corresponding plot. one oblique to nearly perpendicular Knethichnus between on the edge area of the shafts, preserve two or more them, outlining an N- or Z-shaped composite trace traces of Linichnus in a V-shape (Figs 2C, D, 5A, E, F, I); (Figs 2B, 3E, 5A, C, P). Occasionally, this association is sometimes they seem to form a bifurcated trace, corre- the opposite, with Linichnus in-between Knethichnus sponding to a single trace with an abrupt change of ori- traces (Fig. 5D, S). entation with respect to the bone, as observed in traces Nihilichnus is most commonly associated with Linich- produced by Varanus komodoensis (D’Amore & Blu- nus (Data S1), although there are also Nihilichnus– menschine 2009), non-avialan theropods (Jacobsen & Knethichnus associations (e.g. Fig. 5B, P, Q, S, T); in both Bromley 2009), and crocodylians (Njau & Blumenschine cases, there is usually a relatively sharp transition between 2006; Njau & Gilbert 2016; Shale et al. 2017). In fact, both trace morphotypes, though clearly produced by the around these V-shaped Linichnus, there are usually other same tooth. A peculiar association, found in different coupled traces of Linichnus, which are oblique with bones, is that of two gently curved/arched traces of respect the long axis of the bone and extend in perpen- Knethichnus outlining a fan or an open triangle (e.g. dicular directions (Fig. 5I), being somewhat reminiscent SMNS 81109, 97012); in some cases (e.g. SMNS 81210) of the aforementioned fan-shaped composite trace of this association consists of a Knethichnus trace in each Knethichnus–(Linichnus)–Knethichnus (Figs 2H, 5K, L). outer end that changes to Linichnus in the middle of the Some bones display surfaces completely covered by composite trace, where the two traces join, or eventually long traces of Knethichnus, with the different traces being the tooth only generates Linichnus (SMNS 80923) roughly oriented in the same direction, and with only (Figs 2H, 5K, I, L; see below). Some (usually long) slight changes to the angle of the grooves among traces Knethichnus traces show sharp changes in orientation of (Fig. 5P). Otherwise, relatively long but thin Knethichnus the grooves (although direction of the grooves changes a traces (i.e. a few denticles contacted the bone) are also few degrees), occasionally with a Linichnus and/or Nihi- relatively common (Figs 2G–I, 3B, E–F, 5A–B, Q–S). lichnus trace in between (Fig. 5D, S), resembling the These traces are mostly rectilinear, sometimes with slight, pivoting traces produced by crocodylians (Njau & Gilbert though abrupt, changes of orientation. At the change of 2016). orientation point, Linichnus and/or Nihilichnus may Both Brutalichnus morphotypes are very commonly in appear (Fig. 5D, S). association with Nihilichnus, but in some cases they are On the edge of shafts (but also on some bone ends), also in transition from Knethichnus and Linichnus deep traces are frequently observed. They are removed (Figs 2I, J, 3B, D, 4G, 5T). Brutalichnus are mostly found flakes of (mostly cortical) bone with Knethichnus extend- on the end of long bones and bone edges (Figs 3–5; ing obliquely within the entire pit and Linichnus often in Data S1); when found on shafts, it is usually because the one of the short margins of the removed flake (Figs 2D, bone was broken from that part, most likely due to the F, 4G, B, C, E); such traces are likely to be equivalent to bite that generated the Brutalichnus trace (e.g. Fig. 5G). the ‘periosteal/subcambial bone spalling’ of Njau & Gil- Several bones also display areas with dense clusters of bert (2016) and the ‘flakes’ of Drumheller & Brochu diverse bite traces, in parallel or in multiple directions, (2014). In fact, the outline of these flakes with Knethich- and mostly including Nihilichnus and rectilinear to curved nus (i.e. a trace resulting from tooth dragging) is very Linichnus and Knethichnus (Figs 2A, K, 3D, E, 5I). Such similar to the crocodylian traces resulting from bone spal- dense clusters are extremely similar to those produced by ling (Drumheller & Brochu 2014; Njau & Gilbert 2016; crocodylians (Njau & Gilbert 2016; Shale et al. 2017), Shale et al. 2017). even though their teeth are not ziphodont, unlike those Several bones also display bite traces on opposing sur- of Batrachotomus, the bite maker of the traces analysed faces; these are usually Nihilichnus against Nihilichnus, herein (see section Identification of the bite maker, but also Nihilichnus against Brutalichnus, and Nihilichnus below). against Linichnus and/or Knethichnus (Figs 2A, 3–5). Several bones display clusters of (usually abundant) Notably, aligned sets of Nihilichnus also occur, outlining parallel to nearly parallel traces of Linichnus, with some an arch (Figs 3D, 4A, H) representing a partial tooth associated Knethichnus as well, when they build up the row. These traces are reminiscent of the ‘serial bite aforementioned N- or Z-shaped composite traces marks’ (in the sense of Binford 1981) generated during a (Figs 2B, 3A, E, 5C, D, P). Also, some bones, especially single biting event with multiple teeth impacting on the 14 PALAEONTOLOGY TABLE 2. Results of the PERMANOVA and LDA of bite traces. (Figs 7, 8). Thus, all bitten taxa were similarly processed Taxon Bone type Bone by the bite maker. region The distribution of bite traces along each type of bone (limb, rib, pectoral girdle, sacrum, pelvis, vertebral col- PERMANOVA Total sum 373.800 371.500 413.900 umn, skull and teeth) and bone region (for elongated of squares bones, proximal and distal end, and side and edge of the Within-group 367.400 332.800 345.600 shaft; for all other bones, bone edge and bone centre) sum of squares shows remarkable differences (Figs 7, 8; Table 2; F-value 1.052 2.894 1.658 Tables S4–S9). The results of the PCOA and LDA (Error: p-value 0.373 0.000 0.000 LDA Error 0.668 0.560 0.438 56%; Error-JK: 65.4%) indicate that bite trace patterns on ErrorResample 0.750 0.654 0.542 bones from different body regions strongly overlap with ErrorRandom 0.750 0.875 0.833 each other, except for the limb bones. The PERMANOVA results (F = 2.894; p < 0.001) indicate no significant over- The spatial overlap of biting spaces between taxa, types of bones lap of limbs and vertebrae with other bone regions. Nev- and bone regions is indicated (see also Tables S1–S9). Bold ertheless, the results show that bones from most body p-values of the PERMANOVA indicate significant differences regions were processed by the bite maker in a similar between biting spaces. LDA results are shown as errors in com- way, as all ichnotaxona are observed on each bone type, parison to random sampling. skeletal region, and bone region of both Mastodonsaurus and Batrachotomus (Fig. 8). Each bone region shows a strong overlap in the biting space along PCO1 (Fig. 7). bone. Occasionally, such aligned traces are parallel On the other hand, proximal ends, distal ends, side of grooves (i.e. Linichnus; Fig. 4J–K) indicating that the shafts and edge of shafts of long bones are clearly sepa- tooth row moved along the bone, and thus with teeth rated from each other along PCO2. The PERMANOVA analy- moving parallel to one another in a single bite. sis also indicates a common biting pattern in different Machichnus-like traces (Figs 2K, 6D–F) are the only bone regions, which is also evidenced by the presence of ones not associated with any other biting ichnotaxon. equivalent composite bite traces in different bones and Instead, they are only associated with (abundant) traces bone regions (see Associations of bite traces from the of the same morphotype. They are found as clusters of Lower Keuper, above; Figs 3–5; Data S1). Only the bone bite traces, being somewhat aligned with the long axis of edge and bone centre were found to be significantly dif- each trace perpendicular to the major axis of the whole ferent from the other regions (Fig. 7), although this may cluster. Therefore, they represent single bites or multiple, result from the different shape of the corresponding repeated strokes, or most likely both. bones (those that are non-elongated, such as vertebrae and Mastodonsaurus ischia and scapulacoracoids). How- ever, in the LDA (Error: 43.75%; Error-JK: 54.17%) bone Distribution of bite traces along the Lower Keuper bones ends and shafts are well separated from each other along axis 1. Furthermore, proximal and distal ends as well as As aforementioned, bite traces have been identified in sides and edges of shafts are well separated along axis 2. five different tetrapod taxa. With the exception of a rib Both the bone edge and centre regions cluster close to the probably corresponding to a chroniosuchid reptile point of origin (Fig. 7). (Fig. 2H), all bones that were anatomically identified All in all, the results indicate that different bone were included in the statistical analyses. Statistically, regions seem to be more specifically processed by the Mastodonsaurus bones show the largest anatomical distri- bite maker when compared with the skeletal regions, but bution of bite trace types in PCOA ‘biting space’, fol- still show common patterns of bite traces. Such differen- lowed by Batrachotomus and Plagiosuchus. All taxon- tial distribution of bite traces is also observed in the specific biting spaces show strong overlap (see LDA and occurrence of each ichnotaxon on each bone type, skele- PERMANOVA results: Fig. 7; Table 2; Tables S1–S3). The tal region and bone region of Mastodonsaurus and Batra- same associations or composite bite traces are found in chotomus, which encompass the bulk of bones with bite different bone types and bone regions (Figs 3–5), sug- traces, most of which are ribs (Fig. 8; Data S1). Note gesting that the bones were bitten in similar ways inde- that these counts are based on presence/absence, not the pendently from their shape and body region (Figs 7, 8). absolute number of each ichnotaxon (i.e. a given ichno- With the exception of the limbs and vertebrae, taxa taxon on a bone region is counted as one regardless of show a similar distribution of bite traces across anatomi- whether there is a single or several traces of this same cal regions, and the location and identity of the bite ichnotaxon). In Mastodonsaurus, Nihilichnus (punctures traces on individual bones depend on the skeletal region produced by tooth tips) are markedly more abundant MUJAL ET AL.: FEEDING ECOLOGY OF ARCHOSAURS 15 F I G . 