New paper says dinosaurs were endomorphs.

Facebooktwitterredditpinterestlinkedintumblrmailby feather
From left to right: Endomorphic Jay Cutler, Mesomorphic Arnold Schwarzenegger, and Ectomorphic poster-child Frank Zane
From left to right: Endomorphic Jay Cutler, Mesomorphic Arnold Schwarzenegger and Ectomorphic poster-child Frank Zane

Endo-what now? Allow me to explain.

If one studies physical fitness (academically, or practically), then one is bound to come across the three main human body types. The endomorph, mesomorph and ectomorph.

Endomorphs are characterized by their ability to easily gain weight (be it fat, or muscle).

Ectomorphs are characterized by their ability to easily lose weight (fat, or muscle)

Mesomorphs are the middle ground group that appear to have the most malleable bodies.

In general, endomorphs have lower metabolisms than the other two, while ectomorphs tend to “run hot” all the time. Few people are all one way, or the other, but a notable dominance of one type, or another is usually prevalent.

The endo/ecto part can get confusing; especially if one is used to these prefixes in the context of endotherm/ectotherm. The names seem to be reversed from what one might normally hear (ectomorphs being more “warm-blooded” than endomorphs etc). The names have nothing to do with thermophysiology. They were coined after the germinative layers of the body during embryonic development. Endoderm forms the digestive tract, and endomorphs are usually stereotyped as fat. Ectotoderm forms the skin, and ectomorphs are usually stereotyped as being “all skin and bones.”

The reason I went with these specific bodybuilders (Jay Cutler, Arnold Schwarzenegger and Frank Zane) was partly to buck these stereotypes, but also to point out something that the news outlets are missing. Namely that having a lower metabolic state, does not mean one is a “couch potato” or has “forgone exercise.” Bigger, means more massive. That may mean fat, but as one can see above, it also can mean muscle and bone. Dinosaurs were not fatter than mammals. They were bigger.

So what am I rambling on about?

Grab a calculator and come along for the ride.

A new paper on dinosaur growth by Dr. Brian McNab of the University of Florida, has recently come out in the Proceedings of the National Academy of Science. In the paper, Dr. McNab uses mathematical modeling to come up with a hypothesis of dinosaur growth, and its thermophysiological consequences. Perhaps it is because of its heavy use of abstraction (math), or perhaps it is because McNab covers his bases by dedicating an entire section to possible problems with the model, but either way this paper appears to have come and gone without leaving much of a mark. This is unfortunate as the results are really neat.

McNab, B. 2009. Resources and Energetics Determined Dinosaur Maximal Size. PNAS 106(27):1-5. doi: 10.1073/pnas.0904000106

In the paper, McNab uses a non-linear regression equation from Nagy (1999, 2005) for field metabolic rate. The equation is:

FEE = a*mb

  • FEE = Field Energy Expenditure (i.e. the cost of running around gathering food etc. AKA the Field Metabolic Rate [FMR]).
  • a = a coefficient that determines the level at which energy is expended.
  • m = mass of the taxon in question (in grams).
  • b = the power of mass (i.e. the effect of allometry).

McNab points out that when FEE equals the maximal daily field expenditure (i.e. if an animal is pushing itself to the limits in order to gather as much energy as possible in a given environment), then a trade-off occurs between the level at which energy is used up (a), and the overall body size of the taxon in question (m). So as energy expenditure increases, body mass decreases.

To put it more succinctly, this model states that hypermetabolic hamsters are incapable of growing to the size of dogs, because of the fact that they are hypermetabolic.

This seems to be a one-way street relationship. If energy expenditure were to decrease, body mass would not necessarily increase. Further, McNab points out that there is a limit to how much variation is “allowed” between these two variables.

However, at 1 extreme along a continuum, the most sluggish of species would not be able to sustain the largest masses potentially permitted by resources because they could not find a sufficient resource base in a limited area to support a large mass, which therefore would reduce K [Max daily field expenditure] and maximal m.

Despite the elegant math, this is biology we are talking about. That means it’s messy, and full of caveats. However, when boiled down to its most simplistic parts, it does point out something that tends to get overlooked, and might even seem counterintuitive:

Given two taxa living in the same environment, and that are dependent on the same resources, the taxon that costs the least amount of energy “to run,” has the potential to achieve the larger body size.

In the bodybuilder example above, Jay Cutler should require less food to maintain his normal body functions, than Frank Zane. Because of this Cutler should be able to grow more muscle than Zane, using the same amount of food. Indeed, the hard part for many endomorphic people who take on an exercise regime, is to limit the amount of food they normally eat. Meanwhile ectomorphs who try to gain body mass, are forced to eat as much as 6 – 7 times a day.

Okay, so McNab has a nice pretty equation, but how well does it really work?

