Scelidosaurus

Norman, David B, 2021, Scelidosaurus harrisonii (Dinosauria: Ornithischia) from the Early Jurassic of Dorset, England: biology and phylogenetic relationships, Zoological Journal of the Linnean Society 191 (1), pp. 1-86 : 52-54

publication ID

https://doi.org/ 10.1093/zoolinnean/zlaa061

persistent identifier

https://treatment.plazi.org/id/B66BDD2A-081C-FFA3-E09C-756FFA28E450

treatment provided by

Felipe

scientific name

Scelidosaurus
status

 

Scelidosaurus hindlimb structure and pose

The pelvis and hindlimb of Scelidosaurus are now better understood ( Norman, 2020b) and can be more readily subjected to Carrano’s approach to limb structure and functionality. The ilium has a long, laterally deflected preacetabular process and (contrary to previous views) a deep, rectangular postacetabular process and a brevis shelf. The acetabulum forms a partial cupola with a complete, curtain-like medial wall and a supra-acetabular crest with a horizontal rim (unlike the condition in dinosauromorphs and some basal dinosaurs).

The femur has an articular head that is medially offset from the shaft and not deflected anteromedially (as described to be the condition in dinosauromorphs and some basal dinosaurs – Carrano, 2000). The anterior trochanter is proximally placed and prominently positioned on the anterior margin of the greater trochanter. The implication from this configuration is that the femur could be swung parasagittally and was powered by musculature associated with the expanded iliac blade. The femoral shaft is slightly bowed (more so in juvenile individuals – see Fig. 32 View Figure 32 ) and is notable for the presence of a prominent pendent 4 th trochanter. The medium- to large-sized (~ 32 cm long) crushed femur of Scelidosaurus (NHMUK OR41322 – Fig. 34 View Figure 34 ) has a prominent pendent 4 th trochanter. However, on closer inspection this structure is noticeably thicker and considerably more robust than that of NHMUK R6704 ( Fig. 32A View Figure 32 ). The 4 th trochanter of the larger femur is thick, yet an outline of the slender ‘juvenile’ 4 th trochanter morphology remains visible on its surface ( Fig. 34 View Figure 34 , 4 View Figure 4 tr). The original structure has, in effect, been shrouded by a thickness of metaplastic bone ( Fig. 34 View Figure 34 , mpb). It is probable that this secondary feature expanded and strengthened the anchoring of the caudifemoral tendons and reinforced the 4 th trochanter.

A pendent 4 th trochanter is consistently associated with ornithischians that, judged by their posture and limb proportions, were likely to have been cursorial (e.g. Maidment & Barrett, 2014 and references therein). The distal femoral condyles are laterally expanded, but show no clear evidence of an intercondylar extensor groove (but this is also a feature of the femur of the allegedly cursorial ornithischian Hypsilophodon – Galton, 1971, 1974).

Thick cartilages would have been present that capped the bones at the knee-joint, and the structure of the preserved bones does not accurately reflect the extent to which this joint operated as a uni-axial hinge. The crus (shin) comprises a structurally dominant straight tibia and a shorter, but stout and bowed, fibula; this gives the impression that a limited amount of axial torsion might have been possible between the bones of the lower leg.

The tarsus is conventionally mesotarsal but the evidence, judged by what is currently known of the structure of the astragalocalcaneal roller surface, is that the ankle joint was not so strongly constrained to rotate in a transverse uni-axial plane because it does not display a deeply grooved (trochlear) joint surface. The distal tarsals, as preserved, also present an asymmetrical arrangement with the central (dT3) and lateral (dT4) tarsals, forming well-ossified articular pads, whereas the medial tarsal (dT2), if present, was probably an unossified fibrocartilage pad or (less likely) entirely absent from the ankle joint ( Norman, 2020b: figs 86, 87). The pattern of just two distal ankle bones (central and lateral) is found reasonably consistently among more derived ornithischians (in which this anatomy has been preserved, e.g. Norman 1980, 1986). The asymmetrical construction of this ankle joint would have left it potentially susceptible to (or able to accommodate) longaxis torsion during the limb excursion cycle.

In the context of ankle-joint construction and joint mobility, it is interesting to note that the early (Hettangian) but unusually specialized ornithischian Heterodontosaurus has been reported as having three well-ossified distal tarsals (Santa Luca, 1980) or just two ( Sereno, 2012). In either interpretation, the distal tarsal structure that is preserved forms an ossified articular pad that caps all three weight-bearing metatarsals (2–4) that support functional pedal digits. In turn, this tarsal structure articulated with a hinge-like and fused tibiotarsus. This ankle structure suggests that limb excursion was parasagittal and that ankle-joint mobility was restricted to a single transverse plane of rotation.

