What allowed sauropods to get big, and what kept them from getting any bigger?

January 12, 2018

This post started out as a comment on this thread, kicked off by Dale McInnes, in which Mike Habib got into a discussion with Mike Taylor about the max size of sauropods. Stand by for some arm-waving. All the photos of outdoor models were taken at Dino-Park Münchehagen back in late 2008.

I think it’s all too easy to confuse how big things do get from how big they could get, assuming different selection pressures and ecological opportunities. I’m sure someone could write a very compelling paper about how elephants are as big as they could possibly be, or Komodo dragons, if we didn’t have indricotheres and Megalania to show that the upper limit is elsewhere. This is basically what Economos (1981) did for indricotheres, either forgetting about sauropods or assuming they were all aquatic.

Truly, a mammal of excellence and distinction. With Mike and some dumb rhino for scale.

In fact, I’ll go further: a lot of pop discussions of sauropod size assume that sauropods got big because of external factors (oxygen levels, etc.) but were ultimately limited by internal factors, like bone and cartilage strength or cardiovascular issues. I think the opposite is more likely: sauropods got big because of a happy, never-repeated confluence of internal factors (the Sander/et al. [2008, 2011, 2013] hypothesis, which I think is extremely robust), and their size was limited by external, ecological factors.

Take a full-size Argentinosaurus or Bruhathkayosaurus – even modest estimates put them at around 10x the mass of the largest contemporary predators. Full-grown adults were probably truly predator-immune, barring disease or senescence. So any resources devoted to pushing the size disparity higher, instead of invested in making more eggs, would basically be wasted.

If there was reproductive competition among the super-giants, could the 100-tonners have been out-reproduced by the 70-tonners, which put those extra 30 tonnes into making babies? Or would the 100-tonners make so many more eggs than the 70-tonners (over some span of years) that they’d still come out on top? I admit, I don’t know enough reproductive biology to answer that. (If you do, speak up in the comments!) But if – if – 70-tonners could out-reproduce 100-tonners, that by itself might have been enough to put a cap on the size of the largest sauropods.

Another possibility is that max-size adult sauropods were neither common nor the target of selection. In most populations most of the time, the largest individuals might have been reproductively active but skeletally-immature and still-growing subadults (keep in mind that category would encompass most mounted sauropod skeletons, including the mounted brachiosaurs in Chicago and Berlin). If such individuals were the primary targets of selection, and they were selected for a balance of reproductive output and growth, then the few max-size adults might represent the relatively rare instances in which the developmental program “overshot” the selection target.

Dave Hone and Andy Farke and I mentioned this briefly in our 2016 paper, and it’s come up here on the blog several times before, but I still have a hard time wrapping my head around what that would mean. Maybe the max-size adults don’t represent the selective optimum, but rather beneficial traits carried to extreme ends by runaway development. It seems at least conceivable that the bodies of such animals might have been heavily loaded with morphological excrescences – like 15- to 17-meter necks – that were well past the selective optimum. As long as those features weren’t inherently fatal, they could possibly have been pretty darned inefficient, riding around on big predator-immune platforms that could walk for hundreds of kilometers and survive on garbage.

What does that swerve into weird-but-by-now-well-trod ground have to do with the limits on sauropod size? This: if max-size adults were not heavy selection targets, either because the focus of selection was on younger, reproductively-active subadults, or because they’d gotten so big that the only selection pressure that could really affect them was a continent-wide famine – or both – then they might not have gotten as big as they could have (i.e., never hit any internally-imposed, anatomical or biomechanical limits) because nothing external was pushing them to get any bigger than they already were.

Or maybe that’s just a big pile of arm-wavy BS. Let’s try tearing it down, and find out. The comment thread is open.


52 Responses to “What allowed sauropods to get big, and what kept them from getting any bigger?”

  1. David Hone Says:

    I agree that the Sander et al. idea is robust *for the very largest sauropods* but I’m inconvinced that it’s true for sauropods generally, or rather that sauropods were super special. As I noted in my paper on Shantungosaurus, while we’re lacking really robust estiamtes for the size of this thing at first approxiamtion it seems to be larger (well, heavier) than all but a handful of sauropod genera.So while I imagine their bauplan did permit the staggering sizes that they got to, the concept that the avergae sauropod was super special I think doesn’t work given that the a-pneumatic hadrosaurs made it up there in terms of mass. (OK, maybe Shantungosauurs was special but I think the point holds).

  2. LeeB Says:

    There are lots of interesting possibilities with sauropod (or for that matter hadrosaur) growth.
    If selection is for a tiny hatchling to grow up as fast as possible to out grow any predators then any increase in growth rate is likely to lead to further over-shooting the target size and a greater eventual maximum size.
    People have bred dogs in a wide range of sizes and the smaller breeds often live longer than the giant breeds; did this also apply to sauropods with the medium size ones living longer than the giant ones.
    This could then lead to counter balancing selection pressures; a pressure to grow up rapidly leading to giant maximum sizes against a selection for smaller adult sizes conferring a longer reproductive adult life but necessitating slower juvenile growth rates and thus higher juvenile predation rates.
    Also if they had indeterminate growth rates then the adults eventual sizes could to some extent be determined by the nutrition available to the juveniles; with extra well fed juveniles becoming very large adults.
    If the number of juveniles was low, perhaps due to a disease killing most of them then the few survivors may have grown up to be extraordinarily large; likewise if the adults were extending their range into new territory and their population density was low then the specimens at the edge of the range where the population density was low could grow up to be extra large.

  3. Mike Taylor Says:

    the few max-size adults might represent the relatively rare instances in which the developmental program “overshot” the selection target.

    Or they might have been the alpha jocks who got all the chicks.

