Walking Sledgehammers

Scott Persons and Phil Currie made waves late last year with a study that showed everyone’s reconstructions of dinosaur tail anatomy was wrong. We, they said, had incorrectly measured the mass of the m. caudofemoralis longus, the muscle that runs from the mid-femur and along the transverse processes of the caudal vertebra, and as such had undersized the muscle and thus aspect of the tail. We’d drawn the tail way too thin, as shown here in W. Scott Person’s explanation at Dave Hone’s Archosaur Musings and Brian Switek’s Laelaps, but also Scott Hartman’s great review at Skeletal Drawing and Virginia Arbour’s interview with co-author Scott Persons at Psuedoplocephalus.

This paper reviewed the anatomy of crocs and found our dinosaurs wanting. They showed that the leg strength of theropod dinosaurs like Tyrannosaurus rex were massively stronger than estimated. But Tyrannosaurus has nothing on Carnotaurus sastrei, as Persons and Currie now report in PLoS ONE. I am not the first, or only person reporting on this. Take a look not only at Ed Yong’s Not Exactly Rocket Science, as well as Virginia Arbour’s follow-up interview with Scott Persons at Pseudoplocephalus.

Caudal musculature through a proximal section of the tail of Alligator mississippiensis, after Persons and Currie (2010).

The tails, we were told, had not only mis-estimated the masses, reconstructions from renowned artists ranging from Scott to the more famous Greg Paul has seemingly smoothed the margin between the projecting processes of the vertebrae. But, in alligators, these muscles and thus margins are not smooth: they bulge, and strongly. And almost universally, they project laterally further than the spines to which they attach. We’d also forgotten that the M. ilio-ischiocaudalis (pink, above) covers the M. caudofemoralis (deep red), and thus rendered our reconstructions “weak” in the hip. These muscles are important for locomotion:

The caudofemoralis is the primary femoral retractor, pulling the femur, and thus the entire leg with it, posteriorly. Depending on the insertion on the thigh, marked by the fourth trochanter in all archosaurs, or a trochanteric scar in birds, this muscle can also rotate the femur as it retracts it, and in birds at least helps pull the femur outward on its way through its downward arc. Alligators, like birds, have a sprawled femur, although the form of retraction differs largely due to the way the rest of the leg is positioned.

Femoral sprawl between an alligator (A) and a bird (B), from Hutchinson & Gatesy (2000).

As discussed extensively in papers by John Hutchinson and Steve Gatesy (e.g., 2000), femoral sprawl and its muscular relationships are influenced by the lower leg. The muscular action, shown above by the arrow, rotates the femur around its axis due in part to the insertion points of the thigh muscles. In typical theropods, the caudofemoral muscles insert on the posterior femur, and so the pull action of the muscle is almost completely posterior; in crocs and gators, the muscle insertion is on the medial femur, but because the femur sprawls outward, this pulls the femur inward and backward at the same time, and the lower leg follows largely in the same arc of movement, pushing the animal strongly to the sides; but in birds, while the femur sprawls, the insertion is still posterior, and so the femur is pulled around its axis while pulled posterior, and the highly flexed knee barely straightens. Much of this action tends to pull the tail to the body, rather than the leg to the tail, as in alligators, a subject I will get into much later.

This overview is interesting because another, later paper by John Hutchinson and co. (Hutchinson & Garcia, 2002) has assessed the biomechanics of the leg and body mass of Tyrannosaurus rex, and presumed that its bulk may have required it to be a slower, non-running animal than traditionally conceived. Rather than sprinting after prey, it would have trudged after it, and likely held its body in less of a leggy stance, but more of a crouch. Such a model is useful when keeping the body low and steady during the step cycle, rather than potentially tipping with its center of mass to far forward of its center of balance.

An additional element of the limb can be assumed as a proxy to a lower mediportal grade of locomotion, one which appears in medi- and graviportal birds, where the tibia is much shorter, the metatarsus more robust, the femur much longer relative to the leg. Cursorially proportioned animals have the inverse, with longer distal limb elements, and (except in mammals) longer mid-leg elements than proximal (femur). This thus implied that Tyrannosaurus rex, unlike juvenile tyrannosaurids, was a slower animal, and may never have actually ran once it achieved multi-ton weight.

Regardless of the dashing of our fantasies of super-fast tyrannosaurs, the speed demons of the Cretaceous seem to have merely relocated.

As Persons and Currie (2011) argue, the unusual tail vertebrae of abelisaurids allows for some interesting conclusions. First, strongly defined muscle scars on the undersides of the transverse processes (called “caudal ribs” in the paper) delineate the margins between M. caudofemoralis and M. ilio-ischiocaudalis.

MCF-PVPH-236, caudal vertebra of Aucasaurus garridoi (Coria et al., 2002).

