Why bipedal?

[Infinity Blade]

A few of you (though I admit that's next to no one who's still active on this board) might remember a guy named Vobby who posted on Carnivora forum a whole decade ago. I remember one time he made a whole thread on the dinosaur subsection asking why theropods evolved a bipedal stance. Well, this is basically our version of that, but this time I'm going to offer up some ideas to help answer the question, namely by listing advantages and disadvantages that this may have brought about.

Tell me what you think.

Pros:

• Superior reach/feeding envelope from standing point compared to most quadrupeds
• Less physical effort to employ forelimbs for certain tasks (i.e. no need to get weight-bearing limbs off the ground and rear up)
• Forelimb and hindlimb function can be completely segregated (e.g. fossorial forelimbs+cursorial hindlimbs, powered flight wings+perching/walking hindlimbs)
• Reduction of forelimbs may allow for stronger neck muscle attachments on pectoral girdle (not as possible for quadrupeds for obvious reasons)

Cons:

• Less stability all else being equal
• Greater impact to terrestrial locomotion if one leg is impaired
• Probably cannot get quite as fast as the fastest quadrupeds (relying solely on limb length for increased stride, instead of both limb length and vertebral flexion in some quadrupeds)
• Inferior turning ability for whole body

[theropod]

I’ll take the opportunity loosely transfer some material I posted in a recent discussion on Broly’s discord here.

RE: Effect of different tail and body postures on rotational inertia of bipedal dinosaurs

(model is Plateosaurus MSF 15.8B.)

Baseline (green) model 0 has the following mass and inertia properties (the closest approximation is I_elliptical_cs, this is the model that assumes mass distribution as in an elliptical body cross-section, which is the same assumption the mass is based on. As you can see, however, the impact is fairly minor. Due to the elongate body shape the anteroposterior distance from the COM is by far the dominant factor, and the transverse mass distribution does not have a large effect.):
     total_mass   I_point_masses  I_elliptical_cs I_rectangular_cs
      40.256774         4.748666         4.851946         4.886372
Mass in kg, rotational inertia in kg*m². Pneumaticity and limbs not included.

Model 1 (blue): Tail lifted, reduces↓ rotational inertia by ca 16% relative to baseline:

     total_mass   I_point_masses  I_elliptical_cs I_rectangular_cs
      40.775262         3.946151         4.053422         4.089178


Model 2 (red): Tail and torso+neck lifted, reduces↓ rotational inertia by ca. 20% relative to baseline:

     total_mass   I_point_masses  I_elliptical_cs I_rectangular_cs
      41.342844         3.757243         3.863450         3.898852


So a change in posture (similar to postures some have speculated on before) could undoubtedly reduce rotational inertia considerably, and therefore improve turning ability.

One thing I’ve been thinking about is how an (at least intermittent) posture with the head and neck pulled back, and possibly the trunk in a more upright position (similar to model 2), might affect interpretations on forelimb function. Generally people often express strong opinions about the low range of motion of theropod forelimbs at the glenoid, which in neutral posture would mean they couldn’t grasp anywhere near as far forward as the jaws in the majority of theropods. There was a recent debate on the dinosaur mailing list, in which someone argued quite vehemently that theropod forelimbs in general had no significant role in prey acquisition (yes, even the long and highly raptorially adapted ones of Allosaurus) because of the perceived awkwardness of the poses this would require. I disagree with many of their points, but one thing to highlight with regard to the specific topic at hand here:

If the neck was pulled back and upwards, and the torso flexed upwards from the hip, that would have the effect of both increasing the anterior reach of the forelimbs (as it would change the orientation of the scapulocoracoid), and moving the head to a more disadvantageous position, to such a degree that in many theropods the forelimbs would actually be closer to the prey (at least small to mid-sized prey items) than the jaws. No doubt the jaws could still have struck forward quite quickly, especially in theropods where the skulls are quite lightly constructed, but probably they would have still been in a worse position to apprehend small- to mid-sized prey in many realistic scenarios, since bringing the torso down (and back up) would require a much larger, slower movement.