8 . Counts of bite traces on each bone type and bone region for Mastodonsaurus and Batrachotomus. Counts are based on single presence; multiple traces of the same morphotype in a bone region are only counted as one, indicating the presence of the given bite trace morphotype. Bones of the restored skeletons are coloured according to the bars of bite traces counts, and as in the bone biting space (skeletal region) of Figure 7. Unknown bones are excluded. 16 PALAEONTOLOGY based on count data than the other ichnotaxa on both et al. 2018), with a density of 2.8–5.4 denticles/mm on proximal and distal bone ends, whereas in Batrachotomus the mesial and 3–4.8 denticles/mm on the distal carina this only occurs on the distal end (Fig. 8). Long bone (Data S2). In fact, in the Lower Keuper tetrapod assem- ends frequently show traces (mostly Nihilichnus, though blage there are no other known taxa with the same denti- Brutalichnus presence is also notable) on both sides, cle size and density (Schoch et al. 2018). Considering the resulting from a single bite using teeth from the lower association of Knethichnus with the other ichnotaxa (see and upper jaws (Figs 2A, 3A, C–E; Data S1). The side Distribution of bite traces along the Lower Keuper bones, (the widest) and edge (the narrowest) areas of the shaft above), the attribution of most of the identified bite show similar distributions on long bones of both Masto- traces to Batrachotomus is therefore unambiguous donsaurus and Batrachotomus (Fig. 8). On shafts, Linich- (Data S1). In addition, the deep Nihilichnus traces found nus dominates, followed by Knethichnus (which is often on some bones (e.g. Fig. 2A) display a rhomboid outline associated with the former); notably, the N- or Z-shaped fitting with the teeth of Batrachotomus. Similarly, Nihi- traces, V-shaped traces, and the deep flakes covered with lichnus traces corresponding to bisected pits (e.g. Knethichnus are often on the edge of the shaft (Figs 2C– Fig. 5M) fit with Batrachotomus teeth, as occurs in similar D, F, 3B, E, 4A–C, E, 5C–E, P). Also, in several shafts, bite traces attributed to phytosaurs (Drymala et al. 2021). Linichnus traces are perpendicular (with respect to the Furthermore, the distance between aligned traces outlin- bone axis) on the edge area and continue in oblique ori- ing (partial) tooth rows (e.g. Figs 3D, 4A, H, J–K) fits entation to the side area. Nihilichnus is also relatively with the distance between the teeth of Batrachotomus common on shafts, especially on ribs of Mastodonsaurus jaws. Particularly, the four Nihilichnus traces on a Masto- (Fig. 8); in fact, the uncinate processes of Mastodon- donsaurus ischium (SMNS 84244; Fig. 4H) draw a semi- saurus ribs often show Nihilichnus traces, which occa- circular line that fits with the outline of the premaxillae sionally are in transition to Linichnus and/or Knethichnus of a medium-sized Batrachotomus (e.g. Gower 1999); each (Fig. 5A, G, H; Data S1). Both morphotypes of Brutal- bite trace corresponds to a tooth present in the premaxil- ichnus are more commonly found on the ends of long lae, which would be four (two in each side) according to bones and bone edges, which is consistent with the the alternating replacement of teeth (i.e. the first and destructive nature of the trace. Nihilichnus is the domi- third teeth in one premaxilla, and the second and the nant ichnotaxon on non-elongated bones (flat or polygo- fourth in the other) as documented by Gower (1999) and nal elements and vertebrae). observed in the paired premaxillae of SMNS 80260 (Data S2). The only bite traces most likely not to have been pro- Identification of the bite maker duced by Batrachotomus are: (1) small Nihilichnus traces on a Mastodonsaurus tusk (SMNS 91634) that are aligned Since most of the observed ichnotaxa are associated and and outline a curved arch with little spacing between build up regular composite traces with similar distribu- them (Data S1); (2) the traces found on the ventral sur- tion in different bones (Figs 2A–J, 3–5, 8; Data S1), it is face of two palates (SMNS 80704, 81310) and one mandi- evident that they were produced by a single taxon. Within ble (SMNS 92128) of Mastodonsaurus (Fig. 6A–C); and the diverse tooth assemblage from the Lower Keuper (3) all the Machichnus-like traces (SMNS 81210, 97014, (Schoch et al. 2018), only the ziphodont teeth with later- 97013/5) (Figs 2K, 6D–F), because they are not associated ally compressed, blade-like crowns of the pseudosuchian with any other morphotype. Traces on the palates and archosaur Batrachotomus kupferzellensis (Gower 1999; mandible of Mastodonsaurus were all likely to have been Schoch & Seegis 2016; Schoch et al. 2018) match most of produced by a conspecific, especially the paired large the surveyed bite traces. Much more restrictive for the Nihilichnus on the palate SMNS 80704 and the mandible identification of the bite maker is the (relatively high) SMNS 92128 (Fig. 6A–B), which fit well with the mor- occurrence of Knethichnus parallelum, a trace produced phology and position of the paired tusks of Mastodon- by the dragging of denticles on the bone surface, and saurus mandibles (Schoch & Seegis 2016). Considering therefore exclusively produced by ziphodont teeth (Jacob- the small spacing between clusters of Machichnus-like sen & Bromley 2009; D’Amore & Blumenschine 2009, traces (probably produced in single bites and/or repeated 2012; Pirrone et al. 2014). The number of grooves in a strokes on the same area), as well as the notable size and given distance and their spacing in each Knethichnus trace morphological differences with the other ichnotaxa, it can show the minimum size and spacing of the tracing tooth be asserted that the Machichnus-like traces were produced denticles (D’Amore & Blumenschine 2012). All the ana- by a different taxon that could not be identified. In any lysed Knethichnus specimens show a maximum density of case, the bite maker was most probably a relatively small- 3.5 to 5 grooves/mm, falling in the range of Batrachoto- sized diapsid reptile with serrated teeth (possibly an arch- mus teeth (Gower 1999; Schoch & Seegis 2016; Schoch osauriform; Schoch et al. 2018), given the slightly serrated MUJAL ET AL.: FEEDING ECOLOGY OF ARCHOSAURS 17 margins of some grooves in clusters of Machichnus-like wear and notches are generally proportionally inverse on traces (Fig. 6E). Within the Lower Keuper tooth assem- both premaxillary and non-premaxillary teeth: wear blage, several tooth morphotypes (mostly of archosauri- increases from the basal to the apical regions of both cari- forms) that could fit with such bite traces have been nae, while notches show an opposite trend (with the identified, though some have not been associated with exceptions mentioned above; Fig. 11). any particular taxon yet (Schoch et al. 2018). However, Breakage and spalling most commonly occur on the although unlikely considering the inferred behaviour for tooth tips. Usually, the surfaces of broken tips are the formation of bite traces, the possibility that these bite smoothed or polished due to abrasive wear (Fig. 9A–C, traces were produced by a juvenile specimen of Batracho- G–I), indicating that breakage occurred during the tooth tomus cannot be totally excluded. lifetime and thus was most likely to have happened dur- ing feeding. Interestingly, premaxillary teeth often show higher degrees of spalling (running from the tip) on the Morphology and macroscopic wear on Batrachotomus teeth labial than on the lingual side (Data S2), thus indicating that such damage did not result from tooth–tooth con- Given the frequency of tooth–bone contact suggested for tact. Some teeth display severe breakage, with the entire Batrachotomus, we assessed the type and distribution of carina completely broken, starting from the tip. Notably, abrasive macroscopic wear on 82 premaxillary and 232 in some cases, only the tip and the apical distal carina maxillary and dentary teeth of this pseudosuchian archo- display breakage, with a sudden stop of the damage on saur. In fact, the surveyed tooth macroscopic wear the carina (Fig. 9E–F). (including wear, notches, spalling and breakage; Table 1; Fig. 9; Data S2) shows patterns that closely fit with the morphology, associations and distribution of bite traces, Microanatomy of Batrachotomus teeth as further shown in the Discussion, below. Teeth from Kupferzell (N = 258) and Vellberg-Eschenau (N = 56) Longitudinal thin sections along the carina reveal that show similar macrowear patterns, suggesting no ecological denticles in Batrachotomus consist of both enamel and differences between the two localities despite (slight) sedi- dentine, and thus are ziphodont following the definition mentological differences (Urlichs 1982; Schoch & Seegis of Brink et al. (2015) (see also Wang et al. 2015; Whit- 2016; DS, EM, RRS, pers. obs.) (Fig. 1B). ney et al. 2020). The tooth crown consists of enamel Box-plots and counts of macrowear show slight but and bulk dentine, separated by a zone of globular man- statistically significant differentiation of the macroscopic tle