On theropod like mammals
The concept of theropod-like mammals is a fairly controversial one in speculative evolution discourse and one I weighted in very negatively in the past. Some years of retrospect and nature humbling me and I’ve arrived at a more neutral position: they are possible, but only under specific circumstances.
Why few mammals are bipedal?
Bipedalism evolved multiple times among reptiles (particularly archosaurs, but also some lepidosaurs and bolosaurids) but only a select amount of times among mammals. This is due to the nature of their spines; for most reptiles, the vertebral column is inflexible vertically but flexible laterally, providing a torso less efficient for galloping but stable enough to remain stiff during bipedal locomotion. The opposite happens in mammals, which make galloping far more appealing than bipedal running.
This differences in spine anatomy seem to have evolved very early on (Jones 2021) and likely reflect the development of the diaphragm in mammals. In general, mammal spines are very weird, being forced into specific modules other amniotes do not have (Irma Varela-Lasheras 2011), which may further impact performance in bipedality.
Many mammals like kangaroos, jerboas and springhares have evolved a hopping locomotion, a derivation of galloping that relies only on the hindlimbs. However, they are nonetheless obligate quadrupeds at low speeds, so they are seldomly considered true bipeds and certainly not bipedal walkers like dinosaurs.
Most large mammals also have rather thin tails that cannot balance the torso. The center of gravity in mammals synapsids is located forward in the body, which may also explain some evolutionary trends like more erect forelimbs in several groups (eutriconodonts, pyrotheres, chalicothere, et cetera)
How to break the mold?
Nonetheless, a few mammals found ways to bypass these limitations. Humans are the more obvious example, having developed a fully erect stance so there’s no need to balance the torso. But as you can imagine this is not the theropod-like visage speculative evolution projects are looking for.
Pangolins are the most notable example of mammals walking in a theropod-like manner. Allegedly tamanduas are also capable of facultative bipedality as were some extinct ground sloths (Jennifer L. White 1993). All these animals are/were able to do so due to A) massive tails, B) specialised forelimbs whose large claws hinder quadrupedal locomotion, C) short torsos with rigid spines in the case of xenarthrans. None of these animals are obligate bipeds but pangolins at least can get pretty fast:
https://www.youtube.com/watch?v=B95NdS77fZM
Still, being rather specialised toothless insectivores or obligate herbivores in the case of sloths, its unlikely that a mammalian T. rex could evolve from forms like these.
A more promising prospect comes from the hopping mammals. Sthenurine kangaroos are now thought to have been bipedal walkers (Janis 2014), having evolved from hopping ancestors that grew too large to hop. Facultative bipedality can still be seen among modern jerboas, so its easy to see how hopping can transition to bipedal walking. Sthenurines nonetheless had to make several concessions similar to those of anteaters and pangolins: short, robust torsos (in their case so much so that they were incapable of reaching down to the ground and use their forelimbs to walk, thus becoming true bipeds) and heavy tails to balance them. Though fossil footprints seem to suggest that they raised their tails like bipedal dinosaurs (SVP 2021 abstract), they were still slow moving, specialised herbivores like ground sloths.
The closest thing to a mammalian theropod (in terms of being fast running and predatory) per se might have been Leptictidium. There’s still a fierce debate if it was truly a biped or just a hopper and this discussion doesn’t seem to be dying down soon, though it wouldn’t be surprising if it evolved from hopping ancestors like sthenurines.
The lesson to take away
As we’ve seen so far, mammalian theropod-like bipedalism seems to come mostly from two sources: the development of an anteater-like lifestyle and hopping. Most examples are specialised, slow moving animals, but Leptictidium and some pangolins may suggest that cursorial ecologies are possible.
Find a situation in which a large tail can balance the torso and quadrupedal locomotion becomes less accessible than in most mammals. For now I’ve conceived of a few hoppers-turned-bipeds, an otter-like form that developed a large tail and hindlimbs and returned to a terrestrial ecology and a form in which the tail enlarge to provide fat reserves, making facultative bipedalism possible.
refs
Jones, K. E.; Dickson, B. V.; Angielczyk, K. D.; Pierce, S. E. (2021). “Adaptive landscapes challenge the “lateral-to-sagittal” paradigm for mammalian vertebral evolution”. Current Biology. 31 (9): 1883–1892.e7. doi:10.1016/j.cub.2021.02.009. PMID 33657406. S2CID 232093918.
Irma Varela-Lasheras, Alexander J Bakker, Steven D van der Mije, Johan AJ Metz, Joris van Alphen and Frietson Galis. Breaking evolutionary and pleiotropic constraints in mammals. On sloths, manatees and homeotic mutations. EvoDevo, 2011
Jennifer L. White, Indicators of Locomotor Habits in Xenarthrans: Evidence for Locomotor Heterogeneity among Fossil Sloths, Journal of Vertebrate Paleontology Vol. 13, №2 (Jun. 8, 1993), pp. 230–242 (13 pages) Published By: Taylor & Francis, Ltd.
Janis, CM; Buttrill, K; Figueirido, B (2014). “Locomotion in Extinct Giant Kangaroos: Were Sthenurines Hop-Less Monsters?”. PLOS ONE. 9 (10): e109888. doi:10.1371/journal.pone.0109888. PMC 4198187. PMID 25333823.