  4. Mike Taylor Says:

    Come to think of it, do we have any reason to think most sauropod species might not have been like elephant seals, with a few males dwarfing the numerous and relatively tiny females? Might BYU 9024 be merely a male, but otherwise regular, Barosaurus?

  5. Matt Wedel Says:

    Come to think of it, do we have any reason to think most sauropod species might not have been like elephant seals, with a few males dwarfing the numerous and relatively tiny females?

    Yes, we have a good reason to think that sauropods were not like that: it’s pretty rare among mass egg-layers for males to be bigger than females, because egg production goes up very steeply with body size. That’s why I had second thoughts, mid-post, about the ability of 70-tonners to outbreed 100-tonners.

    In snapping turtles (Chelydra serpentina), small females might only lay a couple of dozen eggs a year, but big ones can hit 80. IIRC there is evidence that in some populations of snappers, females delay the onset of sexual reproduction for just that reason: the ones that start reproducing early never get as big as the ones that start late, because they divert so much of their energy budget into reproduction. And the late-bloomers are so much bigger, over the course of an average lifespan they can pump out many, many more eggs.

    …Then again, I wonder if that might not support your point after all. If snapping turtles who start reproducing early suffer a size penalty, and if essentially all sauropods (and dinos generally) were early reproducers, maybe the elephant seal scenario is not out of the question. Or at least maybe early reproduction was a sort of internal limit on sauropod max size.

    Regarding Shantungosaurus and kin, I am more inclined to think that the super-giant hadrosaurs were the special exceptions. AFAIK, we’re talking about a bare handful of genera that were not widespread either geographically or temporally. Whereas 20+ tonne sauropods were around for essentially the entire Jurassic and Cretaceous.

  6. David Hone Says:

    What do you count as a ‘mass’ egg layer? Alligators have one of the highers recorded levels of sexual dimorphism for reptiles and females often produce >30 eggs at a time.

    Agree the Shantungosaurus is slikely soem form of exception. But iirc, Sander et al. argued that what allowed sauropods to even get into that 20 ton+ range was exceptional and that doesn’t look like it can be the case if at least some hadrosaurs made it there too while lacking all of these supposed special advantages.

  7. arctometatarsus Says:

    I still think that Carrano & Janis have a key point: that the question “why did dinosaurs get so big compared to placental mammals?” is reversed, and the real issue is “why don’t placental mammals reach big dinosaurian size?” Here, issues like gestation period & litter size scaling with body mass and (re: Holliday) the negative allometry of mammalian articular cartilage may be keeping fuzzballs from reaching what should otherwise be “normal” giant size.

  8. Anonymous Says:

    “I’m sure someone could write a very compelling paper about how elephants are as big as they could possibly be, or Komodo dragons, if we didn’t have indricotheres and Megalania to show that the upper limit is elsewhere. This is basically what Economos (1981) did for indricotheres, either forgetting about sauropods or assuming they were all aquatic.”

    This is actually pretty common. Modern biologists (as in, study extant animals) tend to think the modern biota is the “best of all possible worlds” and extrapolate from that. Case in point, one oft used argument for gliding rather than volant flight in Pteranodon is “they were much larger than albatrosses, which represent the upper limit for any flying animal”, or in “Why Big Fierce Animals are Rare” it was said the tiger was the largest possible terrestrial carnivore and theropods got away with it by being sluggish obligate scavengers.

    “I admit, I don’t know enough reproductive biology to answer that. (If you do, speak up in the comments!)”

    I dooooo. Generally among egg-laying vertebrates, bigger adults produce a greater number of eggs per sitting. This is one of the reasons people are freaking out over the diminished adult size of many ocean fish.

  9. Anonymous Says:

    Sander et al. actually mention Shangtungosaurus. They note that Shangtungosaurus and other giant hadrosaurs and indricotheres are about the same size, and suggest that this reflects a cap on the size of these herbivores due to the fact that they engage in oral processing of food. Chewing surface increases by a factor of 2 as body mass increases by a factor of 3, resulting in disproportionately large chewing surfaces in larger taxa and disproportionately large heads. Sauropods, with their “binge-eating” mouths and mostly internal processing, don’t have these limitations and in fact their optima is much higher.

    With mammals, I don’t know if there is anything really keeping them from regularly reaching dinosaur size. Indricotheres did, and many creodonts and Andrewsarchus got to be big theropod size. Sauropod size may have been out, but hadrosaur/large theropod/ceratopsian/ankylosaur up to Camarasaurus-sized in the case of indricotheres may not be. It might be historical contingency and lack of hindsight. Mammals have only had half the time of dinosaurs to get huge. They got beaten over the head several times with rapid climatic changes that kept things from getting stable (not saying there were no changes in the Mesozoic, but the Cenozoic was weird). The period where they started to get big happened at the same time that the dominant vegetation shifted from leaves to grass, and while grass might have been enough to fuel big dinosaurs, most grass-eating mammals (i.e., ruminants, macropodids) also tend to not be gigantic for reasons and most of the lineages disposed to gigantism got wiped out. Also note the pile-up of megamammals in the last two million years. It’s possible that could have been the beginning of a size shift if the extinction hadn’t happened. Especially if the “sauropods got big and dragged everyone with them” idea holds water.

    It’s worth noting that most Permian and Triassic ecosystems didn’t produce giants on the scale of the later Mesozoic, and squamates, birds, and other terrestrial non-non-avian dinosaurs didn’t seem to get the metaphorical memo to go huge.

    If the cartilage thing is an issue, it is probably something unique to dinosaurs that was likely lost or is hidden in birds. Many neontologists and comparative anatomists believe cartilage “has” to be relatively thin because thick cartilage isn’t seen in anything today and there are good physiological reasons (avascular, can only get nutrients via diffusion) to support this. But if you take one look at dinosaur bones it’s clear they must have had cartilage caps much thicker than anything alive today, so it’s clear we don’t have the whole story.