If theropods actually did adopt such a posture to reduce inertia when fast turns were required (like when chasing after prey), this might affect a lot of the arguments made about the forelimbs’ role in predation. It also might explain why the forelimbs seem to have had such a (perceived) limited anterior range of motion while having such an extensive posterior range of motion (as they would likely be pulled back during maneuvering to minimize inertia, and be brought forward when needed, without requiring a change to the position of the torso)

[Infinity Blade]

I saw you had an earlier point about feeding envelope (although I see you edited it out). That was something I sort thought about for a while. If you're 12 meters long, and the first 6 meters of that is your head, neck, and torso, your business end is already past a certain point when it's time to turn. Really good if you're trying to catch something, or even reach out past some deadly prey item's weapons and to its flanks.

Maybe not quite as useful when a mountain-sized rock slams into the Earth and you have to dodge meteorites raining down->?

[Supercommunist]

For some animals, bipedalism could have also originated as an intimidation/bluffing display. An standing its hind legs looks larger.

[theropod]

Oct 11, 2024 2:39:58 GMT 5 Infinity Blade said:

I saw you had an earlier point about feeding envelope (although I see you edited it out). That was something I sort thought about for a while. If you're 12 meters long, and the first 6 meters of that is your head, neck, and torso, your business end is already past a certain point when it's time to turn. Really good if you're trying to catch something, or even reach out past some deadly prey item's weapons and to its flanks.

Maybe not quite as useful when a mountain-sized rock slams into the Earth and you have to dodge meteorites raining down->?

Yeah I edited it out because I noticed you had already mentioned it in your first post. I definitely do agree. Being longer means it will take you longer to turn a certain angle, but in exchange for that the distance covered by turning a certain angle will be greater.

Now admittedly, being longer will increase your rotational inertia by length², but only increase the distance covered by your extreme points by length¹. So the greater length won’t offset the higher rotational inertia in terms of simple turning speed. But we also have to think in terms of area and even volume (the ground area the jaws can cover, and the volume of space that they can reach for any given range of angles covered by the whole body axis) to some extent. This is where (lateral) neck flexibility will play a huge role in terms of determining how far inside that feeding envelope the animal can actually reach, and whatever use the animal can get out of its forelimbs too (maybe this is why the forelimbs are small in most giant theropods, because their prey items were all so large that them evading the jaws is less of an option?).

Oct 11, 2024 2:44:46 GMT 5 Supercommunist said:

For some animals, bipedalism could have also originated as an intimidation/bluffing display. An standing its hind legs looks larger.

That is actually a really good point, although I’d make a slight modification; I think it would play a bigger role in why bipedalism was maintained in animals that were already bipedal (for example in large theropods), but where some might ask the legitimate question why they did not revert back to a quadrupedal stance.

Bipedalism arose in fairly small animals (where it would seem that it has a more favorable benefit to penalty ratio in terms of mobility), and got repeatedly lost or reduced in large animals (ceratopsians, sauropodomorphs, thyreophorans, even ornithopods at least to an extent), but the evolutionary impact of threat display in lineages that are already evolving gigantism is likely at least as great or greater (given that it might play a role in why those lineages are gigantic in the first place).



To elaborate on why I think bipedalism is more advantageous for mobility at smaller sizes, let me loosely quote myself again:

––
Presumably there is some natural limit to how fast even quadrupeds can turn that does not depend on their inertia/torque ratio, but rather on how fast their limbs can keep up with the rotation. But it depends on the size of the animal in question. If the inertia is very low, the torque might be sufficient to turn it more quickly than the limbs can actually move.

