Search

Supplementary Material 2

After death, bone is exposed to various abiotic and biotic processes. Microbial bioerosion has been reported from recent, archaeological and palaeontological bones and is deemed to have been caused by fungi, cyanobacteria or bacteria. It is recognizable as microscopic tunnelling, caused by microbial enzymes. However, fossil (i.e., fully permineralized) bone exhibits strikingly few signs of bioerosion. Fossilization only happens if bone is preserved long enough to allow diagenesis by mineral uptake and loss of collagen.Upper Jurassic sauropod dinosaur long and girdle bones from Kenton quarries 1, 5, and 6 (Oklahoma, Morrison Formation) present abundant and distinctive filamentous microstructures. This kind of preservation is absent in hundreds of other Morrison Formation dinosaur bone samples studied.Methods used to evaluate the nature of microstructures include literature research (comprehensive review of palaeontological and archaeological literature) and thin section analysis. Polarized light microscopy was used to describe the general bone histology, ontogeny, and density and distribution of bioerosive microstructures. The aim of this work is to identify bioerosive microstructures and to elucidate possible causative agents, bioerosive mechanisms and invasion patterns. The conclusion is that the filamentous microstructure are most likely Wedl tunnels caused by fungal attack. Presence of tunnelling seems to be unrelated to the specific quarry, ontogenetic stage, taxon, and bone type.Further research on the taphonomic and depositional environments of the Kenton quarries are needed to identify the cause of these unique microstructures. Heavily bioeroded bone and its rarity allows to constrain hypotheses regarding bone fossilization.

The preservation of fossils over long periods of geological history usually has special characteristics, and the specific conditions for successful fossilization have not been established yet. Triassic bones from SW China are characterized by an exceptionally broad variation in the style of preservation. This region presents a perfect natural lab for the study of conditions advantageous for bone preservation. Twenty-five specimens of marine reptiles from the Middle to Late Triassic black shales were sampled and examined in petrographic thin sections, by Raman spectroscopy and wavelength-dispersive X-ray spectroscopy. Our results revealed evidence that the bones were heated, possibly by hydrothermal fluids or by regional geological events, and we observed fluorite in fossil bone for the first time. We discriminate five stages of bone alteration: 1) no alteration beyond diagenetic mineral precipitation in the bone porosity; 2) weak alteration of apatite and migration of kerogen from the shale into the fossil bone; 3) formation of authigenic fluorite needles; 4) Recrystallization of bone apatite to idiomorphic apatite while bone structure is still present; 5) complete loss of bone structure and replacement by idiomorphic apatite. Systematic carbon Raman thermometry returns temperatures up to 150-180°C, indicating that loss of histological information might be linked to temperatures near to low-grade metamorphism. This study provides reference data for different stages of bone alteration, which is important for the study of bone diagenetic processes in dependence of higher than near-surface temperatures.

Varanids are the only non-avian sauropsids that are known to approach the warm-blooded mammals in stamina. Furthermore, a much higher maximum metabolic rate (MMR) gives endotherms (including birds) higher stamina than crocodiles, turtles, and non-varanid lepidosaurs. This has led researchers to hypothesize that mammalian endothermy evolved as a second step after the acquisition of elevated MMR in non-mammalian therapsids from a plesiomorphic state of low metabolic rates. In recent amniotes, MMR correlates with the index of blood flow into the femur (Qi), which is calculated from femoral length and the cross-sectional area of the nutrient foramen. Thus, Qi may serve as an indicator of MMR range in extinct animals. Using the Qi proxy and phylogenetic eigenvector maps, here we show that elevated MMRs evolved near the base of Synapsida. Non-mammalian synapsids, including caseids, edaphosaurids, sphenacodontids, dicynodonts, gorgonopsids, and non-mammalian cynodonts, show Qi values in the range of recent endotherms and varanids, suggesting that raised MMRs either evolved in synapsids shortly after the Synapsida-Sauropsida split in the Mississippian or that the low MMR of lepidosaurs and turtles is apomorphic, as has been postulated for crocodiles.