  10. LeeB Says:

    If anyone suggests that albatrosses represent the upper flying limit even for birds alone I have two words in reply: Pelagornithidae and Argentavis.

    But yes a lot of modern biologists have no idea just what a depauperate world we live in with respect to large animals.
    Consider modern elephants and then compare them to Palaeoloxodon namadicus or Mammuthus trogontherii.
    The only case where modern creatures really do seem to represent a maximum is large whales where Balaenoptera and Balaena seem to have spent the last three million years or so getting really big; perhaps if people hadn’t come along they might have got bigger still..

  11. Matt Wedel Says:

    Many neontologists and comparative anatomists believe cartilage “has” to be relatively thin because thick cartilage isn’t seen in anything today and there are good physiological reasons (avascular, can only get nutrients via diffusion) to support this.

    Actually, even that’s not true: birds have vascular cartilage. Here’s an excerpt from Mallison (2010: p. 439) on the subject:

    Often, it is claimed that even large dinosaurs had only thin layers of articular cartilage, as seen in extant large mammals, because layers proportional to extant birds would have been too thick to be effectively supplied with nutrients from the synovial fluid. This argument is fallacious, because it assumes that a thick cartilage cap on a dinosaur long bone would have the same internal composition as the thin cap on a mammalian long bone. Mammals have a thin layer of hyaline cartilage only, but in birds the structure is more complex, with the hyaline cartilage underlain by thicker fibrous cartilage pervaded by numerous blood vessels (Graf et al. 1993: 114, fig. 2), so that nutrient transport is effected through blood vessels, not diffusion. This tissue can be scaled up to a thickness of several centimeters without problems.

    Mallison goes on to give an example of several-cm-thick cartilage in a dinosaur: the Dead Olecranon of Kentrosaurus. Figured in this post, with a longer excerpt from Heinrich’s paper – a paper which you can read for free in its entirety here:

    Mallison, H. 2010. The digital Plateosaurus II: An assessment of the range of motion of the limbs and vertebral column and of previous reconstructions using a digital skeletal mount. Acta Palaeontologica Polonica 55 (3): 433–458.

    To sum up, the “cartilage can’t be thick because it’s avascular” myth is now contradicted by at least two lines of evidence – cartilage could in fact be thick (shown by extinct dinos), because it can be vascular (shown by extant dinos).

  12. Marco Says:

    Sauropods, and dinosaur in general, seems to to follow a piramidal ontogeny series in terms of number of individuals.
    Milions of eggs, thousands of juvenile, hundreds of subadult sexually mature, and a few of truly giant super adults.
    The mirtality tax was proportionally inverse to the age, which is quite a paradox for us.
    Considering sauropod probably grown to the whole life (slowing down after sexual maturity) could the only Real limit be the life-span?

  13. Matt Wedel Says:

    There are a few sauropods out there that show an External Fundamental System, representing the cessation of growth. So at least some of them did not grow forever. Given the evidence for plasticity in growth strategies in dinosaurs in general and in sauropodomorphs specifically, I wouldn’t automatically extend that to all sauropods just yet.

    Still, life span might have been a limit, if not always the limit.

    I am intrigued by the convincing evidence that whales can live to be well over a century old. Longevity generally scales positively with body size. In the old days, people assumed sauropods had to live for decades or centuries just to grow to full size at reptilian rates. Now we know they grew much faster, but that doesn’t mean they didn’t survive for decades or centuries anyway. The idea of a Bruhathkayosaurus-size female spamming the environment with a couple thousand eggs a year, for centuries, is something that we are probably poorly equipped to even think about, given the very different extant systems that most of our life history and ecology are based on.

  14. LeeB Says:

    The nearest we probably get to it is with marine turtles which grow big and live long.
    The historical accounts of the numbers of marine turtles in the caribbean when european sailors first arrived there are interesting; you could navigate to islands by following the turtles heading there to breed.
    The numbers of juveniles hatching on beaches before human predation must have been astronomical.

  15. Warren Says:

    “The idea of a Bruhathkayosaurus-size female spamming the environment with a couple thousand eggs a year, for centuries, is something that we are probably poorly equipped to even think about”

    I imagine a fairly small population of adults keeping entire predator guilds running. It begins to feel less like modern mass-gathering spawners and more like… something out of a Miyazaki film.

  16. Matt Wedel Says:

    LeeB, that’s a strong point about the nearly-empty world in which we live – and from which we derive what we think we know about ecology.

    Warren – exactly. We reach for these modern analogues that capture some aspect of sauropod biology, like whales for size or turtles for reproductive strategy. But sauropods were like terrestrial whales that reproduced like turtles, which is way outside of our experience. Or maybe elephant/giraffe/whale/birds that reproduced like turtles. That’s a lot of potentially conflicting biology to try to hold in one’s head while thinking about sauropods – leaving aside the high likelihood that none of those models are all that exact, because sauropods were doing unique things that we perceive only dimly if at all.

  17. Andreas Johansson Says:

    Re: predator immunity, we do have lions on record taking down adult elephants, despite being less than 1/10 their weight, so automatically assuming mega-sauropods being safe from predators seem a bit bold to me.

    Now, obviously, there’s scaling issues – presumably a multi-tonne theropod is more worried about falling than a 200 kg lion, for one – and even if it happened sometimes it might no matter much in terms of evolutionary pressure, but still.