The mathematical models used in this paper, are based off of real world data on FMR of various animals. As such, there actually is plenty of data with which to test this model out.

First we will need some numbers.

For terrestrial mammals, the following numbers get plugged into the above equation:

FEE = 4.82m0.734

Just plug in the mass (in grams) of your favourite land mammal, and find out its field energy expenditure (expressed as kilojoules/day).

For instance: a black rhinoceros (Diceros bicornis) weighing in at 1600kg, would have an FEE of:

4.82(1,600,000g)0.734 = 1.73×105kJ/d

I don’t know about you, dear readers, but I have trouble conceptualizing just how much a kilojoule/day really is. So I took McNab’s formula one step further and converted it to kilocalories (or dietary Calories) per day.

There are two ways of doing this. One can take the kJ/d number, and multiply it by 0.000238845896627.

Or, one can do what I did and just use this handy dandy conversion calculator from Just remember to multiply your kJ/d by 1000, before entering it in the calculator.

If one does this, one finds that a black rhino can potentially burn through 41,000 calories in a day!!

Eat your Wheaties kids.

Just for shits and giggles, I input my average mass into the equation (~76kg), and after going through the conversion calculator, came up with 4,402 calories per day. Upon first glance, this might cause one to cry foul of the formula. This is, after all, almost twice the FDA’s daily caloric requirements for an “average” individual. However, one must keep in mind that this is what my maximum daily energy ependiture should be.

Humans are unique among most animals, in that our social lifestyle has permitted many of us to easily acquire our daily calories with very minimal effort (hence the current obesity problems in many developed parts of the world). Most animals (especially the herbivores) spend most of their free time searching for enough food to satisfy their needs. In the case of herbivores, much of this foraging is related to the relatively low nutritional value of what is ingested. As such, FEE – while likely to overestimate the daily nutritional requirements, is going to hover pretty close to what is actually used on a day to day basis.

Given, that I am an avid gym rat, who focuses heavily on weight lifting, the caloric requirement estimated for me, is actually remarkably close to my actual daily requirement (which is hovering around ~3500 calories/day). If anything, McNab’s formula makes me think I should be working out more. 🙂

The largest mammal known to have ever existed: Paraceratherium. Image from Wikipedia
The largest land mammal known to have ever existed: Paraceratherium transouralicum. Image from Wikipedia

McNab tested his model using a 7.5 tonne (the paper says it’s a short ton, but the calculations make it pretty clear it is not) African elephant (Loxodonta africana), and then on the largest terrestrial mammal ever known to have existed (Paraceratherium transouralicum). His results for the latter were a whopping 7.10×105 kJ/d, or ~170,000 calories a day!!

That critter must have been munching nigh 24/7.

The nutritiona guide for the hypothetical Brachiosaurus in the mammal equation
The nutritiona guide for the hypothetical Brachiosaurus in the mammal equation

Which brings us to the crux of the argument. If an 11 tonne Paraceratherium was packing down that many calories, how much would a “warm-blooded” (in the mammalian sense) sauropod have to eat?

A very big Brachiosaurus altithorax (56 tonnes, if Wikipedia is to be trusted), would have a potential energy expenditure of:

4.82*56000000g0.734 = 2.35×106 kJ/d, or 561,000 calories per day!

According to McNab:

Daily feeding time in terrestrial mammalian megaherbivores increases with body size, a 3-ton[sic] African elephant spending 16 of 24-h feeding.

That’s just a 3 tonne elephant. Paraceratherium was almost 4 times that size, and Brachiosaurus was 18 times it. There just weren’t enough hours in the day.

Other researchers have taken on the “sauropods couldn’t eat enough food in a 24hr period’ argument before (Bakker 1986, Paul 1998). Opponents rightly point out that sauropods have the advantage over mammals in that they don’t waste a lot of ingestion time, chewing their food. All the real digestive work takes place internally with the gizzard, and stomach.

McNab tackles this issue as well, pointing out that while using your stomach to “chew your food for you” might increase the amount of food that can be taken in, digestion is ultimately limited to how much of that food gets absorbed into the system. Even after pulverization, the food still needs to sit, and ferment in the gut long enough for its nutrients to be fully extracted. The long and the short of it appears to be that both methods leave one with the same, or nearly the same results. Whatever possible efficiency that may come from using gastroliths to pulverize plant matter (and McNab argues, that it is not as efficient as mastication), it would not be enough to account for the substantially larger body masses seen in sauropods and other herbivorous dinosaurs.

Okay, so what about the other side. McNab points out that field energetic studies of lizards show that FEE/FMR is approximately 6.2% that of equivalent sized mammals. For the B.altithorax in our equation above, that would give it an energy expenditure of approximately 35,000 calories.

Quite the substantial difference. Too substantial, it would appear.