Cymbospondylus Leidy, 1868 (14) Type species Cymbospondylus piscosus Leidy, 1868 (14) Referred species Cymbospondylus petrinus Leidy, 1868 (14); Cymbospondylus buchseri (15); Cymbospondylus nichollsi (16); *Corresponding author. Email: martin.sander@uni-bonn.de (P.M.S.); em.griebeler@uni-mainz.de (E.M.G.); lschmitz@kecksci.claremont.edu (L.S.) Cymbospondylus duelferi (17); Cymbospondylus youngorum sp. nov. Genus diagnosis For a recent detailed diagnosis of the genus, see (17). In addition, the results of the phylogenetic analysis in this study offer an apomorphy-based diagnosis., Published as part of Sander, P. Martin, Griebeler, Eva Maria, Klein, Nicole, Juarbe, Jorge Velez, Wintrich, Tanja, Revell, Liam J. & Schmitz, Lars, 2021, Early giant reveals faster evolution of large body size in ichthyosaurs than in cetaceans, pp. 1-15 in Science (New York, N. Y.) (New York, N. Y.) 374 (6575) on pages 1-2, DOI: 10.1126/science.abf5787, http://zenodo.org/record/5805321, {"references":["14. J. Leidy, Notice of some reptilian remains from Nevada. Proc. Acad. Nat. Sci. Philadelphia 20, 177 - 178 (1868).","15. P. M. Sander, The large ichthyosaur Cymbospondylus buchseri, sp. nov., from the Middle Triassic of Monte San Giorgio (Switzerland), with a survey of the genus in Europe. J. Vertebr. Paleontol. 9, 163 - 173 (1989). doi: 10.1080 / 02724634.1989. 10011750","16. N. FrObisch, P. M. Sander, O. Rieppel, A new species of Cymbospondylus (Diapsida, Ichthyosauria) from the Middle Triassic of Nevada and re-evaluation of the skull osteology of the genus. Zool. J. Linn. Soc. 147, 515 - 538 (2006). doi: 10.1111 / j. 1096 - 3642.2006.00225. x","17. N. Klein, L. Schmitz, T. Wintrich, P. M. Sander, A new cymbospondylid ichthyosaur (Ichthyosauria) from the Middle Triassic (Anisian) of the Augusta Mountains, Nevada, USA. J. Syst. Palaeontology 18, 1167 - 1191 (2020). doi: 10.1080 / 14772019.2020. 1748132"]}