  18. Matt Wedel Says:

    Yeah, I know about the elephant-eating lions and have mentioned them on the blog before (here and here, for example). I’m not convinced that they have much to tell us about theropod-sauropod interactions except that animals will attempt anything when they’re hungry enough. A few caveats to keep in mind:

    – AFAIK the adult elephant-killing is restricted to just a few prides – it’s not a common lion behavior;
    – even those prides only do it when they’re desperate;
    – even then, they don’t usually target the biggest elephants (although I know they can and do occasionally take down big bulls);
    – it requires a fair amount of coordination, and lions are super-geniuses compared to any of the multi-ton non-avian theropods.
    – given the juvenile- and subadult- heavy age structure of sauropod populations, big theropods had to walk past a lot of winnable fights and achievable calories to find and take on max-size sauropods. Unless maybe there was a bad famine on and all of the smaller animals had already died.

    Now, with all of those caveats in mind, I am certain that in the long span of Mesozoic history, at some point one or more 12- or 15-ton world-record theropods took on a 70- to 100-ton sauropod, and maybe even won. I think the questions to ask are:

    1. Did such encounters end in success for the theropods often enough that they exerted a selection pressure on the sauropods, or was it more of a negative selection pressure on overly-ambitious theropods?
    2. Even if those encounters exerted more of a selection pressure on the sauropods, did they happen often enough for that selection pressure to matter?

    Still, I accept that I should have described max-size sauropods as effectively predator-immune instead of truly predator-immune.

  19. Dale McInnes Says:

    Matt. I don’t agree with your statement about lion prides pulling down elephants when they’re “desperate”. Lions are not only opportunistic but some prides prey almost exclusively on elephants. They learn the behavior of their prey and adapt accordingly. Remember. Prey expends most of their energy trying to escape not turning and attacking to kill as many lions as possible. Numbers count more than size. Once a pride figures out how to take one down … it’s on the menu. It is far more dangerous taking down a water buffalo. And a rhino will in fact turn on a pride to run down as many as it can. You don’t see any lion prides specializing on rhino for the above reason. Of course juveniles are going to be the main target. Adult elephants will rarely stand and fight a pride with exceedingly large numbers (even with calf in tow). I am not certain of this next statement but aren’t those lions that specialize in elephants a little larger as individuals in comparison to other prides? It would be very interesting to know that! Wouldn’t it?

  20. Ronald Says:

    What a fascinating discussion on this awesome website! Easily the most relevant discussion of this century so far (together with possible solutions to the Fermi Paradox of course).

    I often say that, for something to evolve, there has to be both the *need* for it (selection pressures) and the *possibility* (what you call ecological opportunities, I like that).

    I just do not agree with the sharp distinction that you make between external and internal factors, when it comes to limitations. I would rather think that limitations (or put differently, the limit to an opportunity) are usually a combination of both. Think for instance of the max. size of insects, which is dictated by their respiratory system (trachea) and the atmospheric pressure plus O2 content.

    With regard to sauropods, I read 3 main reasons in the post and comments, or categories of reasons, for sauropod gigantism:
    – Predators.
    – Feeding.
    – Reproduction.

    Though it is often, and probably here also, a combination of factors, I tend to agree with Anonymous (January 12, 2018 at 5:33 pm) with his strong emphasis on typical sauropod feeding (binge-eating, internal processing) as the main opportunity.

    And reproduction among the true gigapods, I hardly dare say it, maybe even as a limiting factor…

    I am somewhat puzzled that nobody so far brings up the issue of egg size and hatchling size: large egg-laying animals have relatively small eggs and hatchlings (i.e. relative to the mother’s body size). For instance, even medium-sized hadrosaurs and smaller sauropods (Hypselosaurus, Ampelosaurus) had eggs that were only about 1/1000th the weight of the mother. For the biggest sauropods this ratio must have been ridiculous. Particularly, if there is a max. limit to egg size.
    Wouldn’t this have posed problems with regard to the egg laying itself, the tending/caring (trampling!), long maturation times (even with rapid growth rates)?
    I can hardly imagine an Argentino-, Alamo-, Puerta-, Bruhathkayo-, etc., laying puny 10 kg eggs (about the biggest dino eggs found), even if the mother was ‘only’ 30-50 tonnes.

    I had a crazy thought: maybe the biggest sauropods weren’t really egg laying anymore, but ovoviviparous, to allow their offspring to grow to bigger size before birth. Just an idea.

    What are the biggest dinosaurs for which eggs are known?

  21. Marja Erwin Says:

    I’m not a paleontologist, but I think the consensus is that sauropods, unlike birds, crocs, and most other archosaurs, abandoned their young to fend for themselves, like turtles do.

    Now while smaller sauropods stopped growing, is it developmentally possible that some of the larger ones lost the signals to stop growing?

  22. Allen Hazen Says:

    Ronald’s suggestions are worth thinking about. The only thoughts I have are negative, and I’ll put them out, but I certainly don’t want to seem to be trying to shut down discussion of his ideas!
    (i) Lots of squamates are live-bearers, but I don’t know of any archosaurian (or turtle) examples. I don’t know why, but maybe there is some genetic reason why archosaurs are “pre-in-disposed” to evolving live-bearing?
    ((i.a) If it were EASY to evolve live-bearing, I would have thought it would have been a great thing for sea turtles. Given what happens to eggs buried on beaches, and to baby turtles crawling from the nest to the sea, I’d think a mutant strain of Dermochelys in which the females retained the eggs for a month and then released swimming neonates into the sea would have major selective advantage!)

    (ii) The largest terrestrial mammals tend to have precocial young. (The newborns have to be able to follow the herd on their own legs within hours of birth.) I recall seeing somewhere that the stress of carrying a humongous full-term foetus internally might be an important factor in limiting the size of terrestrial mammals to mere mammoth and indricothere size.

  23. Matt Wedel Says:

    Thanks, all, for the interesting discussion. Dale, thanks in particular for setting me straight on the elephant-hunting lions. Clearly I need to go do some actual reading on this if I’m going to be shooting my mouth off about it.