If you have a very long lever, you can theoretically generate a lot of torque, therefore you could impart a lot of rotational acceleration. But in reality, how fast you can accelerate something will be limited by how fast you can move the other end of the lever.
That’s why, when we throw something relatively light like a small rock or a spear, we get better results with a sling or atlatl that decreases our torque but increases our out-lever length and thus the speed; the speed of our throw is limited by how fast we can move our own body (due to our body geometry and the the limits of muscle contraction speed), not how much force we can exert, so in this case we actually benefit from shortening our end of the lever (in-lever) relative to the other end (out-lever) so that less motion (but more force) on our end results in more motion (but less force) on the other end.
However, this only works for light objects whose speed is not limited by how much force or torque we can produce to overcome their inertia (this is one reason why there is an optimum weight for projectiles to achieve the maximum range for any given throwing mechanism; if you shoot an extremely light arrow on a very heavy bow, the bow will simply not move fast enough to transfer as much of its energy to the projectile as if you shot a heavier arrow, so the efficiency is lower.

The ideal mechanical advantage for accelerating something depends on how much inertia there is to overcome. That’s why there are ideal weights for various projectiles (arrows, bolts, rocks, balls etc.) to achieve optimum range, depending on the mechanism used to propel them.

So it can be expected that for a large animal (=high inertia relative to muscle force), the benefits of greater mechanical advantage and torque dominate (i.e. quadrupedalism), while for small animals (=low inertia relative to muscle force) this might not be the case, or perhaps even turned on its head (I don’t know if this has been studied empirically, but it could be that very small bipeds with powerful limb musculature can actually out-turn similar-sized quadrupeds, because the quadruped isn’t limited by how much torque it can produce but by how fast it can move its limbs).
--
So at small sizes (not sure how small exactly, but you should probably think Velociraptor, certainly not Tyrannosaurus), bipedal animals can probably match or even exceed the turning performance of quadrupedal animals, because at these sizes, how fast animals can turn is not limited by how much torque they can produce (the muscles of small animals are far stronger in proportion to their mass and inertia, so how much torque they can produce isn’t as much of an issue for them).

On the other hand, in a larger animal, a biped (at least one with horizontal posture, like a theropod) will never be able to match a similar-sized quadruped in turning performance unless it has vastly larger musculature (and even then it’s probably a stretch among real animals), because the quadruped can simply produce far more torque and also has far lower rotational inertia. That is only relevant for rotational motion, however, because in linear motion inertia is simply mass, and acceleration is simply force divided by mass.

That’s also one reason for why you tend to get less cursorial limb proportions the larger animals get. At small sizes, an animal’s ability to accelerate its body is limited mostly by how fast they can move the tips of their limbs. At larger sizes, the limiting factor becomes how much force they can produce at those limb tips to overcome the body’s mass (= inertia). There is less point in having limbs that would be able to move very quickly and efficiently once high speed is achieved, if achieving that speed would take excessively long due to the force they can produce being too small.
––

[Supercommunist]

Another possibility:

Some early dinosaur ancestors might have wrestled like monitor lizards. Animals with more bipedal adaptations were more balanced in these stances and had a reproductive advantage. There were other mobility advantages in this lower size range so they continued to become more bipedal.

[Infinity Blade]

I'm still kind of interested in how much lateral flexibility of the torso also figures for theropods when it comes to reaching for the flanks of their prey. Obviously it's not going to be as flexible as the neck, but still, I figure some torso flexion is better than nothing.

Also, I know we've kind of discussed allosauroids vs tyrannosaurids in lateral neck flexibility. Do we know about other theropods in this regard, like megalosaurids, ceratosaurids, spinosaurids, etc.?

[theropod]

I don’t think there have been quantitative modelling studies (in fact the only theropod that seems to have one of those for neck flexibility is Allosaurus jimmadseni), but there are some inferences that can be drawn.
For example, Méndez 2012 noted how the cervical series of Carnotaurus is distinctly more robust and stiffer than that of Majungasaurus (in its case that is mainly due to the highly elongated epipophyses. You can usually at least draw a decent qualitative comparison on the basis of vertebral morphology (how long are various processes that would impede movement in specific directions? How massive are the centra? Are they opisthocoelous or just flat or biconcave etc.).