Cymbospondylus youngorum sp. nov. (Fig. 2 and figs. S1 to S5) Etymology The species is named in honor of Tom and Bonda Young. Holotype and only specimen LACM DI 157871 is the holotype and, as of now, the only recognized individual. LACM DI 157871 is largely articulated and complete from the anterior of the trunk region to the skull, preserved ventral side up. The cervical column back to the middle dorsal vertebrae is present with ribs in articulation. The shoulder girdle is articulated, and the two humeri are situated close to their respective glenoid. At present, the skull (Fig. 2, A to F; figs. S1 and S2; and table S1), the right humerus (Fig. 2, G to K), parts of the shoulder girdle (fig. S3), and some vertebrae are prepared (fig. S3) Horizon and locality LACM DI 157871 comes from the Anisian age Fossil Hill Member of the Favret Formation at Favret Canyon, Augusta Mountains, Pershing County, Nevada, USA. The type locality, LACM 8025, is on the northern slope of the rear of Favret Canyon at an elevation of 1676 m. Exact coordinates are on file at the repository. LACM DI 157871 originates from the same general level as the holotype of the macropredatory ichthyosaur Thalattoarchon saurophagis (8) and the pistosaur Augustasaurus hagdorni (10) in the lower third of the Fossil Hill member and pertains to the middle Anisian Taylori Zone (18). Diagnosis C. youngorum sp. nov. is diagnosed by a unique combination of the following eight characters (see data S1 for character descriptions and data S3 for a list of synapomorphies): squamosal, participates in supratemporal fenestra (character 72, state 0; Fig. 2); dentary, labial shelf present (character 117, state 1; fig. S2C); angular, extent of anterior lateral exposure is extensive, at least as high and anteriorly as the surangular’ s exposure (character 120, state 1; Fig. 2); angular, extent of posterior lateral exposure is extensive, with surangular exposure reduced to a thin strip on the lateral surface of the retroarticular process (character 121, state 1; Fig. 2); lower jaw glenoid, deeply excavated and present (character 126, state 1; Fig. 2); dentition, definition of the base of the enamel layer is well defined and precise (character 147, state 1; fig. S2, E and F); humerus, anterior flange is absent (character 200, state 0; Fig. 2 and fig. S6); and humerus, relative anteroposterior width in dorsal view, excluding dorsal and ventral processes, is approximately equal or the proximal end is wider than the distal end (character 206, state 1; Fig. 2 and fig. S6). The new taxon is characterized by the following autapomorphies: a thick base of bone of attachment of the teeth (Fig. 2 and fig. S2, C and E), the distinctive shape of the scapula with a very large and wide dorsal blade and a narrow ventral part (figs. S3 and S6), the distinctive humerus morphology with a wider proximal than distal end, and a triangular proximal head and triangular shaft cross section (Fig. 2 and fig. S6). Note that these autapomorphies were not added as characters to our character matrix. A differential diagnosis and detailed anatomical descriptions (figs. S2 and S3) and comparisons (figs. S4 to S6 and tables S2 and S3) are provided in the supplementary materials (10). Phylogenetic position Phylogenetic analyses (10) (table S4) indicate that C. youngorum sp. nov. is nested within a clade of closely related Cymbospondylus species that account for much of the lineage diversity and morphological disparity of large-bodied Early and Middle Triassic ichthyosaurs (Fig. 3 and fig. S7). The close relationship of these species points to an adaptive radiation (as much as one can be recognized in the fossil record of Mesozoic reptiles). Further evidence for such a radiation is that there are no other four seemingly sympatric species of any ichthyosaur genus in the ichthyosaur record and that other finds of Cymbospondylus from the late Early and early Middle Triassic are widely distributed across the Northern Hemisphere (16). Our analysis with TNT (a software for phylogenetic analysis) and its “new technology” search algorithm resulted in a tree length of 1225 steps. The four most parsimonious trees were retained (table S4), and the nearly fully resolved strict consensus of the four trees is shown in Fig. 3 and fig. S7. The consistency index of this tree is 0.259, coupled with a retention index of 0.627. The absolute Bremer support of the nodes varies from 1 to 5 (fig. S7). Additional analyses (table S4) confirmed the placement of LACM DI 157871 in a clade with other cymbospondylids, yet the position of the Cymbospondylus clade varies with the selection of taxa that were included in the analyses (10). We note that the interrelationships of ichthyosaurs remain difficult to resolve, both in the Triassic and the Jurassic part of the tree. This uncertainty reflects the difficulty in resolving ichthyosaur interrelationships in general (19) and the need for a concerted effort of redefining and rescoring characters. Inferred diet and estimated body size The conical, bluntly pointed tooth crowns of C. youngorum sp. nov., in conjunction with the elongate snout, suggest a generalist diet of fish and squid (20), as inferred for most ichthyosaurs from teeth and stomach contents (21). Considering its size, C. youngorum sp. nov. could also have preyed on smaller and juvenile marine reptiles (10). The right lower jaw of C.youngorum sp. nov. measures 1970 mm from the tip of the dentary to the end of the retroarticular process (table S1). At a total length of 1890 mm (table S1), the skull of LACM DI 15787 is one of the largest complete ichthyosaur skulls known. Although post-Triassic ichthyosaurs never reached the size of Triassic ones again (7), there are skulls of Temnodontosaurus from the Lower Jurassic of England and Germany (22) that are the same length as that of LACM DI 15787 within the limits of preservation. However, these large ichthyosaurs probably were less than 9 m long, having proportionally larger skulls (22). Larger skulls than those of LACM DI 15787 and these largest Temnodontosaurus specimens are only known from Late Triassic ichthyosaurs, specifically Shonisaurus popularis and Shastasaurus sikanniensis, with estimated skull lengths of 2750 and 3000 mm,respectively (7). Humerus length is another commonly used proxy for ichthyosaurian body size [(16, 17, 23); fig. S8A], even though it is available for fewer ichthyosaur taxa than skull length. At 453 mm, the humerus of the holotype of C. youngorum sp. nov. is the second largest ichthyosaurian humerus known, translating into a total length of 17.65 m (10). The lower 95% prediction interval of its length is 12.48 m; the upper 95% prediction interval is 24.96 m (10) (fig. S8A). We also estimated body mass based on a recent dataset for ichthyosaurs (10, 24) (table S5). The regression function (fig. S8B) yielded a body mass estimate of 44,699 kg for the 17.65-msized LACM DI 15787 specimen (table S5). The lower mass based on the 95% prediction interval of both regressions is 14,712 kg (lower limit of the 95% prediction interval of body mass against length evaluated for 12.48 m); the upper mass is 135,809 kg (upper limit of the 95% prediction interval of body mass against length evaluated for 24.96 m). These body size metrics of C. youngorum sp. nov. result in one of the highest length and mass estimates for any ichthyosaur and the largest taxon of the Middle Triassic: Based on length and scaled-up mass data in Gutarra et al. (24), S. popularis from the late Carnian (about 230 Ma ago) was 13.5 m long and had a mass of 21,651 kg, whereas the middle Norian (about 212 Ma ago) S. sikanniensis was 21 m long and had an estimated body mass of 81,497 kg. Taken together, these length and mass estimates place ichthyosaurs in the range of body sizes of living cetaceans, but it appears that ichthyosaurs reached their largest sizes much earlier in clade history than whales.