The pes of Scelidosaurus is functionally tridactyl, being dominated by pedal digits 2–4 ( Norman, 2020b: fig. 90). Metatarsals 2–4 are sutured proximally and form a splayed structure distally. Metatarsal 1 is short, but supports two phalanges, the terminal one of which forms a small, pointed claw. Metatarsal 5 is represented by a splint bone. The three functional toes, as preserved in the lectotype (NHMUK R1111 – the only currently known example that includes articulated feet), curve medially along their lengths and the terminal unguals are pointed, slightly arched, but, as with the phalanges, they are twisted medially along their lengths ( Norman, 2020b: fig. 95).

Scelidosaurus pelvis and hindlimb: comparative comments

The pelvis of Scelidosaurus ( Fig. 31A View Figure 31 ) bears a much closer structural similarity to that of early ornithischians and euornithopods than to those of more derived stegosaurian and ankylosaurian thyreophorans ( Fig. 31B, C View Figure 31 ; Carrano, 2000: figs 6, 7). The ilium has a long, arched (in mature individuals) dorsoventrally flattened preacetabular process, but it is not broadly expanded or downcurved to the extent seen in ankylosaurs and stegosaurs. The dorsal part of the iliac blade, though out-turned when articulated with the sacrum, is not laterally flared, and its postacetabular blade is rectangular and has a modest brevis shelf. In both stegosaurs and ankylosaurs ( Fig. 31B, C View Figure 31 ), the postacetabular blade is short and tilted horizontally. The ischium of Scelidosaurus is a Y-shaped thickened blade with a long stem. In stegosaurs, the ischium is more transversely compressed and tapers distally ( Maidment et al., 2015), whereas those of ankylosaurs are considerably shorter, bar-shaped and decurved ( Coombs, 1978a). The pubis of Scelidosaurus has a long, slender pubic shaft and a blade-like, laterally deflected prepubic process. The pubis of ankylosaurs is diminutive (a small oblong block fused to the ilium) from which projects a short, finger-shaped pubic shaft ( Coombs, 1978a). In stegosaurs the pubis is a large, obtusely V-shaped bone with a parallel-sided pubic shaft that was ligamentously bound to the ischium, and a long rectangular prepubic blade ( Gilmore, 1914).

Stegosaurs and ankylosaurs have much more robust and straight (pillar-like) femora ( Gilmore, 1914; Coombs, 1978a). The femoral head is less clearly medially offset on the shaft (this is particularly so in ankylosaurid ankylosaurs, where the femoral head is terminal – Coombs, 1978a). The anterior trochanter tends to become indistinguishably fused to the greater trochanter. The 4 th trochanter is represented by either a low mound or a large, depressed muscle scar on the lower half of the femoral shaft. Their crural (shin) bones of are straight and the tibia is massive, with greatly expanded proximal and distal ends; the fibula is a comparatively slender, straight bone that had little to do with structural support. The ankle comprises a single, proximal tarsal that is reduced to a flattened, warped plate formed by the astragalus, which is fused to the distal surface of the tibia. Distal tarsals have not, so far, been reported. The pes comprises short, dumbbell-shaped metatarsals and ‘stubby’ toes; these form a divergent ‘spreading’ arrangement that was most likely supported by a plantar pad of elasticated fibrous tissue. Their unguals are flattened, broad and rounded, distally forming a hoof, rather than a claw.

FUNCTIONALITY OF THE HINDLIMB AND PES SUMMARIZED

1. Acetabulum. The acetabular joint surface is unusually large compared to the femoral articular head. It offers the possibility of a wide range of femoral head positions and consequential femoral excursions ( Norman, 2020b: fig. 98) but the extent to which soft tissues (notably cartilages) influenced the potential range of femoral excursions is uncertain ( Hutson & Hutson, 2012).

2. Hindlimb joints.The knee has some of the osteological attributes of a simple hinge, but the extent to which this was constrained by soft tissues to a transverse uni-axial plane of rotation is unknowable. The structure of the lower leg, which retains a short but stout, bowed fibula, indicates that limited (modest) axial torsion may have occurred between the tibia and fibula. The ankle joint is weakly trochlear, as well as being osteologically asymmetrical. If there was a fibrocartilage pad covering the medial portion of the ankle, the entire ankle joint may have been susceptible to (or able to accommodate) long-axis torsion.