    I had a crazy thought: maybe the biggest sauropods weren’t really egg laying anymore, but ovoviviparous, to allow their offspring to grow to bigger size before birth. Just an idea.

    Yep, an intuitively appealing idea, which was entertained semi-seriously for a long time. Now it’s dead, thoroughly buried under many, many lines of evidence. For the short version, see Darren’s Tet Zoo post on the topic, and for all the gory details, see this paper (the main points of which are covered in the Sander et al. 2011 paper cited and linked in the main post):

    Sander PM, Peitz C, Jackson FD, Chiappe LM. 2008. Upper Cretaceous titanosaur nesting sites and their implications for sauropod dinosaur reproductive biology. Palaeontographica A 284:69-107.

    What are the biggest dinosaurs for which eggs are known?

    Sauropods! From Argentina, Spain, France, and India at least, including eggs with embryos. Loads of literature on these – see those Sander et al. papers for starters.

    Before anyone goes there, it’s no good arguing that small sauropods were egg-layers but big ones were live-bearing. Given that all known archosaurs are egg-layers, viviparous sauropods would require extraordinary evidence. But the live-bearing sauropod hypothesis is just kind of a nifty idea with no supporting evidence that isn’t better explained by oviparity + gigantism. Plus, from bone histo it seems that big sauropods started reproducing when they weren’t all that big, which throws a wrench in things. The reasons why live-bearing would have been advantageous, and egg-laying supposedly disadvantageous, apply much less to Dicraeosaurus-sized individuals than to super-giants. So even the ‘logical’ push for viviparity falls apart in light of what we know of sauropod maturation.

  24. nwfonseca Says:

    It is strange that whales reach sizes as large or larger than super sauropods. What do whales (and sauropods) do different than their land lubber counterparts? Maybe the answer isn’t 100% physiological. Could it also have something to do with having the right environmental/circumstantial conditions? During the Mesozoic did sauropods have more access to the right environments/circumstances over time that allowed for the evolution of super size? Did land mammals not have the right conditions on land to allow them to reach super sauropod size over time? Did the oceans provide better conditions than land for mammals to attain “super size”? Maybe it isn’t that land mammals “can’t” attain super size but haven’t had the “right” combination of circumstances to make getting any larger worth the expenses of growing so large?

  25. Matt Wedel Says:

    I’m not surprised that whales get bigger than sauropods. We know now from digestibility studies that horsetails and pine needles are more nutritious than it might seem, but they’re still pretty poor fare compared to krill. Krill are little shots of protein and fat in a thin carbohydrate wrapper (chitin), so whales are basically eating bacon all the time. Given the vast difference in calorie density of whale chow vs sauropod chow, I’m surprised the mass gap isn’t bigger. Plus whales not having to support their bodies in a gravitational field seems like a pretty big constraint release.

    Could it also have something to do with having the right environmental/circumstantial conditions?

    Sander et al. (2011) dealt with this at length – the question of whether sauropods got big because of advantageous abiotic conditions or a novel combination of traits – and Sander et al. (2013) revisited that question in light of additional tests. Both times, they concluded that abiotic factors weren’t enough to explain sauropod gigantism. Now, people can accept that conclusion or not. But if not, they should say why not, not just generally, but specifically rebutting the arguments of Sander et al. (2011, 2013). Both papers are free to read, and IMHO pretty readable, and I encourage everyone who is interested in sauropod gigantism to read them.

  26. Ronald Says:

    Matt Wedel and Allen Hazen: thanks!
    (I read Darren Naish’s post and I will read Sander).

    Ok, so that idea of (giant) sauropod vivipary has been put to rest, I stand corrected and humbled and will bring this fallacy up no more.

    So, reality is that sauropods, even the biggest, probably had the typical archosaurian reproduction of many (relatively) small eggs, the r-select approach.

    However, that does then leave the issue of (max.) egg size and hatchling size, relative to adult size. How big could a sauropod egg (and hatchling) get, max.? And what impacts would that have on reproduction, survival, and ‘lifestyle’ (such as herding)?

    It would be interesting, even fascinating, to know, whether giant sauropod mums had the ‘fire-and-forget’ way of sea turtles or the more crocodilian approach, with maternal protection of the nest and even some care. I think the major difference is that sea turtles are (mostly) migratory and crocodylians are (mostly) not.

    It would also be interesting to know (and maybe this is known?) the smallest individual sauropod size found together with adult size.

  27. Ronald Says:

    With regard to herding behavior, I found some very interesting information on a population of Shantungosaurus, ok not a sauropod, but a duckbill of sauropod proportions. It is so interesting, because it is not so common to find such a large group of the same species together (I know of an even much larger congregation of Edmontosaurus/Anatotitan):


    Remarkable is the fact nearly all individuals are of approx. adult size, with only 3 or so subadults, and no juveniles.

  28. LeeB Says:

    Archosaurs having to lay eggs is probably the reason that Metriorhynchid crocodiles didn’t grow as big as mosasaurs; although the hips of metriorhynchids suggest that they would have been very poor at moving on land and digging nests.

    Whales of gigantic sizes are a recent development; it is only as conditions started to get cold as we entered the Pleistocene that whales much above 15m seem to have evolved; at the same time a lot of smaller baleen whales died out as did the large to gigantic sharks that fed on them.
    The attrition rate on small hatchling sauropods must have been hugely high but the adults could presumably have churned out vast numbers of eggs, even more so than turtles do.
    What would be interesting would be if they migrated to localised areas to lay their eggs and effectively saturated the local predators with eggs and young in such numbers that only a fraction could be eaten; the same strategy that ridley turtles do with their arribadas.

  29. Matt Wedel Says:

    What would be interesting would be if they migrated to localised areas to lay their eggs

    I believe there is some evidence of that from the big Argentinian nesting sites – lots of nests at different horizons indicating site fidelity over a loooong period of time.