dentine (Fig. 9N–T). Enamel is thickest on the sur- wear from premaxillary and non-premaxillary teeth face of the denticles (c. 80 µm), but also towards the (Figs 10, 11), which indicate different usages of each mesial and distal carinae (up to 60 µm) (Fig. 9N, Q–S). tooth type (see Usage of teeth: correlating tooth wear and Enamel thickness also decreases on both the carinae and bite traces, below). Premaxillary teeth display relatively the sides of the tooth in an apicobasal direction. The more wear on the tips than on the carinae, also account- enamel of each denticle is separated from that of the ing for a higher degree of abrasion than the non- adjacent denticle by a channel continuous with the premaxillary teeth. Wear generally decreases from the tip interdental fold (Fig. 9N–O) (Brink et al. 2015). Within to the base of both premaxillary and non-premaxillary a denticle, enamel thickness thins toward the interdental teeth (e.g. Fig. 9C, D). The tips of premaxillary teeth pos- fold (Fig. 9N, Q, R). Enamel spindles are rare overall, sess more frequently severe (extreme) wear than those of but occasionally present in the denticles (Fig. 9P). Den- maxillae and dentaries. Wear on non-premaxillary teeth is tine tubules originating within the fold curve pulpward more common and severe on the mesial carina than on towards the base of the denticle, becoming more densely the distal carina, whereas premaxillary teeth show rela- distributed between folds (as with plicidentine). The base tively homogeneous wear over both carinae, though the of the interdental fold consists of globular dentine, but basal mesial carina (where denticles are often lacking) is globular mantle dentine is thin to absent in the denticles significantly less worn than the other regions. Notches themselves (Fig. 9N–P, T). Sclerotic dentine was difficult show much more differentiation in distribution. On pre- to recognize: there are fewer dentine tubules surround- maxillary teeth, notches are more abundant and severe on ing the interdental fold, and this region appears to min- the middle and basal distal carina, and a very few occur eralize more heavily between the least worn and a on the basal mesial carina. On non-premaxillary teeth, heavily worn tooth. However, wear on individual denti- notches occur more frequently on the mesial carina; cles (Fig. 9Q, R) does not appear to influence the devel- nonetheless, the middle distal carina accounts for the opment of sclerotic dentine, nor does sclerotic dentine highest occurrence of notches, with a marked difference develop on the tips of denticles worn down to the bulk relative to the other regions. Interestingly, the counts of dentine. 18 PALAEONTOLOGY A B C J K L M D E F N O G H I P Q R S T FIG. 9. Macroscopic wear and microanatomy of Batrachotomus teeth. A–C, left premaxillary tooth SMNS 97009/4. D–M, non- premaxillary teeth: SMNS 97009/5 (D–F), 97009/3 (G–I), and 97009/2 (J–M). Images show medial (A, D, G, J), labial (B, K), lingual (E, H, L), and distal (C, F, I, M) views. N–P, longitudinal thin section of denticles from the apical distal carina of the non- premaxillary tooth SMNS 97009/16 in transmitted light. Q, worn denticles from the middle mesial carina of SMNS 97009/8 in longitu- dinal section. R–S, transverse section of SMNS 97009/17, showing thicker enamel close to the carina (also showing a worn denticle) (R) compared to the lateral portion (S). T, detail of the dentine–enamel junction of SMNS 97009/16. Dashed lines in Q and R indicate the supposed original (unworn) denticle shape. Abbreviations: B, breakage; d, dentine; dt, end point of denticles; e, enamel; es, enamel spindles; gd, globular dentine; if, interdental fold; N, notch; S, spalling; W, wear. Scale bars represent: 10 mm (A–C; D–F; G–I; J–M); 100 µm (N, Q); 40 µm (O, T); 25 µm (P); 200 µm (R–S). Comparison of Batrachotomus teeth with those of theropod spinosaurid Megalosauroidea and Spinosauridae, and are dinosaurs and other archosauromorphs further surrounded by the tooth morphospaces of Allosaur- oidea, Ceratosauria, Tyrannosauroidea and Archosauro- According to our LDA analysis of dental measurements in morpha morphotype A (Fig. 12). All these taxa represent different archosauromorphs, including non-avialan thero- apex predators with ziphodont teeth. If Spinosauridae are pods (see Hendrickx et al. 2015; Hoffman et al. 2019), the treated as an independent group, the LDA (Error: 43.05%; teeth of Batrachotomus plot in between the area of non- ErrorResample: 46.03%) identifies 19 out of 23 Batrachotomus MUJAL ET AL.: FEEDING ECOLOGY OF ARCHOSAURS 19 FIG. 10. Box-plots of wear inten- sity on Batrachotomus teeth (1 = low; 2 = high; 3 = extreme). Above, intensity of the different types of macrowear. Below, intensity of macroscopic wear (including all four types) for each portion of the carinae, tip, and base of the crown. teeth correctly (Table 3). One tooth is identified as Coelo- teeth are found to belong to Megalosauroidea (Table 3). physoidea and three as non-spinosaurid Megalosauroidea. Treating Batrachotomus teeth as unknown, they are Treating Batrachotomus teeth as unknown, they are most identified 16 times as Megalosauroidea (Table 3). Based commonly identified as Allosauroidea (five times), Archo- on this grouping, the PERMANOVA found additional over- sauromorpha morphotype A (six times) and Spinosauridae lap with the morphospaces of Megalosauroidea and (five times). The PERMANOVA indicates an overlap of the Silesauridae. Batrachotomus tooth morphospace with that of Archosaur- omorpha morphotype A, Ceratosauria, Erpetosuchidae, ‘Rauisuchia’ and Suchia. DISCUSSION When Spinosauridae are grouped together with the other Megalosauroidea, LDA (Error: 42.38%) classifies Usage of Batrachotomus teeth: correlating tooth wear and 14 teeth correctly, while the other teeth are identified bite traces either as Allosauroidea, Ceratosauria, Dromaeosauridae, Megalosauroidea or Tyrannosauroidea. After cross- Associated and composite bite traces, combined with a validation of the LDA (ErrorResample: 45.14%), only 12 differential distribution of each bite trace morphotype Batrachotomus teeth are correctly identified, while nine in each skeletal and bone region (Fig. 8; Data S1) and 20 PALAEONTOLOGY FIG. 11. Percentage of abrasion (wear and notches) on Batrachotomus teeth. The degree of abrasion (none, low, high, extreme) is accounted. Numbers within colour bars represent counts (absolute occurrences) of each degree of abrasion in each tooth portion. Unknown values (‘?’ in Data S2) are excluded, thus the sum of values of each bar is lower than the total number of teeth (N) analysed. the macroscopic wear on Batrachotomus teeth, allow the on opposing bone surfaces, outlining the tooth rows movement of teeth on bones, and thus the most proba- (Figs 3D–E, 4H). These features indicate a potential dis- ble behaviours and usages of the carcasses, to be memberment of the carcasses in order to reach fleshy and reconstructed. nutrient-dense regions, as done by extinct and extant cro- As shown in ‘Morphology and macroscopic wear on codylians (Njau & Blumenschine 2006; Njau & Gilbert Batrachotomus teeth’ above, tooth tips, especially from the 2016). Dismemberment and defleshing strokes would premaxilla, showed the most wear (Figs 10, 11), correlat- have been violent in some cases, as the Brutalichnus traces ing with the dominance of Nihilichnus traces, especially and the relatively high amount of breakage of tooth tips on the ends of ribs and limb bones (Fig. 8), and also on and carinae suggest. Brutalichnus is often associated with the uncinate processes of Mastodonsaurus ribs (Fig. 5A, Nihilichnus, but also with Knethichnus and Linichnus G, H). Such puncture traces are often aligned and found (Figs 2–5; Data S1), indicating that there were pullbacks MUJAL ET AL.: FEEDING ECOLOGY OF ARCHOSAURS 21 FIG. 12. LDA results of archosauromorph teeth. Plot comparing teeth measurements of Batrachotomus with that of different archo- sauromorphs, including various non-avialan theropod dinosaur groups. Spinosaurids are separated from Megalosauroidea; the dashed line shows the whole area occupied by megalosauroids, mostly overlapping that of Batrachotomus. Percentage of total variation for each ordinate is given in parentheses. The factor loadings of the LDA are visualized as a biplot. 