[Editor’s note: A response from the authors can be found here. It answers many of the questions I had about the paper, though I feel the biggest question remains open for debate. I appreciate the authors taking their time to answer my questions, and PLoS ONE for allowing this type of open communication.]
This post has taken an inordinate amount of time to write up. Mostly because it required finding enough free time to sit down and just type it out. So I apologize ahead of time for bringing up what is obviously old news, but I felt this paper was an important one to talk about, as it relied on a old, erroneous, but very pervasive, popular and rarely questioned hypothesis for how automatic endothermy (mammal and bird-style “warm-bloodedness”) evolved.
Back in November, a paper was published in the online journal: PLoS ONE. That paper was:
Pontzer, H., Allen, V. & Hutchinson, J.R. 2009. Biomechanics of Running Indicates Endothermy in Bipedal Dinosaurs. PLoS ONE.Vol 4(11): e7783.
Using muscle force data for the hindlimbs of theropods, and applying it to a model based on Pontzer (2005, 2007), the authors were able to ascertain the approximate aerobic requirements needed for large bipedal theropods to move around. Their conclusion was that all but the smallest taxa had to have been automatic endotherms (i.e. warm-blooded).
Time to stop the ride and take a closer look at what is going on here.
In 2004, John Hutchinson – of the Royal Veterinary College, London UK – performed a mathematical study of bipedal running in extant taxa. He used inverse dynamics methods to estimate the amount of muscle that would be required for an animal to run bipedally. He then tested his models on extant animals (Basiliscus, Iguana, Alligator, Homo, Macropus, Eudromia, Gallus, Dromaius, Meleagris, and Struthio). The predictive capacity of his model proved to be remarkably substantial and stable (Hutchinson 2004a). A follow up paper in the same issue (Hutchinson 2004b) used this model to predict bipedal running ability in extinct taxa (Compsognathus, Coelophysis, Velociraptor, Dilophosaurus, Allosaurus, Tyrannosaurus and Dinornis). Results from this study echoed previous studies on the running ability of Tyrannosaurus rex (Hutchinson & Garcia 2002), as well as provided data on the speed and agility of other theropod taxa.
Meanwhile in 2005, Herman Pontzer – of Washington University in St. Louis, Missouri – did a series of experiments to determine what was ultimately responsible for the cost of transport in animals. To put it another way: Pontzer was searching for the most expensive thing animals have to pay for in order to move around. One might intuitively assume that mass is the ultimate cost of transport. The bigger one gets, the more energy it requires to move a given unit of mass, a certain distance. However experiments on animals found the opposite to be the case. It actually turns out that being bigger makes one “cheaper” to move. So then what is going on here?
Pontzer tested a variety of options for what could be happening; from extra mass, to longer strides. In the end Pontzer found that the effective limb length of animals, was ultimately the limiting factor in their locomotion. Effective limb length differs from the entirety of the limb. Humans are unique in that our graviportal stance has us using almost our entire hindlimbs. Most animals, however, use a more crouched posture that shrinks the overall excursion distance of the hindlimb (or the forelimb). By taking this into account Pontzer was able to find the one trait that seemed to track the best with cost of transport in animals over a wide taxonomic range (essentially: arthropods – birds).
This latest study combines these two technique in order to ascertain the minimum (or approx minimum) oxygen requirements bipedal dinosaurs would need in order to walk, or run.
As with the previous papers, the biomechanical modeling and mathematics are elegant and robust. However, this paper is not without its flaws. For instance in the paper the authors mention:
We focused on bipedal species, because issues of weight distribution between fore and hindlimbs make biomechanical analysis of extinct quadrupeds more difficult and speculative.
Yet this did not stop the authors from applying their work on bipeds, to predicting the maximum oxygen consumption of quadrupedal iguanas and alligators. No justification is ever really given for why the authors chose to do this. Making things even more confusing, just a few sentences later, it is mentioned (ref #s removed to avoid confusion):
Additionally, predicting total muscle volumes solely from hindlimb data for the extant quadrupeds simply assumes that the fore and hindlimbs are acting with similar mechanical advantage, activating similar volumes of muscle to produce one Newton of GRF. This assumption is supported by force-plate studies in other quadrupeds (dogs and quadrupedal chimpanzees)
The force plate work cited is for quadrupedal mammals. However, mammals are not reptiles. As Nicholas Hotton III once mentioned (1994), what works for mammals, does not necessarily work for reptiles. This is especially so for locomotion.
In many reptiles (including the taxa used in this study) the fore and hindlimbs are subequal in length; with the hindlimbs being noticeably longer and larger. Most of the propulsive power in these reptiles comes from the hindlimbs (which have the advantage of having a large tail with which to lay their powerful leg retractor on). The result is that – unlike mammals – many reptiles are “rear wheel drive.”
The last problem is by far the largest, and ultimately proves fatal to the overall conclusions of the paper. The authors operated under the assumptions of the aerobic capacity model for the evolution of automatic endothermy.
It is here that we come to the crux of the problem, and the main subject of this post.