  30. Dale McInnes Says:

    Actually, I’m quite surprised that super-sauropods AREN’T BIGGER than whales. It may not be nutrition that spiked sauropods into growing big. It may have been the sheer LACK of nutrition. The way Nature counters that is BUILD BIG FERMENTING GUTS. That’s a key innovation for BIG size. However, some good-sized land mammals have big fermenting guts also. But sauropods may have had another disadvantage which helped them grow EVEN BIGGER than mammals. It’s the teeth. One thing sauropods could never do is “CHEW THEIR CUD” to further aid in the digestive process. So if you can’t chew it, perhaps you would need a much BIGGER more COMPLEX group of fermenting chambers than would be necessary if you were a very large mammal that could actually chew properly. It might have given sauropods an edge over the largest Pachyderms and Indricotheres. It might also explain their entire evolutionary history towards whale-sized denizens. A lack of nutrition combined with the lack of proper masticatory tools. It would give them just a slight edge over the biggest mammals. One other thing. If this is even remotely correct, then MOST sauropod species should tend towards super-giant-ism. Just a thought.

  31. Andrew Stuck Says:

    Good point on the relatively recent size increase in whales, LeeB. If I’m not mistaken, isn’t the temperature change associated with this phenomenon also associated with a sudden massive increase in planktonic productivity? Or is that merely the hypothesized connection between the temperature change and the size increase?

  32. LeeB Says:

    Have a look at the paper here: http://rspb.royalsocietypublishing.org/content/284/1855/20170546
    It discusses how baleen whale lineages suddenly got large, and why.

  33. Matt Wedel Says:

    One thing sauropods could never do is “CHEW THEIR CUD” to further aid in the digestive process. So if you can’t chew it, perhaps you would need a much BIGGER more COMPLEX group of fermenting chambers than would be necessary if you were a very large mammal that could actually chew properly.

    You don’t necessarily need more complex fermenting chambers, you just need a longer retention time. Basically, for a given digestive efficiency, particle size x retention time is a constant. If you retain food longer, you can get away with swallowing larger particles. Somewhere around the 3-day mark, which is how long it would have taken food to get through a medium-sized sauropod, the effective particle size is up to the size of a leaf, and you can stop chewing entirely with no loss of digestive efficiency.

    I hate to sound like a broken record, but this is another thing covered by Sander et al. (2011), and established in the literature even earlier.

  34. Anonymous Says:

    I don’t think there’s really any good reason to suspect that metriorhynchids crawled out of the water like sea turtles did. Currently the only evidence in favor of it is that they are archosaurs and all other archosaurs are known to lay hard-shelled eggs. Mosasaurs and plesiosaurs were both thought to have also crawled out of the water to lay eggs, until we started finding embryos in their body cavities.

    Right now we have no direct evidence at all of how metriorhynchids gave birth. No embryos that would indicate viviparity or ovoviviparity, no eggs from a gravid female that would support oviparity. We can’t even use medullary bone, since crocodilians and their relatives don’t seem to have it.

    At the same time, there’s a lot of indirect evidence that there was something weird going on with metriorhynchids. Their limbs are a lot smaller and a lot more specialized for marine life than sea turtles, more like marine reptiles we know were obligately aquatic like plesiosaurs and mosasaurs, and at least one study (Herrera et al. 2017, Earth and Environmental Science Transactions of the Royal Society of Edinburgh) found their pelves to more closely resemble viviparous reptiles than egg-laying ones.

    On top of that it’s worth noting that no other marine crocodyliformes ever really became as specialized for marine life as metriorhynchids. Dyrosaurs and Terminonaris seem to have remained relatively conservative, suggesting that metriorhynchids may very well have been the exception that proves the rule. Of course, in the absence of direct evidence, it’s impossible to say with any confidence how metriorhynchids reproduced.

    The fact that could just be due to the fact that there were a lot of other big marine reptile around during the same time as metriorhynchids filling those niches. If pliosaurs or something similar survived to the end of the Cretaceous and kept mosasaurs from growing into Tylosaurus-sized animals, and at the same time we never discovered embryonic mosasaurs, would we be debating if mosasaurs laid hard-shelled eggs on land? Right now it doesn’t seem possible to disentangle the factors.

  35. Anonymous Says:

    The evidence from Auca Mahuevo seems to suggest a “fire and forget” strategy of reproduction. I think Fowler had a paper that suggested laying sauropods actually made multiple nests per season to prevent (literally) putting all their eggs in one basket. Which is especially odd given that there is evidence that Massospondylus did not practice this type of reproductive behavior, something some have suggested may indicate that “turtle-style” reproduction is a derived trait in sauropods. It wouldn’t be surprising given how diverse, say, avian reproductive behavior is today. The SV-POWsketeers would know better than I would, but I think there was also something about Hypselosaurus possibly having a slightly different reproductive style than mainland sauropods.

    Hadrosaurian social behavior is weird. In Edmontosaurus and Gryposaurus, there are known examples of herds containing mixed-age groups of 2-8 years old (the oldest being adults), but yearlings are pointedly absent. It’s also hard to say whether a given composition of a herd represents a normal aggregation of animals or could be influenced by taphonomy or circumstance.

    @more generally
    One thing I would really like to know is exactly how dinosaurs compensated for the growth issue during mating. It is becoming increasingly apparent that dinosaurs exhibited a high degree of intraspecific variation, especially in growth, with some individuals stopping growth at smaller sizes and others growing much larger. At the same time, many dinosaurs became sexually mature far earlier than they reached adult size.