22 PALAEONTOLOGY TABLE 3. Results of the PERMANOVA and LDA of tooth measurements. Group All Megalosauroidea Spinosauridae separated PERMANOVA LDA PERMANOVA LDA F-value p-value Classification Unknown F-value p-value Classification Unknown Batrachotomus NA NA 14 (12) NA NA NA 19 (19) NA Allosauroidea 42.430 0.012 3 (0) 2 42.430 0.014 0 (0) 5 Archosauromorpha morph. A 7.425 0.096 0 (0) 2 7.425 0.095 0 (0) 6 Archosauromorpha morph. B 22.410 0.024 0 (0) 0 22.410 0.027 0 (0) 0 Basal Saurischia 71.430 0.012 0 (0) 0 71.430 0.014 0 (0) 0 Ceratosauria 6.244 0.588 2 (0) 0 6.244 0.843 0 (0) 0 Coelophysoidea 82.920 0.012 0 (0) 0 82.920 0.014 1 (1) 2 Dromaeosauridae 123.100 0.012 1 (0) 0 123.100 0.014 0 (0) 0 Erpetosuchidae 2.271 1.000 0 (0) 0 2.271 1.000 0 (0) 0 Megalosauroidea 3.626 1.000 2 (9) 16 15.340 0.014 3 (3) 3 Megaraptora 16.090 0.024 0 (1) 1 16.090 0.014 0 (0) 0 Rauisuchia 11.910 0.072 0 (0) 0 11.910 0.054 0 (0) 0 Silesauridae 171.400 0.072 0 (0) 0 171.400 0.041 0 (0) 0 Spinosauridae NA NA NA NA 28.590 0.014 0 (0) 5 Suchia 2.495 1.000 0 (1) 2 2.495 1.000 0 (0) 2 Troodontidae 128.300 0.012 0 (0) 0 128.300 0.014 0 (0) 0 Tyrannosauroidea 35.390 0.012 1 (0) 0 35.390 0.014 0 (0) 0 The spatial overlap between Batrachotomus and other archosauromorph teeth (including various non-avialan theropod dinosaurs) is indicated. On the left, Spinosauridae are included within Megalosauroidea; on the right they are separated. In each case, Batrachotomus teeth were treated as ‘classified’ and ‘unknown’ for the LDA. Bold p-values of the PERMANOVA analysis indicate a significant degree of overlap between Batrachotomus and other archosauromorphs. Numbers in parentheses in the LDA sections show number of classifica- tions after resampling. NA, not applicable. and/or repeated strokes associated with these more of tooth crowns. In contrast, the common (short) bite destructive bites. traces on the edge of the shafts were likely to have been The amount and distribution of wear and notches produced by the distal carinae. A pullback when tearing along the tooth carinae are consistent with the genera- and dismembering carcasses with the distal carina tion of most Linichnus, Knethichnus, and their associa- wedged into the bone surface (D’Amore & Blumenschine tions (N/Z-, fan- and V-shaped composite traces, and 2009; Njau & Gilbert 2016) would have generated both clusters of Linichnus; Figs 2–5). Tooth crowns contacted Knethichnus within relatively deep flakes or hacks bones in repeated strokes to deflesh and tear carcasses, (Figs 2F, 3B, 4A–C; see also Jacobsen & Bromley 2009, leaving composite traces similar to those of crocodylians fig. 3B) and Linichnus, with tooth crowns perpendicular (Njau & Blumenschine 2006; Njau & Gilbert 2016), and oblique to the bone long axes, and resulted in the occasionally misinterpreted as bone modifications by observed markedly higher number of notches on the dis- hominids (Shale et al. 2017). Nonetheless, each individ- tal carinae of premaxillary teeth and the middle distal ual bite trace is more similar to those of non-avialan carinae of non-premaxillary teeth (Fig. 11). The usual theropods and V. komodoensis (D’Amore & Blu- lack of denticles on the basal portion of the mesial menschine 2009; Jacobsen & Bromley 2009), in other carina of premaxillary teeth (Fig. 9B; Data S2) probably words, taxa with ziphodont teeth. Higher amounts of responds to a low functionality of this region (D’Amore abrasion in the apical and middle mesial carina, espe- 2009), which is much less worn than the rest of the cially of non-premaxillary teeth, are consistent with this crown (Figs 10, 11). Thus, the observed pattern of mac- interpretation of repeated strokes (Fig. 11). The mesial roscopic wear in the teeth of Batrachotomus provides carina would have more easily reached the bone and complementary support for its identity as the bite also had a wider range of motion than the distal carina maker. (D’Amore & Blumenschine 2009). Similarly, the parallel In some fossil taxa, tooth–tooth contact (attritional and oblique orientations of Knethichnus with respect to wear) has been invoked to explain patterns of carinal the bone long axes (Data S1) indicate lateral movement wear and basal facets (Schubert & Ungar 2005; Young MUJAL ET AL.: FEEDING ECOLOGY OF ARCHOSAURS 23 et al. 2012). We consider such explanations unlikely in of macrowear, variation in enamel thickness across the Batrachotomous. The bite trace assemblage clearly sup- tooth crown in Batrachotomus is correlated with varia- ports extensive carina–bone contact as described above. tion in wear, with enamel thickness greatest in those Spalled teeth usually display damage on both labial and areas of the tooth most prone to contact with bone dur- lingual sides (Fig. 9K, L); if such damage was generated ing feeding (Fig. 9Q–S). The overall tooth microstructure due to occlusion, these teeth would have contacted the of Batrachotomus has been interpreted as a convergent opposing teeth on both their lingual and labial sides, character of hypercarnivores engaging in puncture and which is very unlikely or even biomechanically impossi- pull feeding (D’Amore 2009; Whitney et al. 2020), ble. In fact, when the mouth of Batrachotomus was closed increasing wear resistance of the denticles and while (see reconstruction by Gower & Schoch 2009), teeth from maintaining overall tooth elasticity (Brink et al. 2015; the upper jaw remain externally placed, thus only their Wang et al. 2015; Gignac & Erickson 2017). This func- lingual side could have contacted the labial side of teeth tional interpretation is congruent with the wear and from the lower. Therefore not all spalling (if any) could breakage patterns of Batrachotomus teeth described here result from tooth–tooth contact. Similarly, premaxillary (see Morphology and macroscopic wear on Batrachoto- teeth are usually more worn towards the labial side mus teeth, above; Fig. 9; Data S2) indicating specializa- (Data S2), the side that would have never occluded with tions for puncture and pull feeding by Batrachotomus on teeth from the mandible. various large sympatric vertebrates. The role of microanatomy on the usage of Batrachotomus Palaeoecological implications of the feeding habits and teeth taphonomy of Batrachotomus The microanatomy of Batrachotomus teeth (Fig. 9N–T) Feeding ecology and taphonomy. The distribution and fre- shows adaptations conferring resistance to carinal wear quency of bite traces is suggestive of different behaviours and breakage caused by tooth–bone contact seen in by Batrachotomus, including scavenging, cannibalism and other hypercarnivores engaging in puncture and pull predation. We found a strong correlation between the fre- feeding (non-avialan theropod dinosaurs and gorgonop- quency of bone types in a Mastodonsaurus skeleton and sian synapsids: D’Amore 2009; Brink et al. 2015; Wang the frequency with which each bone is bitten (Fig. 13), et al. 2015; Whitney et al. 2020). The loss of globular suggesting a systematic and virtually complete exploita- mantle dentine at the tips of the denticles, combined tion of carcasses by Batrachotomus. with its presence at the enamel–dentine junction else- Different bones of Mastodonsaurus and Batrachotomus where along the tooth crown, has been interpreted as an from Kupferzell were found in clusters reliably referred to adaptation for wear resistance in the denticles; in addi- single individuals (Wild 1980; Gower 1999; Schoch 1999; tion, elasticity elsewhere in the tooth has been inferred Gower & Schoch 2009; Data S1). These associations, to confer resistance to breakage (Brink et al. 2015). The together with the lack of bone sorting, indicate that there deep interdental folds would have helped to dissipate was little-to-no transport of carcasses, at least in Kupfer- stress forces on the teeth (Brink et al. 2015; Gignac & zell. In fact, in Vellberg-Eschenau, where bite traces are Erickson 2017); globular mantle dentine as found at the notably lower in number than in Kupferzell (Data S1), bases of the folds has been linked to blocking crack evidence of bone sorting has been reported (Schoch & propagation and increasing tissue elasticity (Wang et al. Seegis 2016). This would indicate a higher washout of 2015). The sudden stop of breakage on the distal carina carcasses in Vellberg-Eschenau, further noting slight sedi- observed in Batrachotomus (Fig. 9E–F; Data S2) is con- mentological differences (Urlichs 1982; DS, EM, RRS, sistent with this functional interpretation of deep inter- pers. obs.; Fig. 1B). Nonetheless, between Kupferzell and dental folds. Enamel thickness in Batrachotomus is quite Vellberg-Eschenau there are no differences in the location thin relative to that in both therapsids and Tyrannosau- of bite traces on bones nor in the macroscopic wear of rus (Whitney et al. 2020), but is within the reported Batrachotomus teeth (Data S1, S2). These similarities sug- range of ‘rauisuchians’ and other theropods (Sander gest that, as a whole, this bite trace assemblage is not an 1999; Wang et al. 2015). Sander (1999) reported that unusual scenario under specific environmental conditions, ‘rauisuchian’ enamel was thinnest on the tops of the but the record of the usual feeding behaviour of the bite denticles; this is in contrast to what we observed in maker. Hypothetically, this is also indicated by the pre- Batrachotomus (Fig. 9R, S), but might be attributable to dictable distribution of composite bite traces in carinal wear in the tooth sampled by Sander (1999) (see different taxa, different bones, and different bone regions worn denticles in longitudinal and transverse section in (Figs 3–5, 7, 8) suggesting that, during feeding, Batracho- Fig. 9Q, R). Here, considering the distribution and types tomus modified bones in a regular, monotonous manner. 24 PALAEONTOLOGY F I G . 