3. Hindlimb retractors. The lines of action of the caudifemoral muscles can be reconstructed (compare Figs 33 View Figure 33 , 35 View Figure 35 ). The origin of the m. caudifemoralis brevis lies on the underside of the postacetabular process of the ilium and runs anteroventrally to the base of the 4 th trochanter and its line of action is posteromedial ( Fig. 35A View Figure 35 , cfb). Musculus caudifemoralis longus is known to originate from the lateral surfaces of the caudal centra and the edges of the caudal ribs in all sauropsids (and even the equivalent pygostyle in birds) and inserts on the 4 th trochanter. The line of action ( Fig. 35A View Figure 35 , cfl) of this retractor muscle runs anterolaterally from close to the caudal midline. In addition, the adductor muscle ( Fig. 35 View Figure 35 , add) would both retract the femoral shaft (from its fully protracted state) and rotate the femoral shaft outward along its longaxis (balancing some of the inward rotation imposed by the far more powerful caudifemoral muscles).

4. The pendent 4 th trochanter is positioned on the posteromedial surface of the femoral mid-shaft ( Fig. 35C View Figure 35 ), increasing the axial rotation generated by the femoral retractor muscles.

5. Hindlimb protractors. The lines of action of the mm. iliotibialis, iliofemoralis and puboischiofemoralis internus can be reconstructed with varying degrees of confidence. Musculus iliotibialis is consistently found to originate along the dorsal edge of the iliac blade in sauropsids (including birds) and inserts on the cnemial crest of the tibia ( Fig. 33 View Figure 33 , it). Musculus iliofemoralis originates on the dorsolateral surface of the iliac blade and inserts variously on the shaft of the femur and (with personal equivocation) on the anterior trochanter ( Fig. 33 View Figure 33 , if) and lateral surface of the greater trochanter. At least one significant portion of the puboischiofemoralis complex originates from the proximomedial surface of the pubis, the surfaces of the posterior dorsal centra and the first sacral rib (in sauropsids). It has been suggested that a slip of this muscle attached to the ventral surface of the preacetabular process in ornithischians ( Maidment & Barrett, 2011 – see also Fig. 33 View Figure 33 , pifi). This muscle inserts on the medial surface of the proximal femoral shaft and a major tendon of this muscle inserts adjacent to the 4 th trochanter in crocodiles (there is a wellpreserved muscle scar in this position on the femur of the lectotype – Fig. 32B View Figure 32 , pifi). The lines of action of the first two muscles can be reconstructed with some confidence and vary a little along the length of the iliac blade because the preacetabular process swings markedly laterally along its length. Musculus puboischiofemoralis has a line of action that not only serves to protract the femur, but imposes a torsional force that rotates the femoral shaft outward (laterally) in preparation for the next stride ( Fig. 35C, E View Figure 35 ).

6. Dimensions of the abdominal cavity. Scelidosaurus was herbivorous and, judged from its teeth and jaw action, would have pulped and partly sheared its food prior to swallowing. This style of oral processing is unlikely to have prepared plant matter for immediate absorption in the intestine, so it would need to be processed further in the gut. A gizzard, lined with gastroliths, cannot be dismissed simply because no gastroliths have been recorded with any of the skeletons of this animal that have been discovered to date. A gizzard is present in living crocodiles and birds, so may well have been present and used to comminute plant food in the anterior gut. However, further enzyme-mediated digestion would be necessary to release plant cell contents and can only have been achieved within gut caecae that housed permanent populations of microbes capable of secreting enzymes that can break down the tough polysaccharide (cellulose-based) cell walls of plants. Gut enlargement to accommodate caecae requires a capacious abdomen. The dorsal rib cage is indisputably broad ( Norman, 2020b: fig. 34; see Fig. 35D, E View Figure 35 ), and the body shape is more similar to that seen in ankylosaurs (e.g. Gaston et al., 2001) than stegosaurs (see also Fig. 22 View Figure 22 ), which suggests that Scelidosaurus had a barrel-shaped abdomen.

7. Deflection of the prepubic blades. The lateral deflection of the prepubic blades that is evident in Scelidosaurus is also suggestive of the physical accommodation of a broadly expanded abdomen.

8. Asymmetry of the pes. The medial curvature that is evident when the digits of the pes are articulated might be an artefact of preservation in this specimen. However, the ungual phalanges of digits 2–4 are well-preserved and show comparatively little in the way of crushing plastic distortion, and yet they all indicate clear medial curvature. In the tridactyl pes of a ‘normal’ parasagittally gaited dinosaur, the axis of support tends to run through digit 3, which displays bilateral symmetry. The digits on either side splay: digit 2 curves medially, whereas digit 4 curves laterally, so that the foot as a whole exhibits some degree of symmetry on either side of the principal axis of support (e.g. Norman, 1980: fig. 71; 1986: fig. 63). The persistent asymmetry evident across the digits of the pes of Scelidosaurus implies that torsional forces in the long-axis of the hindlimb were acting on the foot during the supportretraction phase of the locomotor cycle ( Fig. 35E View Figure 35 ).

Darwin Core Archive (for parent article) View in SIBiLS Plain XML RDF