For example, things like Acrocanthosaurus and Spinosaurus obviously have long spinous processes that would limit dorsiflexion, and also (by the the elastic limits of the ligaments connecting them, though those could be quite lax) probably would limit ventroflexion. But they also have relatively long, slender, opisthocoelous centra, suggesting that they would have been fairly flexible laterally.

--
Méndez, A.H. 2012. The cervical vertebrae of the Late Cretaceous abelisaurid dinosaur Carnotaurus sastrei. Acta Palaeontologica Polonica 59 (3): 569–580.

[theropod]

Another thing that I think probably goes in favor of bipedalism (and may or may not have been an important factor in its evolution), is that it likely is iherently more energetically efficient. A bipedal strider has a lot fewer moving parts than a quadrupedal one, and more moving parts equals more energy losses. A biped has two legs that it has to extend and retract in order to move, and the rest of its body can remain pretty much unmoving. A quadruped has four legs that need to move, and additionally will almost necessarily incorporate its vertebral collumn into its stride to varying degrees (very prominently in sprawlers, less so in erect-limbed animals, which is what makes the latter more efficient after all).
While that is a key advantage for peak performance (that they can utilize more power from muscle groups in the axial skeleton than bipeds can is a key reason the fastest quadrupeds outrun the fastest bipeds, and for speed in short bursts energy efficiency does not matter as much as peak power putput), it is probably inferior to taking the entire locomotory power from the limbs when it comes to efficiency.

Now, of course some quadrupeds have evolved lots of adaptations (most of which are available to bipeds too, e.g. springy tendons for energy storage) to minimize their energy-use even while remaining quadrupedal, because evolving into bipeds just wouldn’t be an evolutionarily viable step for them (certainly a camel won’t become a biped any time soon, just because it might be a tad more efficient, as the intermediate steps it would need to go to would be a lot less efficient than either). But it can’t be denied that the two largest extant bipedal striders, humans and ratites, are also in the running for the two best long-distance runners.

[Supercommunist]

Oct 12, 2024 18:42:02 GMT 5 theropod said:

Another thing that I think probably goes in favor of bipedalism (and may or may not have been an important factor in its evolution), is that it likely is iherently more energetically efficient. A bipedal strider has a lot fewer moving parts than a quadrupedal one, and more moving parts equals more energy losses. A biped has two legs that it has to extend and retract in order to move, and the rest of its body can remain pretty much unmoving. A quadruped has four legs that need to move, and additionally will almost necessarily incorporate its vertebral collumn into its stride to varying degrees (very prominently in sprawlers, less so in erect-limbed animals, which is what makes the latter more efficient after all).
While that is a key advantage for peak performance (that they can utilize more power from muscle groups in the axial skeleton than bipeds can is a key reason the fastest quadrupeds outrun the fastest bipeds, and for speed in short bursts energy efficiency does not matter as much as peak power putput), it is probably inferior to taking the entire locomotory power from the limbs when it comes to efficiency.

Now, of course some quadrupeds have evolved lots of adaptations (most of which are available to bipeds too, e.g. springy tendons for energy storage) to minimize their energy-use even while remaining quadrupedal, because evolving into bipeds just wouldn’t be an evolutionarily viable step for them (certainly a camel won’t become a biped any time soon, just because it might be a tad more efficient, as the intermediate steps it would need to go to would be a lot less efficient than either). But it can’t be denied that the two largest extant bipedal striders, humans and ratites, are also in the running for the two best long-distance runners.

Studies on human endurance indicate that anthropologists seriously overestimated how much physical endurance we had. It seems our species
ability to traverse long distances mostly boils down to our ability to carry water with us and pacing.

While some human populations have persistence hunted (and on occasion still do), the success of this unlikely foraging strategy may be best explained by the application of another adaption – high cognitive capacity. With dedication, experience and discipline, capitalising on their small endurance advantage in high temperatures, humans have a chance of running a more athletic prey to exhaustion.

physoc.onlinelibrary.wiley.com/doi/full/10.1113/EP088502

If we had more efficent bipedal stance maybe we would be able to compete with true endurance runner likes horses, pronghorns, wolves or hyenas.