McNab points out:

If dinosaur FEEs conformed to this lizard curve, they would have to weigh 330 tons[sic] to attain the elephant’s estimated FEE, a mass that is much greater than the 40 to 80, and possibly 100, tons[sic] found in the largest sauropods. Therefore, it is unlikely that dinosaurs had such low FEEs.

The assumption here is a simple one. Sauropods got as big as they could, because they could. Argentinosaurus huinculensis pushed the bar for max sauropod size, with estimates nearing 100 tonnes, and Amphicoelias fragillimus is likely to push that number even further up (assuming we ever find more of it), so it would appear that sauropods found a way around that whole physics issue. Theoretically the only thing stopping them from achieving even greater sizes would be the environment (the ultimate limit to any animal’s potential). That no sauropod comes close to the estimated size imposed by the “standard lizard” curve, is highly suggestive that they didn’t have the “standard lizard” physiology.

Brachiosaurus brancai pic from Wikipedia
Brachiosaurus brancai pic from Wikipedia

So dinosaurs appear not to have been mammals. Nor do they appear to have been “standard lizards” either. Then again, what the heck is a standard lizard anyway?

The mathematical models used in this paper were all based off of data from dozens of species of mammals, reptiles, amphibians, and birds. This gives these models weight in the real world, as they are based off of real world data. However, since it takes a look at the whole group, it is possible to have certain members skew the results one way, or the other.

In the case of our “standard lizards,” the issue is a collection bias in favour of those oh so popular iguanians.

On the family tree of Squamata, there is an early split that occurs between one group of lacertilians that focused heavily on visual cues, and sit-and-wait style hunting (Iguania), and another group that focused more on chemosensory cues gathered through the tongue (Scleroglossa). The latter group are characterized as active foragers. Both groups are immensely successful in this day and age. Yet despite the greater species count for scleroglossans, iguanians are almost always the lizards of choice for pet enthusiasts, and researchers. It is the latter that has lead to many popular misconceptions about reptile energetics. There’s a pretty notable difference between getting an iguanian like a fence lizard (Sceloporous) to run on a treadmill, and getting a scleroglossan like a whiptail (Cnemidophorous) to do it. In fact, getting the latter to sit still is likely to be the bigger chore.

Take iguanians out of the equation, and things start to look quite a bit different, as McNab points out:

Aggressively predatory lizards of the genus Varanus, including the largest living lizard, the Komodo dragon
(V. komodoensis), which feeds on deer, pigs, and occasionally water buffalo and people have FEEs that average 3.6 times those of most lizards (Fig. 3), reflecting their high level of activity and high body temperatures.

Using field data obtained from six varanid species ranging from 2.2kg to 47kg, McNab produces the following formula:

FEE = 1.07m0.735

Which results in a 59 ton sauropod with an FEE of 5.13×105kJ/d, or 123,000 calories per day; which is practically the same max caloric expenditure of the 7.5 ton African elephant tested with the first equation.

Giganotosaurus carolinii pic, like most, from Wikipedia
Giganotosaurus carolinii, the largest theropod currently known (reliably), but not the largest theoretical theropod, according to McNab’s equation. Pic from Wikipedia (of course)

What about all those beloved theropods? Does this equation cover carnivores as well?

Carnivores, theoretically, should have a higher FEE for their body size, as the food they eat, often fights back. So what happens to carnivores in these models?

McNab points out that carnivorous mammals have FEEs that are about 1.5times greater than the standard terrestrial mammal curve. Because of this, mammalian carnivores appear to be more restricted in their body mass potential.

According to McNab:






This analysis implies that the largest terrestrial mammalian carnivores should fall between 0.6 and 1.0 ton, which encompasses the largest species known to have existed, including the creodont Megistotherium osteothlastes (880 kg) and the short-faced bear (Arctodus simus, 0.7 to 1.0 ton), a species that was committed to carnivory.

Interestingly, McNab argues that terrestrial mammalian carnivores appear to be limited to maximum masses of 1 tonne. I can’t help but retain some skepticism here. Certainly part of the reason for smaller body sizes in mammalian carnivores, is their rather low efficiency of ingestion (mammalian carnivores tend to leave between 30-50% of a carcass behind after a kill). Should a mammalian carnivore evolve to be “less picky” I’d expect that max size to go up at least a little bit.

Using the varanid curve, an 8 tonne theropod like Giganotosaurus carolinii, would have an FEE equivalent to 594kg mammalian carnivore; which is approaching the theoretical top-end of the terrestrial mammal spectrum.

So by operating at 22% the rate of mammals, dinosaurs were able to grow up to 8 times larger than their fuzzy ecological equivalents.

McNab goes on to compare the biomass of dinosaurs in the Mesozoic, to those of extant mammals in Africa today, and cites work that suggests that dinosaur biomasses were greater, and more geographically limited than those of today’s mammals. It is interesting work, but I have to take issue with his Aldabra tortoise (Geochelone gigantea) FEE calculation. Admittedly, FMR data for chelonians is pretty sparse, and McNab used what data was available. The problem is that the data comes from that of gopher tortoises (Gopherus agassizii); an animal that spends a whopping 157hrs a year, being active. Other cited turtle data comes from Russian tortoises (Testudo horsfieldii), and spur-thighed tortoises (Testudo graeca). The problem I have with all of these taxa in question, is that they live very similar lives which include large periods of the year where they are inactive. This is something that Aldabras, or any other giant tortoise (to my knowledge) does not do. As such, I expect the FEE for the Aldabras in McNab’s equation, to be excessively low.

While comparing diet data for dinosaurs, with those of mammals, McNab operated under the assumption that Mesozoic plant communities were of equal productivity to those of the East African plains (i.e. the same basic setup as African elephants). Prior to a few years ago, this would seem unlikely. The major plants of the Mesozoic (prior to the Cretaceous, at least), were gingkoes, ferns, horsetail and conifers. All are plants that were thought to contain little nutritional value compared to extant angiosperms. However, a new study by Hummel et al (2008) has cast doubt on this old view, and found that these plants can actually hold their own (nutritionally) with angiosperms. The model is likely oversimplifying the nutritional value of the flora in the Jurassic, but even so, no one has argued that Mesozoic plant communities carried more nutritional “bang for the buck” than extant grass dominated communities. This would seem necessary to counter the current argument over dinosaur metabolism.

The last section of the paper (save the conclusions) is devoted to covering potential difficulties with the results of this study, compared to other studies of dinosaur energetics. McNab covers the issues with growth rate and thermophysiology, dinosaur bones in latitudes that appear to have been too cold for bradymetabolic “traditional” reptiles, as well as O2 isotope comparison studies. McNab reminds everyone that the greatest differences between bradymetabolic animals, and automatic endotherms occurs at masses less than 50 grams. As sizes increase, the differences between both groups start to get harder to distinguish.

In the end, we have a paper that takes a detailed mathematical view of the energy expenditure and theoretical limits of extant animals. Does it give the final word on dinosaur thermophysiology?

No, but it does serve to greater elucidate the relationships between animals and their respective environments. Pushing these formulae out to animals that were many times the size of any terrestrial animal today is a bit dicey, but it does at least give us an idea of what we can potentially expect.

So there you go. Dinosaurs were endomorphs.



Bakker, R. 1986. The Dinosaur Heresies. William Morrom. New York. ISBN:0821756087, 978-0821756089.
McNab, B. 2009. Resources and Energetics Determined Dinosaur Maximal Size. PNAS 106(27):1-5. doi: 10.1073/pnas.0904000106
Hummel, J., Gee, C.T., S?dekum, K-H.,? Sander, P.M., Nogge, G., Clauss, M. 2008. In Vitro Digestibility of Fern and Gymnosperm Foliage: Implications for Sauropod Feeding Ecology and Diet Selection. Proc. Biol. Sci. 275: 1015-1021
Nagy, K.A. 2005. Review. Field Metabolic Rate and Body Size. J. Exp. Biol. 208:1621?1625.
Nagy, K.A., Girard, I.A., Brown, T.K. 1999. Energetics of Free-Ranging Mammals, Reptiles, and Birds. Ann. Rev. Nutr. 19:247?277.
Paul, G. S. 1998. Terramegathermy and Cope’s Rule in the land of titans. Modern Geology (Special Issue) 23:1?4179?217.
Facebooktwitterredditpinterestlinkedintumblrmailby feather

6 Responses to New paper says dinosaurs were endomorphs.

  1. Avatar Tim Morris
    Tim Morris says:

    I find this very interesting. I would estimate that I am a Mesomorph, because I am fairly sedentary, and at the same time only slightly overweight. If I were a Endomorph, I would probably be a real lardass, or if I were an Ectomorph, I’d be one of those lanky idiots who doesnt need to excercise to stay thin.

    Also, I’m not sure how relevant this is to a science blog, but as a homosexual I personally view all three as very attractive in the ideal sense. I dont find any of them “weird or creepy” as some people view extreme bodybuilders etc

  2. Zane = ectomorph
    Arnold = ecto-mesomorph
    Jay = endo-mesomorph or even mesomorph
    These people are bodybuilders and they don’t look like ordinary people because they can control the “shape” of the body and may seem as having a different body type .

    • While true that the ecto-meso-endo body type is more like a spectrum than discrete character states, I did not want to confuse people by throwing in ecto-meso and endo-meso (I almost didn’t throw in mesomorph altogether), as the entire point was just to illustrate what McNab’s paper was referring to regarding dinosaurs.

  3. Paraceratherium was the largest land mammal, not the largest mammal. The blue whale gets that honor.

  4. yeah