FIGURE 1. Skeleton in a standing posture as if dip fishing in water following the wading model, and in a swimming posture (based on Ibrahim et al., 2020a) following the pursuit predator model. A non-exhaustive set of lines of evidence as described in the text are indicated by arrows that either directly support either model (white arrow), are ambiguous or do not contradict the model (grey arrow), or actively contradict the model (black arrow). Key traits are as follows: A) laterally compressed skull, B) nares position, C) mechanical jaw performance, D) orbit position, E) neck stiffness and posture, F) non-hydrodynamic shape, G) instability in water, H) sub-anguilliform locomotion, I) thin caudal neural spines, J) tail propulsion, K) distal tail flexibility, L) low swimming efficiency, M) somewhat reduced hind limbs, N) enlarged 1st toe, O) pachyostosis, P) pneumatic elements, Q) forelimbs not reduced, R) neck ventriflexion, S) quadrate shape, T) head posture (as determined for Irritator), U) isotopic data from teeth, V) tooth enamel ridges, W) rostral sensory system. Skeleton modified from the original by Genya Masukawa (used with permission) and scaled to the size of the neotype. Scale bar is 1 m.
FIGURE 2. Principal Components Analysis of various measurements of the skull rescaled to skull length. Principal Component 1 (83.5% of variance) plotted against Principal Component 2 (13.7% of variance), plotted using eigenvalue scale. The red point is Spinosaurus, yellow are other spinosaurids, green are terrestrial taxa, pale blue are semi-aquatic, and dark blue, fully aquatic animals. Silhouettes are from PhyloPic.org and color-coordinated with the lines of the convex hulls for the groups of taxa they represent: the red Suchomimus (representing Spinosauridae; red Xs), the light green Allosaurus (representing non-spinosaurid Theropoda; open light green circles), and the orange Paleorhinus (representing phytosaurs: light brown pluses) are by Scott Hartman; blue Peloneustes (representing Plesiosauria: solid dark blue circles) by Nobu Tamura; dark green Varanus (representing terrestrial lepidosaurs: green asterisks) and dark brown Crocodylus (representing Crocodyliformes: dark brown pluses) by Steven Traver. Additional taxa plot include thallatosuchians (solid light blue circles), the mosasauroid Plotosaurus (blue asterisk), the nothosauroid Lariosaurus (solid aqua circle), and freshwater semi-aquatic lepidosaurs (open orange squares). The inset shows a reptile skull and how measurements were taken for the data used here and in Figure 3.
FIGURE 3. Graphs of various skull measurements to show the relationship between skull shape for different ecotypes. The red point is Spinosaurus, yellow are other spinosaurids, green are terrestrial taxa, pale blue are semi-aquatic and dark blue, fully aquatic animals. Least squares regressions are given for the terrestrial, semi-aquatic and aquatic datasets (the various spinosaurids were not included in these calculations), and the R2 values for these regressions are given.
FIGURE 4. Graph of theropod ungual curvature vs ungual length. The inset shows how the curvature of the unguals was measured. In lateral view a line AB is drawn between the ungual tip and the base. This is bisected by a perpendicular line until it contacts the ungual at point C. Lines are drawn from A to C and A to B and the internal angle measured. Unguals of Spinosaurus are in red, a further specimen attributed to a spinosaur is in yellow, and individual specimens are abbreviated as follows: Ab, abelisaurid; Ac, Acrocanthosaurus; Ai, Alioramus; Al, Allosaurus; Ca, Caudipteryx; Ce, ceratosaur; Co, Compsognathus; Di, Dilophosaurus; Ga, Gaulicho; Gg, Gigantoraptor; Gl, Gallimimus; Gu, Guanlong; Ha, Halszkaraptor; Ju, Juravenator; Ki, Kileskus; Li, Limusaurus; Mj, Majungasaurus; Sc, Spectrovenator; Sd, spinosaurid; Sn, Sinraptor; Sp, Spinosaurus; SB, Spinosaurus B; Tt, Tyrannotitan; Ty, Tyrannosaurus.
FIGURE 5. Comparison of skull shape of Spinosaurus and Baryonyx scaled to the same size. The two are very similar, which although this may be expected from their shared evolutionary history would suggest that they fundamentally forage in similar ways for similar prey, which contradicts the idea that one is an aquatic specialist. Not to scale.
FIGURE 6. Suggested head positions of Spinosaurus relative to water. A) when dipping the snout in the water to forage while leaving the naris above the waterline as per the wading model. Head angle of 45 degrees based on Schade et al. (2020) for Irritator. B) while lying submerged, keep the naris and orbit clear of the water while minimising the amount of head that is exposed as per Arden et al. (2019), C) fully submerged as if coming up for air and trying to expose only the nares to breathe. Scale bar equals 1 m.
FIGURE 7. A) Skull of a stork (Leptoptilos - scale bar is 100 mm) with a posteriorly retracted naris allowing them to forage while keeping the nares free of the water as in B) showing Ephipporhynchus senegalensis feeding. Although proportionally much further back here than in Spinosaurus, the absolute distance of the naris from the anterior tip of the jaw is less in the stork. C) Skull of crocodylian (Crocodylus - scale bar is 100 mm) with dorsally positioned naris allowing them to rest with minimal exposure of the head as in D) Crocodylus niloticus resting at the surface (image courtesy of Jonathan J. Meisenbach).
FIGURE 8. Depth of water required for Spinosaurus to avoid the considerable effects of wave drag. Even with the hind limbs lifted up, the animal is nearly 3 m in dorsoventral height so to avoid wave drag (fully submerged by over 3.5 m) the water would need to be close to 6 m in depth for Spinosaurus to swim efficiently. This is a minimum and the real value is likely to be higher (see text for details). Outline modified from Ibrahim et al. (2020a) and scale bar equals 1 m.
FIGURE 9. Line drawings of mid caudal vertebrae and chevrons of assorted reptiles (all in left lateral view) compared to A) Spinosaurus (Ibrahim et al., 2020a). Taxa with known or inferred signaling structures linked to their elongate neural spines (top row), B) Bagaceratops (Tereschenko and Singer, 2013), the sail-finned lizards C) Hydrosaurus* and D) Basiliscus* (courtesy of Jeroen Costeseque), E) the drepanosaur Drepanosaurus (redrawn from Sues, 2019), F) the chameleon Trioceros (courtesy of Steven Huskey), and those which show adaptations for aquatic locomotion (bottom row), G) a juvenile specimen of the crocodylian Tomistoma* (courtesy of Mathew Wedel), H) the phytosaur Mystriosuchus (Renesto and Lombardo, 1999) I) the mosasaur Mosasaurus (modified from Lindgren et al., 2011), J) the sea snake Pelamis (modified from Lindgren et al., 2011), and K) diapsid Hovasaurus* (redrawn from Sues, 2019). Scale bars are A) 200 mm, B) 20 mm, C) 10 mm, D) 10 mm, E) 20 mm, F) 10 mm, G), 20 mm, H) 20 mm, I) 100 mm, J) 2 mm K) 20 mm.