[theropod]

That humans are indeed able to compete with those animals in terms of endurance, though of course not speed, is impressive considering our cardiovascular inferiority (as explained in the first paragraph of the paper you cite).

As your paper also explains, humans may not be efficient runners compared to other mammals (which considering the distinct lack of cursorial limb proportions does make sense, one would need to compare a human to another plantigrade animal of similar size and femur/tibia ratio for the comparison to be informative with regard to bipedalism), and our advantages at high heat are probably mainly related to sweating and the ability to carry water, but what humans are above average in is efficiency while walking:

Humans have a net cost of transport (NCOT; the whole-animal energy cost to move a unit distance) while walking that is 25% lower than predicted based on the best fit NCOT–mass relationship across mammalian species, but an NCOT while running that is 27% higher (Halsey & White 2012). Horses have an NCOT about 20% above predicted (Wickler, Hoyt, Cogger, & Myers, 2003).

Whereas of course horses, pronghorns, wolves or hyaenas all have far more cursorial limb proportions than us, so it makes some degree of sense for them to be more efficient while running. But running is not the locomotory mode in which we humans tend to cover the most distance.
In fact, as your paper also explains, average speed of humans during endurance hunts was around 6 km/h, which means the humans in questions must have been walking, rather than running, for at least a large portion of that hunt. The analogous speeds for horses and dogs on hunts or long-distance races that they cite were also low (8-11 km/h), which suggests that the animals were walking or at best trotting for much of that time (11 km/h is not a galloping horse). In other words, it is easy to see how greater efficiency while walking would potentially be very important during such activities, as well as during other normal, day to day activities of a nomadic hunter gatherer, or any other creature reliant on covering large distances at moderate speed on a daily basis.

So none of that refutes that a bipedal stance might be inherently more energy-efficient than a quadrupedal one.

[Supercommunist]

Oct 26, 2024 18:32:30 GMT 5 theropod said:

That humans are indeed able to compete with those animals in terms of endurance, though of course not speed, is impressive considering our cardiovascular inferiority (as explained in the first paragraph of the paper you cite).

As your paper also explains, humans may not be efficient runners compared to other mammals (which considering the distinct lack of cursorial limb proportions does make sense, one would need to compare a human to another plantigrade animal of similar size and femur/tibia ratio for the comparison to be informative with regard to bipedalism), and our advantages at high heat are probably mainly related to sweating and the ability to carry water, but what humans are above average in is efficiency while walking:

Humans have a net cost of transport (NCOT; the whole-animal energy cost to move a unit distance) while walking that is 25% lower than predicted based on the best fit NCOT–mass relationship across mammalian species, but an NCOT while running that is 27% higher (Halsey & White 2012). Horses have an NCOT about 20% above predicted (Wickler, Hoyt, Cogger, & Myers, 2003).

Whereas of course horses, pronghorns, wolves or hyaenas all have far more cursorial limb proportions than us, so it makes some degree of sense for them to be more efficient while running. But running is not the locomotory mode in which we humans tend to cover the most distance.
In fact, as your paper also explains, average speed of humans during endurance hunts was around 6 km/h, which means the humans in questions must have been walking, rather than running, for at least a large portion of that hunt. The analogous speeds for horses and dogs on hunts or long-distance races that they cite were also low (8-11 km/h), which suggests that the animals were walking or at best trotting for much of that time (11 km/h is not a galloping horse). In other words, it is easy to see how greater efficiency while walking would potentially be very important during such activities, as well as during other normal, day to day activities of a nomadic hunter gatherer, or any other creature reliant on covering large distances at moderate speed on a daily basis.

So none of that refutes that a bipedal stance might be inherently more energy-efficient than a quadrupedal one.

One study suggests that persistence walking would be less efficient than intermittent running for hunts.

www.sciencedirect.com/science/article/abs/pii/S0047248422001075

[quoe] Our simulations predicted that walking would be successful in persistence hunting of low- and nonsweating prey, especially under hot and humid conditions. However, simulated persistence hunts by walking yielded a 30–74% lower success rate than hunts by running or intermittent running. In addition, despite requiring 10–30% less energy, successful simulated persistence hunts by walking were twice as long and resulted in greater exhaustion of the hunter than hunts by running and intermittent running. These shortcomings of pursuit by walking compared to running identified in our simulations could explain why there is only a single direct description of persistence hunting by walking among modern hunter-gatherers. Nevertheless, walking down prey could be a viable option for hominins who did not possess the endurance-running phenotype of the proposed first persistence hunter, Homo erectus. Our simulation results suggest that persistence hunting could select for both long-distance walking and endurance running and contribute to the evolution of locomotor endurance seen in modern humans.[/quote]

This makes sense to me because while a trotting wolf or a hyena would be using more energy than a walking human, they would cover distance a lot quicker and can rest up and recontinue trotting once the walking human catches up.

www.sciencedirect.com/science/article/abs/pii/S0047248422001075

I could have sworn I once read a study that suggested lizards don't actually use less energy while running on two legs than on all four, but I can't find it at the moment. Maybe I imagined it.

Historically, we do know that it was much more efficient for people to ride on horses than walk. For instance, it was basically impossible for foot based armies to catch up to steppe nomads that were all mounted on horses.

There are other studies on the subject that suggest human persistence hunting in general really wasn't an efficient hunting method for us. It probably only worked in very specific circumstances when the heat was extremely oppressive.

From direct observation over a twenty-year period of
well over 100 successful hunts by Hadza using bows and arrows,
and of an even greater number of unsuccessful hunting and scavenging opportunities, we have never observed persistence hunting by the Hadza, nor have informant interviews revealed this as
a strategy they practice. Independent research on the Hadza by
others has, likewise, produced no reference to persistence hunting. In fact, the Hadza often decide to abandon the blood trail of
a wounded animal headed into heavy vegetation, choosing to
satisfy immediate hunger by collecting some honey and figuring
that the odds of encountering another prey animal are likely
enough to obviate the need for further, difficult tracking (Bunn
et al., 1988).

We agree with the opinion of Bramble and Lieberman
(2004) that early Homo required a high-quality diet, which included a substantial meat component, and that it was thus a capable carcass forager. However, our understanding of the
paleoenvironment, paleoecology, and archaeology of early
Homo sites, reviewed here, makes us dubious about their further suggestion that ER might have been employed regularly
and successfully in service of that foraging pursuit. We are reluctant to assign to early Homo the impressive tracking skills
of the Kalahari San, when the cognitive and meat-foraging
abilities of Plio-Pleistocene Homo are active research issues.
The behavioral pattern that selected for ER in the genus
Homo remains unclear, but it seems likely that hunting and
scavenging contributed minimally, if at all.

Another important thing to note is that human hunters can quickly kill or fatally injure with a single spear throw, whereas other endurance predators have to bite their prey to death which takes more energy.

www.originalwisdom.com/wp-content/uploads/bsk-pdf-manager/2019/05/Pickering-and-Bunn_2007_TheEnduranceRunningHypothesis.pdf

I am a bit iffy on whether bipedal animals naturally have more endurance than quadruped. Ostriches and ratites are very impressive endurance runners but they have a unidirectional respiratory so its hard to directly compare them to quadrupedal mammals with a less efficient respiratory system. Similarly, humans are very efficient sweaters and can transport water which allows us to travel long distances that a lot animals can't but that's not due to sheer endurance.

There are studies that show ostriches have very energy efficient strides but I am not sure if that's because bipedalism is inherently more energy efficient or if its because a bipedal animals speed caps out sooner than a quadrupeds so there was more evolutionary pressure for ratites to specialize in long distance running compared to fleet quadrupeds.