Using these as markers, results in a very strongly developed caudofemoral muscles. Moreover, due to the elevation of the processes, the height of the muscle becomes extremely developed, although the muscles have simply broadly expanded the muscle laterally in keeping with the constraint the biology of the alligator and birds play in this region of the body (Persons & Currie, 2010). In birds, at least, this muscle can expand lateral to the processes, as it must to reach the femur. This results in a very wide and deep tail:

Section through tail of Carnotaurus sastrei, from Persons & Currie (2011).

Of course, you must be wondering why I brought up that bit about Tyrannosaurus rex (aside from the previous work by Persons and Currie).

First, there’s Majungasaurus crenatissimus:

Skeleton of Majungasaurus crenatissimus. Based mostly on FMNH PR2100.

One will notice the extremely short legs. The body is long, low-slung, and there are the huge knees. How do these work together? The long femur and the large cnemial process of the tibia work to increase the size of the femoral musculature, and while the tail was not as robustly muscled as in Carnotaurus sastrei, in Majungasaurus crenatissimus there were nonetheless substantial mass in the leg to contend with. These include the M. iliotibialis and M. iliofibularis, the latter which inserted on the large process of the fibula below the “knee” of the cnemial process. As in virtually all abelisaurids, this process is large, and in some it is positively gigantic (a tibia included in the holotype series of Lametasaurus indicus [Matley, 1923] includes a cnemial prcoess so large the anteroposterior length of the bone is almost 1/3 the proximodistal length, and almost certainly pertains to an abelisaurid [Chakravarti, 1935 argued, and Walker, 1964 agreed, that the tibia be excluded from the holotype, which was inferred to be an ankylosaur; the tibia is thus not considered part of Lametasaurus indicus, based on large osteoderms; Rajasaurus narmadensis Wilson et al., 2003, appears to be a suitable container for the material, although this is tentative]).

Tibiae of Indian abelisaurids. A, tibia from type series of Lametasaurus indicus (Matley, 1923); B, GSI 21141/29, tibia referred to Rajasaurus narmadensis (Wilson et al., 2003). A is after Chakravarti (1935), albeit a poor scan copy of the original plate.

Such a giant knee is similar to birds, which always keep the knee strongly flexed. In my skeletal reconstruction above, this emphasizes an unusually low body posture, one in which a speed leg seems unreasonable.

Enter Carnotaurus sastrei. The holotype includes an ilium, a femur, and the proximal portion of the tibia. The rest of the leg is unknown.

Skeleton of Carnotaurus sastrei Bonaparte, 1986.

Proportionately, the femur is longer relative to the ilium, and the knee seems smaller, although the cnemial process is missing. I’ve reconstructed the tibia to be equal to the femoral length, and over-estimated the metatarsus and thus pes to more cursorial proportions, but based on other abelisaurids, I wonder. This leads us to Aucasaurus garridoi which, sadly, I’ve not done a skeletal reconstruction of, but others have:

Skeleton of Aucasaurus garridoi, from Scott Hartman.

If we were proportioning the limbs correctly to a close relative, Aucasaurus seems a closer fit for Carnotaurus than Majungasaurus. Here, the body is shorter, the legs “leggier”, and the animal more slender and bird-like in general form. We’d not get the bizarre look so easily, and instead a “standard” looking long-legged slender theropod.

Persons and Currie argue that the strong femoral retractors are heavily related to locomotion, a feature most biomechanics would agree with. Carnotaurus, they argue, was a swift-legged animal, using its powerful leg retractors and robustly muscled thighs to power large chicken legs into action. The question then, is “Why?” The authors argue that, in comparison to the large carcharodontosaurids among which these smaller abelisaurids are found, niche partitioning may have required different prey selection and behavior; carcharodontosaurids would be adapted to preying upon the large sauropods, giants like Antarctosaurus and Argentinosaurus, while the smaller abelisaurids would take smaller, swifter-footed prey such as the ornithopods Talenkauen or Anabisetia.

I am curious about this, though. The crouch model, instead, might imply some functional differences. First, it means the muscle moment arms and masses might change effective leg strength. Second, FEA analysis (Mazzetta et al., 2009) and mechanical analysis of the head and neck of Carnotaurus sastrei (Mazzetta et al., 1998) has been used to argue that this animal’s head was adapted to high-speed velocity impacts, much akin to the use of a sledgehammer with teeth at the end. Rather than the use of large robust impact-resistant crowns as in tyrannosaurs, carnotaurine abelisaurids have thinner, lower crowned teeth and massively constructed skulls. While the jaws seem less adaptive to high-impact biting, this doesn’t seem to prevent the upper skull from being used to ram into and downward onto prey.

Lunch vs. Carnotaurus

If anything, this emphasizes the acquisition of smaller prey, which can be assaulted from above rather than the side, or where the head could be elevated backwards in order to slam forward and downward, the peen to the sledge, while the neck would be stiff. This is not too dissimilar to the model used to describe the tails of ankylosaurids, although in this case instead of a side-to-side swing, the sledge is up-to-down in an “overhand” swing.

The massive M. caudofemoralis longus, running along the underside of the tail, is a powerful muscle. It extends the length of over half the tail in most archosaurs, and pulls the tail around to counterbalance the walk-step during undulation of the spine. But it serves its primary purpose to pull the femur backward, causing the body to “jump” forward. The higher the position of the origins on the caudal transverse processes, the longer the moment arm around the femoral hip joint, and thus the greater leverage. When the femur is oriented further forward in a “crouch” posture a la Groucho Marx, this further increases the moment arm of the thigh, as well as the length of the muscle itself.

Muscles of the thigh and their relative sizes and relationships between the "high-step" and "crouch" models. The latter stance is 92% the height of the higher step.

The biomechanics of the “lunge” or propulsion during the crouch have only been modeled so far for tyrannosaurs (Hutchinson & Garcia, 2002), where it is useful to save energy and increase speed without running. Given the extremely large knees of abelisaurids, the leg-straightening muscles of the upper thigh would increase the propulsive speed of the leg, an advantage seemingly more suited to lunging than for pursuit running. In this, I tend to think the biology compares well with ambush predators, rather than pursuit runners: more lion, less cheetah. Despite their smaller size and relatively longer legs, then, Carnotaurus and kin may not have been much different from the large carcharodontosaurids, although I am certainly massively estimating things here.

It may be that the extreme legginess of Carnotaurus permitted it to be a pursuit predator, advantaging hunting the smaller ornithopods of the floodplains and forests of Patagonia, while its even shorter-legged kin Majungasaurus would have preferred ambush predation on Rapetosaurus and larger prey in the Maeverano forests of Madagascar.

Carnotaurus by Lida Xing and Yi Liu, from Persons and Currie (2011).

Theropods may be much more massive than we expect, with slimmed-down muscle masses and thus body masses resulting in too-svelte theropods, a motif that has been popular in the last few decades by the ultra-slim theropods of Greg Paul or the modeling required to be conservative on mass rather than generous due. Another paper by Hutchinson and co., (Hutchinson et al., 2011), has revised the body mass of Tyrannosaurus rex using several models and a proposed growth trajectory between two specimens (“Sue” and “Jane,” one quite larger than the other), saying:

What body mass estimate methods are most reliable?
Our methods again raise this question that has been recurring in recent studies. Even with consideration afforded to artefacts of reconstruction and investigator biases in fleshing out skeletons (above), it is clear that even minimal body mass values for Tyrannosaurus rex estimated directly from skeletons tend to be larger than published values extrapolated from scale models or limb bone dimensions[. … R]esearchers have tended to favour ‘skinnier’ reconstructions in the production of scale models, with scaling perhaps magnifying the effects of such subjective choices.

[Hutchinson et al., 2011:pg.13]

This study emphasizes the issues inherent in mass estimation, but also the quandary involved in developing metrics and methods “off the cuff.” Despite this, I think all three of these works can be constructed into a whole, a model where leggy theropods with heavy mass and certain morphological features may be useful in emphasizing slower animals, in favor of power as in ambushers (dinosaur lions), rather than swifter, “dancing queens” (dinosaur cheetahs).

Chakravarti, D. K. 1935. Is Lametasaurus indicus an armored dinosaur? American Journal of Science 30(5):138-141.
Hutchinson, J., Bates, K., Molnar, J., Allen, V. & Makovicky, P. 2011. A computational analysis of limb and body dimensions in Tyrannosaurus rex with implications for locomotion, ontogeny, and growth. PLoS ONE 6(10):e26037.
Hutchinson, J. R. & Garcia, M. 2002. Tyrannosaurus was not a fast runner. Nature 415:1018-1021.
Hutchinson, J. R. & Gatesy, S. M. 2000. Adductors, abductors, and the evolution of archosaur locomotion. Paleobiology 26(4):734-751.
Matley, C. A. 1923. Note on an armoured dinosaur from the Lameta beds of Jubbulpore. Records of the Geological Survey of India 55:105-109.
Mazzetta, G. V., Cisilino, A. P., Blanco, r. E. & Calvo, N. 2009. Cranial mechanics and functional interpretation of the horned carnivorous dinosaur Carnotaurus sastrei. Journal of Vertebrate Paleontology 29(3):822-830.
Mazzetta, G. V., Fariña, R. A. & Vizcaíno, S. F. 1998. On the palaeobiology of the South American horned theropod Carnotaurus sastrei Bonaparte. in Pérez-Moreno, Holtz, Sanz & Moratalla (eds.) Aspects of Theropod Paleobiology. Gaia 15:185-192.
Persons, W. S., III & Currie, P. J. 2010. The tail of Tyrannosaurus: Reassessing the size and locomotive importance of the M. caudofemoralis in non-avian theropods. The Anatomical Record 294:119-131.
Persons, W. S., III & Currie, P. J. 2011. Dinosaur speed demon: The caudal musculature of Carnotaurus sastrei and implications for the evolution of South American abelisaurids. PLoS ONE 6(10):e25763.
Walker, A. 1964. Triassic reptiles from the Elgin area: Ornithosuchus and the origin of carnosaurs. Philosophical Transactions of the Royal Society of London B 248:53-134.