    This begs the question of how dinosaurs mated. In general it’s hard enough for most animals to find a mate of their own species, but now dinosaurs have to worry about whether a mate is of an appropriate size that they can mate. If not, there are issues of the male crushing the female during or being unable to physically inseminate her. This would effectively reduce the effective mating pool for any given individual to a fraction of the actual sexually mature population. I think John Long made a convincing argument for dinosaurs having particularly flexible genitalia: in general, animals with stiffer bodies (e.g., turtles) and have less room for error when mating (e.g., large animals like elephants) seem to have more prehensile genitalia. But even that wouldn’t completely solve the issue.

  36. Dale McInnes Says:

    Hmmm. Just a thought Anonymous. Couldn’t metriorhynchids perhaps like some serpents, still have been egg-layers but simply hatched them internally?

  37. William Miller Says:

    I’m not sure I understand this part of the post –

    ” if max-size adults were not heavy selection targets, […] or because they’d gotten so big that the only selection pressure that could really affect them was a continent-wide famine –”

    Even if a 50 ton female *Argentinosaurus* was already immune to predators etc., if a 75 ton female of the same species would be laying more eggs, wouldn’t that in itself still be a selective pressure to get larger and larger?

  38. Anonymous Says:

    Yes, that’s ovoviviparity. However, the general argument for archosaurs being egg layers is that all archosaurs for which eggs are known (including birds, several groups of non-avian dinosaurs, and several extinct in addition to extant crocodyliformes including one notosuchian) have hard-shelled eggs. Hard-shelled eggs permit less gas exchange than leathery eggs or just straight live birth because of the thicker shell. Retaining a hard-shelled egg inside the body would be a death sentence for any offspring, hence the argument why there are no viviparous or oviviparous birds. Pretty much all ovoviviparous reptiles have leathery eggs and there doesn’t appear go be a transition from hard-shelled egg to live birth without going through a leathery egg state, at least as far as I know.

    Actually, as I was writing this, I just realized. Pterosaurs have leathery eggs, don’t they? Which means hard-shelled eggs either aren’t plesiomorphic for archosaurs (as some have argued), or it’s possible to go from having a hard shell to a leathery one. And looking it up, it looks like at least one notosuchian has leathery eggs and live birth is known in the non-archosaurian archosauriform Dinocephalosaurus. This is especially interesting given that there have been some suggestiom that thallatosuchians are really basal on the Crocodyliformes tree (though I don’t know if that hypothesis is still supported.

  39. Matt Wedel Says:


    Even if a 50 ton female *Argentinosaurus* was already immune to predators etc., if a 75 ton female of the same species would be laying more eggs, wouldn’t that in itself still be a selective pressure to get larger and larger?

    Maybe! This is where my knowledge runs out. That’s exactly the question I was wrestling with in the paragraph that starts, “If there was reproductive competition among the super-giants…” Would the 75 ton female make enough more eggs over her lifespan to beat the 50-tonner, considering that the 50-tonner started reproducing a few years earlier and could plow the equivalent of those 25 extra tons into making eggs?

    I suppose it depends on a couple of unknowns: just how fast eggs per year scaled up with body mass, and life expectancy of adult females. So maybe we’ll never be able to answer the question for certain in sauropods. But I’m hoping that someone will come along who knows enough about how this works in sea turtles or ratites or what have you to tell us if one outcome is more likely than the other.

  40. LeeB Says:

    Another interesting thing is that birds and dinosaurs get the calcium for their egg shells from medullary bone, but crocodilians get it from their osteoderms. And metriorhynchids of course have lost their osteoderms; so they either had another way of getting calcium for egg shells or they produced soft shelled eggs. Of course if the latter there then arises the question of where the embryonic metriorhynchid crocodilian got the calcium to form it’s bones and teeth.

  41. William Miller Says:

    @Matt Wedel:”considering that the 50-tonner started reproducing a few years earlier”

    Yeah, that might make it advantageous to grow more slowly (putting energy into reproduction vs further growth).

    Hmm, I guess the more fundamental question is, why do egg-laying animals have a maximum size (determinate growth) at all? Does it have to do with the body plan not scaling up appropriately, like how the tallest humans have had various size-related problems?

  42. Matt Wedel Says:

    This is an area of the literature that I have NOT kept up with very well since grad school. But my half-remembered impression is that the popular idea of “indeterminate growth” is not a great description of what really happens, in that even animals that supposedly have “indeterminate growth” still pretty much level off at some asymptotic size, and some even form external fundamental systems on the outsides of their bones, which indicate that growth has truly ceased. Here’s a relevant bit from the Introduction of Woodward et al. (2011: p. 339), which appears to be a free download here:

    “Numerous observational ecological studies suggest that some reptile taxa do become skeletally mature, such as gartersnakes (Bronikowski and Arnold, 1999), Green Anoles (Lailvaux et al., 2004), Blanding’s Turtle (Congdon et al., 2001), and even alligators (Chabreck and Joanen, 1979; Wilkinson and Rhodes, 1997; Wilkinson, 2008). In fact, some lepidosaurs are known to possess secondary centers of ossification in their long bone epiphyses similar to those of mammals. These secondary centers fuse to the bone shaft at the metaphysis in mature individuals, effectively ceasing growth in length, although there is a range of individual maximum sizes for any particular taxon (Andrews, 1982). Secondary centers of ossification also develop in the tuatara (Andrews, 1982) after skeletal maturity is achieved within 15 to 25 years (Castanet et al., 1988).

    “Despite these lines of evidence, references to the indeterminate growth of reptiles can still be found in recent literature (e.g., Kozlowski, 1996; Olsson and Shine, 1996; Bronikowski and Arnold, 1999; Heino and Kaitala, 1999; Moore et al., 2000; Charnov et al., 2001; Chinsamy-Turan, 2005; Sparkman et al., 2007; Stamps, 2007; Tumarkin-Deratzian, 2007). The idea that reptiles have indeterminate growth may be perpetuated by the long lifespan of many taxa and the ability to adjust growth rates to changing environmental conditions, but it is important to establish which reptile groups are capable of attaining skeletal maturity as this affects how growth rates and other aspects of physiology are modeled. The current study reports on an external fundamental system within another reptile, the American Alligator (Alligator mississippiensis).”