1 3 . Frequency of bitten bones of Mastodonsaurus and Batrachotomus. Numbers of bones (N) account for the bones that are known for each taxon and skeletal region (e.g. forelimb includes six bones: two humeri, two radii and two ulnae); teeth, phalanges, tarsals, carpals and vertebrae are excluded because their total amount per skeleton is unknown. Sacral ribs (differentiated in the statisti- cal analyses) are here counted together with thoracic and caudal ribs. Restored skeletons are coloured according to the skeletal regions shown in the table and in Figures 7 and 8. Note the high percentage of bitten bones from the pelvic girdle of Batrachotomus, being nearly three times the percentage of these bones per skeleton (highlighted in bold). The distribution and frequency of bite traces produced associated with defleshing rather than disarticulation, as by Batrachotomus across bone regions (Fig. 8) is conspic- has been suggested for crocodylians (Njau & Blu- uously similar to that of crocodylians (e.g. Njau & Blu- menschine 2006). All in all, the differential distribution of menschine 2006; Drumheller & Brochu 2014, 2016; Njau bite traces supports the hypothesis of different usage for & Gilbert 2016; Shale et al. 2017). As documented in each bone region, in agreement with the statistical results extant crocodylians by Drumheller & Brochu (2014), the (Fig. 7) and further pointing to conserved feeding behav- bones bitten by Batrachotomus display more penetrative iour in pseudosuchians (Fig. 14).The incompleteness of bite traces on the ends of long bones (deep Nihilichnus skeletons would suggest habitats roamed by scavengers and Brutalichnus). Njau & Blumenschine (2006) associ- (Hungerb€ uhler 1998), tearing carcasses and thus facilitat- ated the presence of relatively deep traces to disarticula- ing the loss of specific parts that otherwise would have tion attempts, which was also likely to have been the case been preserved. Similarly, the high abundance of shed for the bones bitten by Batrachotomus (see Usage of teeth within mudstones also indicates a low energy envi- Batrachotomus teeth: correlating tooth wear and bite ronment inhabited by several individuals (Hungerb€ uhler traces, above). Otherwise, also like in crocodylians 1998; Augustin et al. 2020). Further evidence of scaveng- (Drumheller & Brochu 2014), the shafts of long bones ing is provided by the presence of bite traces produced by from the Lower Keuper more commonly (though not much smaller organisms than Batrachotomus. The exclusively) display relatively shallow grooves (Linichnus Machichnus-like traces suggest abandoned corpses con- and Knethichnus) that usually do not remove the entire sumed by small-sized scavengers (probably archosauri- bone cortex and are oriented oblique to perpendicular forms; see teeth in Schoch et al. 2018) that could reach with respect to long bone axis (e.g. Fig. 3D, E; Data S1). carcasses when relatively high nutrient-value regions (such Similarly, spalled flakes of bone, often covered with as the rib cage) were still available (Drumheller et al. Knethichnus, are mainly found on the shaft of long bones 2020). Therefore, even though there may have been (e.g. Figs 2D, F, 3A–C), also reminiscent of those pro- periods of stressed environments (e.g. droughts) as indi- duced by crocodylians (Drumheller & Brochu 2014; Njau cated by histological data from Batrachotomus bones & Gilbert 2016). Such bite traces are likely to be (Klein et al. 2017), scavenging was not an unusual habit. MUJAL ET AL.: FEEDING ECOLOGY OF ARCHOSAURS 25 FIG. 14. Simplified phylogeny of amniotes focused on taxa with ziphodont teeth. Age distribution of each clade follows literature cited throughout the manuscript. Potential high-frequency of bone tracing by tyrannosaurids (*) after Hone & Rauhut (2010) (but see discussion in Drumheller et al. 2020). Silhouettes from: PhyloPic (http://phylopic.org: Steven Traver (crocodile; CC0 1.0); Scott Harman (non-avialan theropods; CC BY-NC-SA 3.0); Wikimedia Commons, Dmitry Bogdanov (gorgonopsian; CC BY 3.0); Batrachotomus, phytosaur and Komodo monitor are author originals. Remarkably, Batrachotomus also fed on conspecifics gregariousness in Triassic pseudosuchians (Francßa et al. (Figs 2C–E, G, 3F, 4I–K, 5E, N, 8; Data S1), representing 2011; Nesbitt et al. 2013, 2020). one of the few cases of cannibalism in the archosaur fos- Batrachotomus was less frequently bitten than Masto- sil record, previously only reported in non-avialan thero- donsaurus (Fig. 8; Data S1), being a more occasional food pods (Rogers et al. 2003; Longrich et al. 2010; source, and possibly representing evidence of scavenging Drumheller et al. 2020), but well-known in present-day behaviour. Thus, under unstable environmental condi- crocodylians (Cott 1961). Most of the bite traces on tions, Batrachotomus possibly adopted cannibalism. None- Batrachotomus bones are concentrated on ribs and pelvic theless, the relatively high number of Batrachotomus elements (Fig. 13), which give access to the nutrient-rich bones with bite traces suggests that cannibalism was not a viscera and muscles of the tail and hindlimbs. Consider- totally unusual feeding habit. The frequency of bitten ing their location on the bones and the lack of evidence bones with respect to their number in a skeleton of of healing, these traces could not have been produced in Batrachotomus (Fig. 13) suggests that carcasses of this a living specimen; therefore, they indicate feeding rather taxon were not completely exploited and, as the than non-trophic intraspecific interactions such as social Machichnus-like traces on Batrachotomus ribs suggest, behaviour. This latter behaviour is known from both small-sized scavengers did have access to them (Drumhel- fossil pseudosuchians (including crocodylians) (Abel ler et al. 2020) (see above). 1922; Buffetaut 1983; Avilla et al. 2004; Katsura 2004) The large Nihilichnus found on opposing surfaces of and present-day crocodylians (K€alin 1936; Cott 1961; Mastodonsaurus femora (Fig. 3C–E), even preserving the Webb & Manolis 1983), and mostly results in wounds outline of the tooth (Fig. 2A), indicate active predation, on skull bones (either healed or not) and/or traumatic as they are strikingly similar to those produced by large pathologies on regions of the skeleton that can be Triassic predatory archosaurs (Niedzwiedzki et al. 2011; reached in a living individual. The Batrachotomus teeth Drumheller et al. 2014; Qvarnstr€ om et al. 2019), Creta- preserving Knethichnus and Linichnus traces on their ceous crocodyliforms (Noto et al. 2012; Boyd et al. 2013) labial surfaces (Data S1) provide conclusive evidence of and Miocene caimans (Pujos & Salas-Gismondi 2020), all intraspecific interaction, possibly competition, among resulting from biting the limbs to disable prey during living individuals and are consistent with aggressive active hunting. This suggests that Batrachotomus individ- interactions seen in fossil pseudosuchians and extant uals would have attacked Mastodonsaurus on their hin- crocodylians (Abel 1922; Cott 1961; Buffetaut 1983). dlimbs from the rear, as observed in crocodylians (Cott This also agrees with purported evidence for 1961; Pujos & Salas-Gismondi 2020), to avoid a 26 PALAEONTOLOGY potentially dangerous face-to-face encounter. Bones with Batrachotomus and Ticinosuchus highlight the potential more destructive bite traces, such as Brutalichnus and for a broad dietary spectrum in basal pseudosuchians. In deep Nihilichnus, show fractures with saw-toothed mar- this regard, some regurgitalites including skeletal remains gins (Figs 2I–J, 3C–D, 4G–I 5G, J), indicative of damage from Vellberg-Eschenau (Schoch & Seegis 2016) could to fresh bones (Britt et al. 2009). However, healed bite have been produced by Batrachotomus, as suggested for traces indicating attack survival (Drumheller et al. 2014) Triassic ‘rauisuchians’, representing another potential have not been found. Thus, if hunting did occur, all the behaviour shared with crocodylians (Gordon et al. 2020) evidence indicates a predominance of successful attacks. and further showing conserved feeding ecology in pseu- Further interaction between Mastodonsaurus and Batra- dosuchians (Fig. 14). Multidisciplinary studies including chotomus is recorded by two Mastodonsaurus tusks show- bite traces, tooth macrowear and microanatomy, as well ing Linichnus and Knethichnus traces (Data S1). Since no as comparisons with taxa with similar dentition, allow for bite traces on Mastodonsaurus skull bones were produced the better identification of feeding habits and ecological by Batrachotomus (Fig. 6A–C), it is unlikely that these roles of extinct taxa. In the same way, studies specifically bite traces on teeth represent feeding; rather they proba- focused on tooth macrowear and/or microanatomy may bly represent a face-to-face interaction. already provide information of the feeding ecology of the Considering that Mastodonsaurus most commonly analysed group (e.g. D’Amore 2009; Brink et al. 2015; roamed (sub-)aquatic habitats (Schoch & Milner 2000; Wang et al. 2015). This is in line with the functionality Mujal & Schoch 2020), Batrachotomus would potentially study of denticulated (ziphodont) teeth by D’Amore also have hunted in such environments. Nonetheless, evi- (2009). As pointed out in that work, the mesial carina of dence of desiccation and relatively long-term aerial expo- theropod teeth tend to be less denticulated, suggesting sure in the Kupferzell deposits (Wild 1980; Urlichs 1982; that the mesial carina was used less than the distal carina RRS, DS, EM, pers. obs.; Fig. 1B) could indicate that Mas- during feeding. In Batrachotomus, this lack of denticles is todonsaurus individuals were trapped in shallow pools only observed in the basal portion of the mesial carina of where they were hunted by Batrachotomus. The ability of some premaxillary teeth (Fig. 9B; Data S2) and, in con- Batrachotomus to forage in an aquatic habitat is consistent trast to what is observed and expected in theropods (cf. with the dominance of piscivorous tooth types in the Fiorillo 1991), Batrachotomus teeth show higher degrees Untere Graue Mergel deposits (Schoch et al. 2018) and the of abrasion (wear) on the mesial carina (Figs 10, 11), bite traces of Batrachotomus on the bones of the aquatic indicating that these parts were more used during feeding. sauropterygian reptile Nothosaurus (Fig. 5R). ‘Rauisu- Such wear can be linked to the relatively high frequency chians’ (e.g. Postosuchus) are usually interpreted as large, of Knethichnus (Fig. 8), especially also when this morpho- terrestrial hypercarnivores (Nesbitt et al. 2013); however, type outlines fan-shaped traces and when it is associated the ability of some genera to exploit aquatic food sources with Linichnus in clusters of traces (e.g. Figs 2B, H, 3D, has previously been noted (Nesbitt 2011). Furthermore, 4K–L). Such traces represent repetitive biting strokes and the strongly convex ventral margin of the maxilla and dor- show changes in the direction of movement of teeth con- soventral expansion of the anterior end of the dentary in tacting the bone (probably indicating defleshing attempts; Batrachotomus (Gower 1999) resembles the morphology of see also Feeding ecology and taphonomy, above). large-bodied theropod dinosaurs with ichthyophagous Knethichnus is a bite trace morphotype exclusively linked behaviour (Charig & Milner 1997; Marsh & Rowe 2020). to denticulated (ziphodont) teeth (Jacobsen & Bromley 2009) also produced by theropod dinosaurs and Palaeoecological implications for archosaurs. The afore- V. komodoensis. However, in these groups, this bite trace mentioned morphological similitudes between the den- morphotype is much less frequent than others, such as taries of Batrachotomus and ichthyophagous theropods Nihilichnus and Linichnus (Jacobsen & Bromley 2009; are in accordance with the similarities between the teeth D’Amore & Blumenschine 2009; Drumheller et al. 2020). of Batrachotomus and megalosauroids (see Comparison of This notable difference is more likely to be explained by Batrachotomus teeth with those of theropod dinosaurs different bone usages and feeding behaviours between and other archosauromorphs, above; Fig. 12; Table 3), a clades, showing that Batrachotomus and pseudosuchians, theropod clade that evolved piscivorous adaptations including extant crocodylians, process the bones more (Charig & Milner 1997; Sadleir et al. 2008). At first extensively than theropods and V. komodoensis (Fig. 14). glance, this is unexpected, given the evidence for hyper- Fiorillo (1991) noticed that isolated theropod teeth are carnivory in Batrachotomus documented herein. However, common in dinosaur fossil localities and that few of them as discussed above, such similitudes suggest a potentially show signs of wear, explaining the low frequency of bone broad diet not reflected in the bite-trace record. Ticinosu- tracing of the clade (and possibly the lack of denticulation chus, another Triassic pseudosuchian, is also known to of some teeth: D’Amore 2009). Indeed, low levels of tooth have included fish in its diet (Nesbitt 2011). Thus, wear are in contrast with our findings from the MUJAL ET AL.: FEEDING ECOLOGY OF ARCHOSAURS 27 Batrachotomus tooth assemblage (Figs 9–11; Data S2), than by theropod dinosaurs, probably resulting from the including only a few pristine tooth crowns probably cor- deep phylogenetic split between these two archosaur responding to unerupted teeth. Fiorillo (1991) concluded clades (Fig. 14). that theropod dinosaurs did not usually crush bones (potentially with a few exceptions, see Hone & Rauhut 2010) and considered that this ecological role was unoc- CONCLUSION cupied during the Mesozoic. While agreeing with a low level of bone crushing activity by theropods, considering By analysing bite traces on bones produced by the pseu- the high frequency of bone tracing by pseudosuchians, we dosuchian Batrachotomus kupferzellensis together with suggest that this ecological role was occupied by the pseu- tooth morphology, microanatomy and macrowear, and dosuchians. Potential evidence for this hypothesis is comparing them with those of other pseudosuchians found in the Lower Cretaceous vertebrate lagerst€atte from (including extinct and extant crocodylians), theropod France analysed by Rozada et al. (2021). These authors dinosaurs, and Varanus komodoensis, we suggest that phy- reported numerous bite traces only associated with croco- logeny can be a better predictor of feeding behaviour than dylomorphs (see also G^ onet et al. 2019), even though tooth morphology (Fig. 14). Even if tooth–bone contact megalosauroid theropods are also present in the locality during feeding was mostly unintentional in Batrachoto- (mostly represented by teeth; Rozada et al. 2021). Nota- mus, as suggested for non-avialan theropods and bly, our analysis of archosauromorph teeth shows that V. komodoensis (Fiorillo 1991; D’Amore & Blumenschine those of Batrachotomus are mostly similar to those of 2009), the high frequency of bitten bones, the predictable megalosauroids (Fig. 12; Table 3). Therefore, although distribution of traces within the Lower Keuper assem- Batrachotomus and megalosauroids probably shared some blage, and the severe damage of many teeth indicate that feeding habits (see above), it is likely that they also had Batrachotomus did not actively avoid such contact. This is different ecological roles, again suggesting a broad diet comparable to the feeding behaviour of crocodylians for Batrachotomus. (Njau & Gilbert 2016; Shale et al. 2017). In this sense, All in all, the relatively lower frequency of bone- even if bite traces produced by Batrachotomus morpho- tracing by theropods with respect to pseudosuchians logically resemble those of taxa with ziphodont teeth (including present-day crocodylians) may relate to dif- (producing the same ichnotaxa; Jacobsen & Bromley ferent behaviours in the manipulation of prey or car- 2009), associations, relative abundances and frequencies casses; the former groups would mostly engage in of bite traces are more similar to those of the pseudosu- pullback strokes (possibly similar to the behaviour of chian lineage (Njau & Blumenschine 2006; Noto et al. V. komodoensis; e.g. D’Amore 2009; D’Amore & Blu- 2012; Boyd et al. 2013; Drumheller & Brochu 2014, 2016; menschine 2009, 2012; Snively et al. 2013) resulting in a Drumheller et al. 2014; Njau & Gilbert 2016; Shale et al. lower likelihood of tooth–bone contact than that of 2017; G^ onet et al. 2019; Pujos & Salas-Gismondi 2020; pseudosuchians (Fig. 14), which usually engage in vio- Drymala et al. 2021; Rozada et al. 2021), a likely synapo- lent and abrupt moves to tear prey and carcasses (e.g. morphy of this long-lived hypercarnivorous group. This Drumheller & Brochu 2014, 2016; Njau & Gilbert 2016). points to a dichotomy in the interpretation of bite traces: Recently, Drumheller et al. (2020) documented a bone on the one hand, trace morphologies correlate with mor- assemblage with unusually high frequencies of theropod phology of teeth and their movement along the bone; on bite traces. These authors linked this to long exposure the other hand, the association, location and frequency of times for the remains and late access to carcasses (scav- bite traces reflect the behaviour of the bite maker, proba- enging, including cannibalism) in a stressed ecosystem. bly differing between carnivorous clades (Fig. 14). Fur- Nonetheless, Drumheller et al. (2020) also highlighted thermore, our results show that tooth gross morphology potential collecting biases against remains with bone alone may not reflect the complete dietary spectrum of a surface modifications (see also Pobiner 2008), which given taxon; this can be overcome by combining analyses could be the case for tyrannosaurs, probably overrepre- of macroscopic wear and microanatomy, which provide sented in the literature (see also Hone & Rauhut 2010). independent lines of evidence of dietary preference and Even if collecting biases exist in dinosaur localities (as feeding behaviour. shown by Drumheller et al. 2020), in the case of the Our multidisciplinary analyses of an exceptional associ- Lower Keuper, many complete (or nearly complete) ation of bite traces, tooth wear patterns, and tooth micro- bones display bite traces (Figs 3–5; Data S1). This indi- anatomy from the Lower Keuper fossil lagerst€atten allow cates that biases against collection and preparation of the feeding ecology of a 240-million-year-old pseudosu- incomplete bones do not necessarily preclude the collec- chian reptile to be understood in detail. Despite morpho- tion of bitten elements, and supports a higher frequency logical affinities of Batrachotomus teeth with those of of bone tracing by Batrachotomus and pseudosuchians megalosauroid theropod dinosaurs, the behaviours 28 PALAEONTOLOGY underlying the produced bite traces share greater similari- REFERENCES ties to those produced by members of the crocodylo- morph lineage. This suggests evolutionary conservation of A B E L , O. 1922. Die Schnauzenverletzungen der Parasuchier und ihre biologische Bedeutung. Pal€aontologische Zeitschrift, 5, feeding behaviour in this clade, indicating that this suc- 26–57. cessful feeding ecology extended from Triassic pseudosu- A N D E R S O N , M. J. 2001. A new method for non- chians until present-day crocodylians. Concurrently, long- parametric multivariate analysis of variance. Austral Ecology, term behavioural and ecological separation between pseu- 26, 32–46. dosuchians and theropod dinosaurs was probably the A U G U S T I N , F. J., M A T Z K E , A. T., M A I S C H , M. W. and result of the deep phylogenetic split between them. P F R E T Z S C H N E R , H.-U. 2020. A theropod dinosaur feed- ing site from the Upper Jurassic of the Junggar Basin, NW Acknowledgements. We thank Norbert Adorf, Isabell Rosin, Mat- China. Palaeogeography, Palaeoclimatology Palaeoecology, 560, thias Boller, Cristina Gasc o-Martın, and Marit Kamenz (SMNS) 109999. for preparation of specimens and aid on access to specimens, A V I L L A , L. S., F E R N A N D E S , R. and R A M O S , D. F. B. Christoph Wimmer-Pfeil (SMNS) for the preparation of tooth 2004. Bite marks on a crocodylomorph from the Upper Creta- thin sections, and Eli Amson (SMNS) for discussion. We thank ceous of Brazil: evidence of social behavior? Journal of Verte- Hans Hagdorn (Muschelkalkmuseum, Ingelfingen) for access to brate Paleontology, 24, 971–973. specimens and discussion. This research was supported by the B E R T L I N G , M., B R A D D Y , S. J., B R O M L E Y , R. G., Swiss National Science Foundation (PZ00P2_174040) (to CF)  S, R., D E M A T H I E U , G. R., G E N I S E , J., M I K U L A and the German Science Foundation (FO 1005/2-1) (to CF). We N I E L S E N , J. K., N I E L S E N , K. S. S., R I N D S B E R G , A. acknowledge the reviews by an anonymous referee and Domenic K., S C H L I R F , M. and U C H M A N , A. 2006. Names for C. D’Amore, and the editors David Button and Sally Thomas, trace fossils: a uniform approach. Lethaia, 39, 265–286. who helped to improve a previous version of the manuscript. B I N F O R D , L. R. 1981. Bones: ancient men and modern myths. Open Access funding enabled and organized by Projekt DEAL. Academic Press, 320 pp. Author contributions. Conceptualization E Mujal, C Foth, EE B L U M E N S C H I N E , R. J., M A R E A N , C. W. and Maxwell, D Seegis, RR Schoch; Data curation E Mujal, C Foth, C A P A L D O , S. D. 1996. Blind tests of inter-analyst corre- EE Maxwell, D Seegis, RR Schoch; Formal analysis E Mujal, C spondence and accuracy in the identification of cut marks, Foth, EE Maxwell; Funding acquisition C Foth, RR Schoch; percussion marks, and carnivore tooth marks on bone sur- Investigation E Mujal, C Foth, EE Maxwell, D Seegis, RR faces. Journal of Archaeological Science, 23, 493–507. Schoch; Methodology E Mujal, C Foth, EE Maxwell; Project B O Y D , C. A., D R U M H E L L E R , S. K. and G A T E S , T. A. administration C Foth, RR Schoch; Resources C Foth, RR 2013. Crocodyliform feeding traces on juvenile ornithischian Schoch; Software E Mujal, C Foth, EE Maxwell; Supervision EE dinosaurs from the Upper Cretaceous (Cenomanian) Kaiparo- Maxwell, RR Schoch; Validation E Mujal, C Foth, EE Maxwell, wits Formation, Utah. PLoS One, 8, e57605. D Seegis, RR Schoch; Visualization E Mujal, C Foth, EE Max- B R I N K , K. S., R E I S Z , R. R., L E B L A N C , A. R. H., well, RR Schoch; Writing – original draft E Mujal; Writing – C H A N G , R. S., L E E , Y. C., C H I A N G , C. C., H U A N G , review & editing E Mujal, C Foth, EE Maxwell, D Seegis, RR T. and E V A N S , D. C. 2015. Developmental and evolutionary Schoch. novelty in the serrated teeth of theropod dinosaurs. Scientific Reports, 5, 12338. B R I T T , B. B., E B E R T H , D. A., S C H E E T Z , R. D., G R E E N H A L G H , B. W. and S T A D T M A N , K. L. 2009. DATA ARCHIVING STATEMENT Taphonomy of debris-flow hosted dinosaur bonebeds at Dal- ton Wells, Utah (Lower Cretaceous, Cedar Mountain Forma- Data for this study, including Data S1 (bite trace database) and tion, USA). Palaeogeography, Palaeoclimatology Palaeoecology, Data S2 (Batrachotomus teeth database), are also available in the Dryad Digital Repository: https://doi.org/10.5061/dryad.stqjq2c4b 280, 1–22. B R O W N , C., T A N K E , M. D. H. and H O N E , D. W. E. 2021. Rare evidence for ‘gnawing-like’ behavior in a small- Editor. David Button bodied theropod dinosaur. PeerJ, 9, e11557. B R U N N E R , H. and B R U D E R , J. 1981. Standardprofile des Unteren Keupers (Lettenkeuper, Trias) im n€ ordlichen Baden- SUPPORTING INFORMATION W€ urttemberg. Jahresberichte und Mitteilungen des Oberrhei- nischen Geologische Vereins, N.F, 63, 253–269. Additional Supporting Information can be found online (https:// B U F F E T A U T , E. 1983. Wounds on the jaw of an Eocene doi.org/10.1111/pala.12597): mesosuchian crocodilian as possible evidence for the antiquity Data S1. Database of bite traces on Lower Keuper bones. of crocodilian intraspecific fighting behaviour. Pal€aontologische Data S2. Database of Batrachotomus teeth, with tooth mea- Zeitschrift, 57, 143–145. surements and macroscopic wear. C H A R I G , A. J. and M I L N E R , A. C. 1997. Baryonyx walkeri, Appendix S1. Includes Tables S1–S9, which summarize the a fish-eating dinosaur from the Wealden of Surrey. Bulletin of statistical analysis on bite traces. the Natural History Museum (Geology Series), 53, 11–70. MUJAL ET AL.: FEEDING ECOLOGY OF ARCHOSAURS 29 C O T T , H. B. 1961. Scientific results of an inquiry into the G O R D O N , C. M., R O A C H , B. T., P A R K E R , W. G. and ecology and economic status of the Nile crocodile (Crocodilus B R I G G S , D. E. G. 2020. Distinguishing regurgitalites and niloticus) in Uganda and Northern Rhodesia. Transactions of coprolites: a case study using a Triassic bromalite with soft tis- the Zoological Society of London, 29, 211–357. sue of the pseudosuchian archosaur Revueltosaurus. Palaios, D ’ A M O R E , D. C. 2009. A functional explanation for denticu- 35, 111–121. lation in theropod dinosaur teeth. The Anatomical Record, G O W E R , D. J. 1999. The cranial and mandibular osteology of 292, 1297–1314. a new rauisuchian archosaur from the Middle Triassic of D ’ A M O R E , D. C. and B L U M E N S C H I N E , R. J. 2009. southern Germany. Stuttgarter Beitr€age zur Naturkunde. Serie Komodo monitor (Varanus komodoensis) feeding behavior and B (Geologie und Pal€aontologie), 280, 1–49. dental function reflected through tooth marks on bone sur- G O W E R , D. J. and S C H O C H , R. R. 2009. The postcranial faces, and the application to ziphodont paleobiology. Paleobi- skeleton of the rauisuchian Batrachotomus kupferzellensis. Jour- ology, 35, 525–552. nal of Vertebrate Paleontology, 29, 103–122. D ’ A M O R E , D. C. and B L U M E N S C H I N E , R. J. 2012. H A G D O R N , H., S C H O C H , R. R., S E E G I S , D. and W E R - Using striated tooth marks on bone to predict body size in N E B U R G , R. 2015. Wirbeltierlagerst€atten im Lettenkeuper. theropod dinosaurs: a model based on feeding observations of 325–358. In H A G D O R N , H., S C H O C H , R. R. and Varanus komodoensis, the Komodo monitor. Paleobiology, 38, S C H W E I G E R T , G. (eds). Der Lettenkeuper – Ein Fenster in 79–100. die Zeit vor den Dinosauriern. Staatliches Museum f€ ur Natur- D E V I T A , D. E. 1979. Niche separation and the broken-stick kunde Stuttgart. model. The American Naturalist, 114, 171–178. H A M M E R , Ø. 2020. PAST: PAleontological STatistics. Version D R U M H E L L E R , S. K. and B R O C H U , C. A. 2014. A diag- 4.03. Reference manual. University of Oslo, Oslo. https://www. nosis of Alligator mississippiensis bite marks with comparisons nhm.uio.no/english/research/infrastructure/past/ to existing crocodylian datasets. Ichnos, 21, 131–146. H A M M E R , Ø. and H A R P E R , D. A. T. 2006. Paleontological D R U M H E L L E R , S. K. and B R O C H U , C. A. 2016. Phyloge- data analysis. Blackwell Publishing, 351 pp. netic taphonomy: a statistical and phylogenetic approach for H A M M E R , Ø., H A R P E R , D. A. T. and R Y A N , P. D. 2001. exploring taphonomic patterns in the fossil record using cro- PAST: paleontological statistics software package for education codylians. Palaios, 31, 463–478. and data analysis. Palaeontologia Electronica, 4, 1–9. D R U M H E L L E R , S. K., M c H U G H , J. B., K A N E , M., H E N D R I C K X , C., M A T E U S , O. and A R A UJ  O , R. 2015. R I E D E L , A. and D ’ A M O R E , D. C. 2020. High frequencies The dentition of megalosaurid theropods. Acta Palaeontologica of theropod bite marks provide evidence for feeding, scaveng- Polonica, 60, 627–642. ing, and possible cannibalism in a stressed Late Jurassic eco- H O F F M A N , D. K., E D W A R D S , H. R., B A R R E T T , P. M. system. PLoS One, 15, e0233115. and N E S B I T T , S. J. 2019. Reconstructing the archosaur D R U M H E L L E R , S. K., S T O C K E R , M. R. and N E S B I T T , radiation using a Middle Triassic archosauriform tooth assem- S. J. 2014. Direct evidence of trophic interactions among apex blage from Tanzania. PeerJ, 7, e7970. predators in the Late Triassic of western North America. Nat- H O N E , D. W. E. and C H U R E , D. J. 2018. Difficulties in urwissenschaften, 101, 975–987. assigning trace makers from theropodan bite marks: an exam- D R Y M A L A , S. M., B A D E R , K. and P A R K E R , W. G. 2021. ple from a young diplodocoid sauropod. Lethaia, 51, 456–466. Bite marks on an aetosaur (Archosauria, Suchia) osteoderm: H O N E , D. W. E. and R A U H U T , O. W. M. 2010. Feeding assessing Late Triassic predator-prey ecology through ichnol- behaviour and bone utilization by theropod dinosaurs. ogy and tooth morphology. Palaios, 36, 28–37. Lethaia, 43, 232–244. E T Z O L D , A. and S C H W E I Z E R , V. 2005. Der Keuper in H U N G E R B UH€ L E R , A. 1998. Taphonomy of the prosauro- Baden-W€ urttemberg. 215–258. In D E U T S C H E S T R A T I - pod dinosaur Sellosaurus, and its implications for carnivore G R A P H I S C H E K O M M I S S I O N (eds). Stratigraphie von faunas and feeding habits in the Late Triassic. Palaeogeography, Deutschland IV. Keuper. Bearbeitet von der Arbeitsgruppe Keu- Palaeoclimatology Palaeoecology, 143, 1–29. per der Subkommission Perm-Trias der DSK. Courier For- J A C K S O N , D. A. 1993. Stopping rules in principal compo- schungsinstitut Senckenberg, 253 pp. nents analysis: a comparison of heuristical and statistical F I O R I L L O , A. R. 1991. Prey bone utilization by predatory approaches. Ecology, 74, 2204–2214. dinosaurs. Palaeogeography, Palaeoclimatology Palaeoecology, J A C O B S E N , A. R. 1998. Feeding behaviour of carnivorous 88, 157–166. dinosaurs as determined by tooth marks on dinosaur bones. FRANC ß A , M. A. G., F E R I G O L O , J. and L A N G E R , M. C. Historical Biology, 13, 17–26. 2011. Associated skeletons of a new Middle Triassic “Rauisu- J A C O B S E N , A. R. and B R O M L E Y , R. G. 2009. New ichno- chia” from Brazil. Naturwissenschaften, 98, 389–395. taxa based on tooth impressions on dinosaur and whale G I G N A C , P. M. and E R I C K S O N , G. M. 2017. The biome- bones. Geological Quarterly, 53, 373–382. chanics behind extreme osteophagy in Tyrannosaurus rex. Sci- € I N , J. A. 1936. Uber K AL € Skelettanomalien bei Crocodiliden. entific Reports, 7, 2012. Zeitschrift f€ur Morphologie und Oekologie der Tiere, 32, 327– ^ E T , J., R O Z A D A , L., B O U R G E A I S , R. and G ON 347. A L L A I N , R. 2019. Taphonomic study of a pleurosternid tur- K A T S U R A , Y. 2004. Paleopathology of Toyotamaphimeia tle shell from the Early Cretaceous of Angeac-Charente, south- machikanensis (Diapsida, Crocodylia) from the Middle Pleisto- west France. Lethaia, 52, 232–243. cene of Central Japan. Historical Biology, 16, 93–97. 30 PALAEONTOLOGY K L E I N , N., F O T H , C. and S C H O C H , R. R. 2017. Prelimi- Fenster in die Zeit vor den Dinosauriern. Staatliches Museum nary observations on the bone histology of the Middle Triassic f€ ur Naturkunde Stuttgart. pseudosuchian archosaur Batrachotomus kupferzellensis reveal N J A U , J. K. and B L U M E N S C H I N E , R. J. 2006. A diagnosis fast growth with laminar fibrolamellar bone tissue. Journal of of crocodile feeding traces on larger mammal bone, with fossil Vertebrate Paleontology, 37, e1333121. examples from the Plio-Pleistocene Olduvai Basin, Tanzania. L O N G R I C H , N. R., H O R N E R , J. H., E R I C K S O N , G. M. Journal of Human Evolution, 50, 142–162. and C U R R I E , P. J. 2010. Cannibalism in Tyrannosaurus rex. N J A U , J. K. and G I L B E R T , H. G. 2016. A taxonomy for PLoS One, 5, e13419. crocodile-induced bone modifications and their relevance to M A R S H , A. D. and R O W E , T. B. 2020. A comprehensive paleoanthropology. FOROST Occasional Publications, 3, 1–13. anatomical and phylogenetic evaluation of Dilophosaurus N O T O , C. R., M A I N , D. J. and D R U M H E L L E R , S. K. wetherilli (Dinosauria, Theropoda) with descriptions of new 2012. Feeding traces and paleobiology of a Cretaceous (Ceno- specimens from the Kayenta Formation of northern Arizona. manian) crocodyliform: example from the Woodbine Forma- Journal of Paleontology, 94, 1–103. tion of Texas. Palaios, 27, 105–115. MIKULA  S, R., K A D L E C O V A,  E., F E J F A R , O. and P I R R O N E , C. A., B U A T O I S , L. A. and B R O M L E Y , R. G. DVOR  , Z. 2006. Three new ichnogenera of biting and  AK 2014. Ichnotaxobases for bioerosion trace fossils in bones. gnawing traces on reptilian and mammalian bones: a case study Journal of Paleontology, 88, 195–203. from the Miocene of the Czech Republic. Ichnos, 13, 113–127. P O B I N E R , B. 2008. Paleoecological information in predator M U J A L , E. and S C H O C H , R. R. 2020. Middle Triassic (Ladi- tooth marks. Journal of Taphonomy, 6, 373–397. nian) amphibian tracks from the Lower Keuper succession of P U J O S , F. and S A L A S - G I S M O N D I , R. 2020. Predation of southern Germany: Implications for temnospondyl locomo- the giant Miocene caiman Purussaurus on a mylodontid tion and track preservation. Palaeogeography, Palaeoclimatology ground sloth in the wetlands of proto-Amazonia. Biology Let- Palaeoecology, 543, 109625. ters, 16, 20200239. M U J A L , E., F O T H , C., M A X W E L L , E. E., S E E G I S , D. € , M., A H L B E R G , P. E. and N I E - Q V A R N S T R OM and S C H O C H , R. R. 2022. Data from: Feeding habits of the  I E D Z K I , G. 2019. Tyrannosaurid-like osteophagy by a D ZW Middle Triassic pseudosuchian Batrachotomus kupferzellensis Triassic archosaur. Scientific Reports, 9, 925. from Germany and palaeoecological implications for archo- R O G E R S , R. R., K R A U S E , D. W. and R O G E R S , K. C. saurs. Dryad Digital Repository. https://doi.org/10.5061/dryad. 2003. Cannibalism in the Madagascan dinosaur Majungatholus stqjq2c4b atopus. Nature, 422, 515–518. N E S B I T T , S. J. 2011. The early evolution of archosaurs: rela- R O Z A D A , L., A L L A I N , R., V U L L O , R., G O E D E R T , J., tionships and the origin of major clades. Bulletin of the Ameri- A U G I E R , D., J E A N , A., M A R C H A L , J., P E Y R E D E can Museum of Natural History, 352, 1–292.  U E S , C., Q V A R N S T R OM F A B R EG € , M. and R O Y O - N E S B I T T , S. J., B R U S A T T E , S. L., D E S O J O , J. B., T O R R E S , R. 2021. A Lower Cretaceous Lagerst€atte from L I P A R I N I , A., D E F R A N C ß A , M. A. G., W E I N B A U M , France: a taphonomic overview of the Angeac-Charente verte- J. C. and G O W E R , D. 2013. Rauisuchia. 241–274. In N E S - brate assemblage. Lethaia, 54, 141–165. B I T T , S. J., D E S O J O , J. B. and I R M I S , R. B. (eds). Anat- S A D L E I R , R., B A R R E T T , P. M. and P O W E L L , H. P. omy, phylogeny and palaeobiology of early archosaurs and their 2008. The anatomy and systematics of Eustreptospondylus oxo- kin. Geological Society, London, Special Publications, 379. niensis, a theropod dinosaur from the Middle Jurassic of N E S B I T T , S. J., Z A W I S K I E , J. M. and D A W L E Y , R. M. Oxfordshire, England. Palaeontographical Society Monograph 2020. The osteology and phylogenetic position of the loricatan London, 160, 1–82. (Archosauria: Pseudosuchia) Heptasuchus clarki, from the ? S A N D E R , P. M. 1999. The microstructure of reptilian tooth Mid-Upper Triassic, southeastern Big Horn Mountains, Cen- enamel: terminology, function, and phylogeny. M€ unchner Geo- tral Wyoming (USA). PeerJ, 8, e10101. wissenschaftliche Abhandlungen A, 38, 1–102.  I E D Z K I , G., G O R Z E L A K , P. and S U L E J , T. N I E D ZW S C H O C H , R. R. 1999. Comparative osteology of Mastodon- 2011. Bite traces on dicynodont bones and the early evolution saurus giganteus (Jaeger, 1828) from the Middle Triassic (Let- of large terrestrial predators. Lethaia, 44, 87–92. tenkeuper: Longobardian) of Germany (Baden-W€ urttemberg, N I T S C H , E. 2015a. Der Lettenkeuper – Verbreitung, Alter, Bayern, Th€ uringen). Stuttgarter Beitr€age zur Naturkunde. Serie Pal€aogeographie. 9–16. In H A G D O R N , H., S C H O C H , R. B (Geologie und Pal€aontologie), 278, 1–175. R. and S C H W E I G E R T , G. (eds). Der Lettenkeuper – Ein S C H O C H , R. R. and M I L N E R , A. R. 2000. Handbuch der Fenster in die Zeit vor den Dinosauriern. Staatliches Museum Pal€aoherpetologie 3B: Stereospondyli. F. Pfeil, 203 pp. f€ ur Naturkunde Stuttgart. S C H O C H , R. R. and S E E G I S , D. 2016. A Middle Triassic N I T S C H , E. 2015b. Lithostratigraphie des Lettenkeupers. 27– palaeontological gold mine: the vertebrate deposits of Vellberg 40. In H A G D O R N , H., S C H O C H , R. R. and S C H W E I - (Germany). Palaeogeography, Palaeoclimatology Palaeoecology, G E R T , G. (eds). Der Lettenkeuper – Ein Fenster in die Zeit 459, 249–267. vor den Dinosauriern. Staatliches Museum f€ ur Naturkunde S C H O C H , R. R., U L L M A N N , F., R O Z Y N E K , B., Z I E G - Stuttgart. L E R , R., S E E G I S , D. and S U E S , H.-D. 2018. Tetrapod N I T S C H , E. 2015c. Fazies und Ablagerungsr€aume des Letten- diversity and palaeoecology in the German Middle Triassic keupers. 285–324. In H A G D O R N , H., S C H O C H , R. R. (Lettenkeuper) documented by tooth morphotypes. Palaeobio- and S C H W E I G E R T , G. (eds). Der Lettenkeuper – Ein diversity & Palaeoenvironments, 98, 615–638. MUJAL ET AL.: FEEDING ECOLOGY OF ARCHOSAURS 31 S C H U B E R T , B. W. and U N G A R , P. S. 2005. Wear facets W A N G , C.-C., S O N G , Y.-F., S O N G , S.-R., J I , Q., and enamel spalling in tyrannosaurid dinosaurs. Acta Palaeon- C H I A N G , C.-C., M E N G , Q., L I , H., H S I A O , K., L U , tologica Polonica, 50, 93–99. Y.-C., S H E W , B.-Y., H U A N G , T. and R E I S Z , R. R. 2015. S H A L E , Y., E L Z A A T A R I , S. and W H I T E , T. D. 2017. Evolution and function of dinosaur teeth at ultramicrostruc- Hominid butchers and biting crocodiles in the African tural level revealed using synchrotron transmission X-ray Plio–Pleistocene. Proceedings of the National Academy of Sci- microscopy. Scientific Reports, 5, 15202. ences, 114, 13164–13169. W E B B , G. J. W. and M A N O L I S , S. C. 1983. Crocodylus john- S M I T H , J. B., V A N N , D. R. and D O D S O N , P. 2005. Dental stoni in the McKinlay river area, N.T.V.* Abnormalities and morphology and variation in theropod dinosaurs: implications injuries. Australian Wildlife Research, 10, 407–420. for the taxonomic identification of isolated teeth. The Ana- W H I T N E Y , M. R., L E B L A N C , A. R. H., R E Y N O L D S , A. tomical Record, 285A, 699–736. R. and B R I N K , K. S. 2020. Convergent dental adaptations in S N I V E L Y , E., C O T T O N , J. R., R I D G E L Y , R. C. and the serrations of hypercarnivorous synapsids and dinosaurs. W I T M E R , L. M. 2013. Multibody dynamics model of head Biology Letters, 16, 20200750. and neck function in Allosaurus (Dinosauria, Theropoda). W I L D , R. 1978. Massengrab f€ ur Saurier. Kosmos, 11, 790–797. Palaeontologia Electronica, 16, 11A. W I L D , R. 1979. Saurier kommen ans Licht. Tierwelt, 4, 38–45. S T E H M A N , S. V. 1997. Selecting and interpreting measures of W I L D , R. 1980. The fossil deposits of Kupferzell, Southwest thematic classification accuracy. Remote Sensing of Environ- Germany. Mesozoic Vertebrate Life, 1, 15–18. ment, 62, 77–89. W I L L S , M. A., B R I G G S , D. E. G. and F O R T E Y , R. A. U R L I C H S , M. 1982. Zur Stratigraphie und Fossilf€ uhrung des 1994. Disparity as an evolutionary index: a comparison of Lettenkeupers (Ob. Trias) bei Schw€abisch Hall (Baden- Cambrian and recent arthropods. Paleobiology, 20, 93–130. W€ urttemberg). Jahresberichte und Mitteilungen des Oberrhei- Y O U N G , M. T., B R U S A T T E , S. L., B E A T T Y , B. L., nischen Geologische Vereins, N.F, 64, 213–224. A N D R A D E , M. B. and D E S O J O , J. B. 2012. Tooth-on- V A L L O N , L. H., R I N D S B E R G , A. K. and M A R T I N , A. J. tooth interlocking occlusion suggests macrophagy in the 2015. The use of the terms trace, mark and structure. Annales Mesozoic marine crocodylomorph Dakosaurus. The Anatomi- Societatis Geologorum Poloniae, 85, 527–528. cal Record, 295, 1147–1158.