Reconstruction of feeding behaviour and diet in Devonian ctenacanth chondrichthyans using dental microwear texture and finite element analyses

Abstract

Devonian ctenacanth chondrichthyans reached body sizes similar to modern great white sharks and therefore might have been apex predators of the Devonian seas. However, very little is known about the diet and feeding behaviours of these large ancestral sharks. To reconstruct their ecological properties, teeth of the large Famennian (Late Devonian) chondrichthyan Ctenacanthus concinnus from the Anti-Atlas, Morocco, were analysed. The teeth show strong tooth wear with deep horizontal as well as vertical scratches. Dental microwear texture analysis, a well-established method for the reconstruction of diet and commonly used in terrestrial vertebrates, was applied for the first time, to our knowledge, to Palaeozoic vertebrates in this study. Furthermore, finite element analysis was performed to test the biomechanical properties of the teeth. By combining both analyses, as well as palaeoenvironmental data and tooth morphology, we demonstrate that the results from only one method can be insufficient and misleading. Ctenacanthus concinnus most likely was an opportunistic feeder like many modern sharks. Direct evidence and the results of our analyses suggest that Ctenacanthus fed on ectocochleate cephalopods, other chondrichthyans and further vertebrates using a combination of head movements including lateral head shaking to cut large prey items.

Keywords: DMTA; Devonian; FEA; chondrichthyans; feeding behaviour; tooth wear.

Figure 1.

(a) PIMUZ A/I 5253 (tooth A, all teeth in electronic supplementary material, figures S1 and S2) and detailed scanning electron microscope image (scale bar = 500 μm) showing the wear in the abapical part of the wear facet. The tooth shows a highly worn wear facet and strong wear in the outermost cusplets, mainly visible on the left cusplet. (b) Wear facet of PIMUZ A/I 5254 (tooth B, all teeth in electronic supplementary material, figures S1 and S2) as well as an overview picture of tooth B. Red lines represent main scratches and their directions. A rose diagram (x-axis = scratch angle; y-axis = number of scratches) of all counted scratches shows that there are two main directions: horizontal and vertical. The vertical scratches are mainly in the apical part of the wear facet while the horizontal scratches are more in the middle to abapical parts of the wear facet. Scale bars 5 mm.

Figure 2.

Overview of Ctenacanthus concinnus teeth. (a) PIMUZ A/I 5253 from labial, lingual, mesial view (left to right). (b) PIMUZ A/I 5262 tooth embedded in matrix in lingual view. The tooth shows the originally pointed apex with little sign of apical wear. (c) PIMUZ A/I 5262 tooth in matrix and surrounded by scales, from the labial side showing the apex and three cusplets on each side. Scale bars 5 mm.

Figure 3.

PIMUZ A/I 5263. Ctenacanthus tumidus from Madene El Mrakib, Famennian. (a) Overview of the entire block containing several Ct. tumidus teeth. The black box indicates which tooth is shown in detail in (b–e). Scale bar 1 mm. (b) Close-up of the apical part of the Ct. tumidus tooth showing the labial striation along the tooth as well as serration along the cutting edge, scale bar = 1 mm. (c) Close-up of the serrated cutting edge of Ct. tumidus, scale bar 0.5 mm. (d) SEM picture along the cutting edge of the tooth, scale bar 200 μm. (e) Detailed SEM picture of one tooth of the serrated cutting edge showing microwear pattern, scale bar 100 μm.

Figure 4.

Results of the surface texture parameters epLsar—anisotropy and Asfc—complexity. The plots in (a) differentiate between the teeth of the two specimens. Teeth A–G, PIMUZ A/I 5253-5259 belong to one specimen and teeth H and I, PIMUZ A/I 5260 and 5261 belong to the skeleton PIMUZ A/I 5262. The plots in (b) differentiate between measurements that were taken in the apical parts of the wear facet and measurements that were taken in the abapical parts of the facet of each tooth. Top plots show data that were collected with a 100× objective. Bottom plots show data that were collected with a 20× objective. NMP_cat = non-measured points category; the percentage of unmeasured points of the measurement.

Figure 5.

Principal components analysis (PCA) of the surface texture parameters displayed in the biplots, each for the first two principal components. (a) Two PCA plots differentiating between different specimens. Teeth A–G, PIMUZ A/I 5253-5259 belong to one specimen and teeth H and I, PIMUZ A/I 5260 and 5261, belong to the skeleton PIMUZ A/I 5262. (b) Two plots differentiating between the different positions (apical and abapical) that the measurements were taken on the wear facet of each tooth. The top plots represent measurements taken with a 100× objective. The bottom plots represent measurements taken with a 20× objective.

Figure 6.

Finite element models of a Ctenacanthus tooth (PIMUZ A/I 5264, tooth J) and modern shark teeth, loaded from different directions. (a) Load applied to the tip, representing a puncture. (b) Load applied from distal and mesial, representing cutting through head shaking movements. (c) Load applied from labial and (d) lingual, representing holding of prey items.

Figure 7.

Chondrichthyan and ammonoid body size, and the numbers of lobes and sutural fractal dimension in ammonoid conchs, across the Devonian and Carboniferous, showing an increase in all four factors towards the Famennian. See also supplementary tables S1 to S4 and supplementary references. Excel sheet on Zenodo: https://doi.org/10.5281/zenodo.13772272.

Figure 8.

PIMUZ A/I 5338. Ctenacanth fossil from the Famennian of the Anti-Atlas, Morocco, with a preserved fin spine of another chondrichthyan in its mouth. (a) Overview of the entire fossil. (b) Overview of the skull, the black box indicates where (c), a chondrichthyan fin spine of a different taxa, is located.