    So possibly the question is, to what extent is indeterminate growth in animals a real thing at all? Even in animals that really never quite stop growing, the size levels off. We don’t see sofa-sized mussels, for example.

  43. Just want to jump in here about archosaur reproduction being limited to eggs. I was going to bring up metriorhynchids, but everybody else already has. But in the last couple years, the marine “protorosaur” (I’m not sure if that group is real anymore) Dinocephalosaurus was revealed to be a live-birther. Now, Dinocephalosaurus is an archosauriform, not an archosaur proper, but I think it demonstrates at least some reproductive plasticity among ruling reptiles.

    Somebody up there wondered why sea turtles haven’t figured out live-birth. One reason must be because their bony shells would prevent the body cavity from expanding to include growing embryos.

  44. Kevin Dann Says:

    This is all very instructive but doesn’t this speculative discussion leave out a much more critical question: how is it that the big dinosaurs completely violate Galileo’s Square-Cube Law?

  45. Matt Wedel Says:

    I wasn’t aware that they did. Explain please.

  46. Dale McInnes Says:

    Nature doesn’t violate its own Laws. You are simply applying the Galileos Square-Cube Law for objects that do not change their 3-D density as they get larger. Everything changes in a bio-system that gets exponentially larger. As dimensions increase … internal structures change abruptly (mostly density and strength to compensate). Therefore, a 60′ sauropod is not the same as a 90′ sauropod of the same species. More than ontogeny is at work here. You can have fifty complete skeletons of a 50′ sauropod but don’t think that finding a 100′ sauropod of the same species isn’t worth a paper. It will tell you a lot that a 50′ specimen can’t. It’s a completely different animal OF THE SAME SPECIES.

  47. […] and also about the impact of Don Glut’s New Dinosaur Dictionary when I was a kid. A post on sauropod gigantism sparked a very active discussion that ran to 47 comments, which is a rarity these […]

  48. […] Тейлор в своем блоге Sauropod Vertebra Picture of the Week рассуждает о том, что позволило завроподам стать такими […]

  49. Brad Lichtenstein Says:

    Ok, I was just pointed to this thread, is why I’m about four years late to the party. A couple of non-paleontologist remarks:

    1. I was going to point out that, just because mammals don’t vascularize their cartilage, doesn’t mean others don’t – I was quite happy to see there’s good evidence for that.

    2. In times of famine, I was under the impression it was advantageous to be small, meaning you didn’t need many calories to survive – like what survived the K-T event. But is the opposite true for short famines that don’t completely collapse the food web? Either makes sense to me. I guess young humans don’t do well in famines, what with their obligatory growth….

    2.5. …but goldfish don’t quite have this obligatory growth, they expand to fit their pond, and/or their food supply – why not other …uh, vertebrates… so in times of plenty, the giants got super-big, others they didn’t. Or, we know that apparent famine (usually any weight loss regimen) to a newly-pregnant human epigenetically triggers lifelong preference for calorie retention in the future newborn, and the opposite for a mother with plenty of food – there’s lots of ways species can change “target size”, fwiw.

    3. Several people mentioned the effects of humans on the modern world – archaeologists have long understood that humans have driven the most prolonged megafauna extinction known. Larger animals are such a cornucopia for most any size human hunter gatherer community (probably no larger than 150, partly for other reasons than food), that any megafauna that survived climate changes were preferentially hunted to extinction. No idea how that applies to turtles; human habitation would generally disrupt turtle beaches whether or not they were food – but it is likely people made them food.

    4. Metriorhynchoidea… since everyone kept talking about them, I read their brief Wikipedia page. Apparently someone has found an embryo in one; no idea when that discovery was made known, but it seems to contradict whoever said there was no direct evidence.

    5. Any new conclusions in the last 4 years – lol, or is this all “settled science”?

    6. Thanks for everyone’s thoughts 4 years ago, and really, no need to rehash stuff for me, or even answer at all. My own dang fault I don’t have the time to read the papers you all point to. Fun to learn how the wider world works!

  50. Matt Wedel Says:

    It’s not so much that goldfish expand to fit their pond, as that they can arrest their growth in a small container. Otherwise they max out at a little over a foot, like most other wild carp. I suspect that goldfish tell us much less about vertebrate growth and development than one would think from how often they come up in these kinds of discussions.

    Any new conclusions in the last 4 years – lol, or is this all “settled science”?

    Well, since the post is mostly questions, ‘maybe’ statements, and frank admissions of ignorance, I’m not sure which part could be considered settled. But if someone has provided definitive answers to what limited the max size of sauropods, I haven’t heard about it.

  51. llewelly Says:

    I think some of the tangential issues that came up in the thread have changed. For example, recent evidence indicates the three major groups of dinosaurs – theropods, sauropodomorphs, and ornithiscians – evolved their hard shelled eggs independently, and their common ancestor had a leathery egg. That implies that one of the oft-used used arguments against live birth in fossil archosaurs doesn’t apply to at least some archosaurs. For metriorhynchids, that matters. But for sauropods, it doesn’t change anything; the big sauropods all post date the evolution of hard shelled eggs.

    And the lost Amphicoelias fragillimus was revised to a rebachiosaur in the new genus Maraapunisaurus later in 2018 by Ken Carpenter, making it less enormous (but nonetheless twice as tall at the hips as a 4 meter giraffe … bigger giraffes were not used in this comparison … : )

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: