The sauropod viviparity meme

March 1992 Omni
ScienceBlogs.com: The sauropod viviparity meme
GO TO THE ORIGINAL SITE TO SEE THE ARTWORK THE AUTHOR, DARREN NAISH, REFERENCES IN THE ARTICLE.

Fossils demonstrate beyond any doubt that Mesozoic dinosaurs laid eggs, as of course do all dinosaurs today. But back during the 1960s, 70s and 80s - back when Robert Bakker and his idea about dinosaur biology were regularly featured in magazines and other popular sources - the scientific community was (sarcasm alert) delighted and enthralled by Bakker's proposal of sauropod viviparity.

Yes, mostly forgotten today is the idea that sauropod mothers were once suggested to give birth to a single, live, proportionally large baby, and to then engage in protracted parental care.

While Bakker is typically associated with this idea, it seems to have first been suggested by William Diller Matthew in 1910. Like many other palaeontologists of the time, Matthew assumed (very wrongly) that sauropods were amphibious or aquatic, and it was within this context that he proposed the possibility of sauropod viviparity (Matthew 1910). The thinking here was that viviparity would eliminate the need to lay eggs on land and hence allow a fully aquatic lifestyle. Copious evidence now shows that sauropods were strongly terrestrial - at best being about as 'aquatic' as are rhinos or elephants - and thus Matthew's proposal can be completely ignored.

Bakker's hypothesis was made within the paradigm of sauropod terrestriality. Let's note from the start that there's nothing obviously, inherently 'wrong' with his idea: viviparity, and substantial maternal investment in juvenile body size and parental care have all evolved elsewhere in non-avian reptiles. Nevertheless, the idea that sauropods were K-selected, elephant-style breeders with a very low reproductive turnover is totally at odds with the bulk of our knowledge on dinosaur biology, much of which indicates that dinosaurs were r-selected animals with large clutches, high juvenile mortality and perhaps little to no post-hatching parental care (e.g., Janis & Carrano 1992, Varricchio 2011).

More importantly, the support for the viviparous sauropod hypothesis was always tremendously poor. Three pieces of evidence were mentioned, all circumstantial and/or weak. So far as I can tell, none were put forward or evaluated in the technical literature: a few scant details are provided in Bakker's 1986 book The Dinosaur Heresies but the longest discussion seems to be in Adrian Desmond's 1975 The Hot Blooded Dinosaurs [the cover of the 1975 Blond & Briggs edition is shown here: quite possibly one of the worst dinosaur book covers ever]. Consultation of these sources reveals that the following proposals were used to support the concept of viviparity in sauropods:-

-- Bakker (1986, p. 357) mentioned a supposed 'newborn' Apatosaurus (estimated to weigh c. 230 kg), said to be too big to have hatched from an egg. His scaled diagram of this specimen - from Bakker (1986) - is shown below.
-- The pelvic canal in sauropods looks wide. Maybe, he said, this is because they gave birth to relatively big babies (Bakker 1986, p. 357).
-- Desmond (1975) noted that - due to physical constraints acting on the maximum sizes of shelled eggs - the babies of an oviparous sauropod would be tiny compared to adults. Such tiny animals wouldn't be able to "find their way to the herd", and their size would mean that they surely be crushed underfoot by their gigantic parents. Ergo, relatively big babies, born live in the middle of the herd (a 'reptilian elephant' model), were surely the answer (Desmond 1975, p. 132).


You don't have to be an expert on sauropods, or on dinosaurs or fossil reptiles of any sort, to show that these contentions are problematic. They neither demonstrate nor suggest the presence of viviparity. Let's go through them in turn:-

-- On the existence of the 'newborn': how do we know it's a 'newborn'? We don't. Other sauropod babies (read on) are substantially smaller than c. 230 kg, so it seems more parsimonious to assume that the c. 230 kg juvenile had been growing for some unknown postnatal period.
-- On the wide pelvic canal, this could be present because sauropods have really wide pelves (for reasons unconnected to the production of eggs and/or babies), and I'm not sure that sauropods do have particularly wide pelvic canals when compared with other oviparous reptiles anyway.
-- Finally, the behavioural scenario discussed by Desmond (1975) is entirely speculative and relies on the assumption that sauropod babies required 'herd protection'. We actually have good reason to think that baby sauropods (and other baby dinosaurs too) lived independently of adults (e.g., Varricchio 2011).

Another possible objection to the concept of sauropod viviparity is that archosaurs are unable to evolve this parturition strategy (Hopson 1977). That is, some people have argued that archosaurs are incapable of evolving viviparity: it's been said that developing archosaur embryos rely on the eggshell as a calcium reservoir and cannot therefore dispense with it (Hopson 1977, Packard et al. 1977, Tarsitano 1982), and also that limited oxygen exchange in the archosaur uterus prevents extended embryonic development (Andrews & Mathies 2000). Indeed, we don't know of any confirmed cases of viviparity in archosaurs, though its possible presence has been suggested at times for hesperornithine birds, metriorhynchoid crocodilians, pachycephalosaurian ornithischians and pterosaurs.

The definitive killer point on the whole viviparous sauropod idea comes from the discovery of numerous definitive sauropod eggs and embryos. It's true that many Cretaceous eggs traditionally identified as those of sauropods* have never been shown beyond doubt to belong to these dinosaurs, but there are now many, many others. At Auca Mahuevo in Argentina, "over a dozen in situ eggs and nearly 40 egg fragments encasing embryonic material were recovered" (Chiappe et al. 1998, p. 258). Six more Auca Mahuevo eggs - this time revealing well-preserved, near-complete skulls - were published a few years later (Chiappe et al. 2001, Salgado et al. 2005) [embryonic Auca Mahuevo sauropod skulls shown here, from Chiappe et al. (2001)]. More recently, sauropod hatchlings discovered right next to hatched and unhatched eggs (oh, and preserved with the large snake Sanajeh indicus as well) have been reported from the Upper Cretaceous of India (Wilson et al. 2010) (the snake was evidently preying on the babies).

* The French 'Hypselosaurus eggs' are the best case.

These eggs and embryos do only belong to one sauropod group (Titanosauria), but we also have eggs and embryos from close relatives of sauropods (e.g., Reisz et al. 2005) and from elsewhere in Saurischia and Dinosauria. The principle of parsimony demands that egg-laying must be assumed for all sauropods in the absence of compelling evidence to the contrary.

What drew me to write about this topic in the first place is the artwork. The best known picture depicting sauropod viviparity is Doug Henderson's painting - shown here - where a mother Apatosaurus is craning her neck round to look at her newborn baby. Unfortunately, what you see here is cropped. In the full-size version (which I don't seem to have anywhere*), what appears to be afterbirth is obviously present next to the baby (here, you can just see part of it overlapping the baby's tail). Incidentally, can you spot the technical error in the illustration? I mean, besides the brand new, just-born baby and all that. Yes, if you've spent years staring at the back ends of reptiles you'll note that the mother sauropod has been given a transverse cloacal slit, as if it's a squamate. In keeping with crocodilians, it should really have been given a longitudinally aligned slit.

The other painting - quite probably inspired by, or based on, Henderson's - is by John Bindon and appeared in a children's book called Dinosaur Mysteries (1989, Hamlyn), written by Mary O'Neill. It's shown here. The human is meant to be Robert Bakker.

Ok, so, I'm only aware of two relevant illustrations, but I always find it interesting when a minority scientific opinion becomes encapsulated in an illustration (other examples: non-avian theropods climbing trees, ceratopsians with frills embedded within their neck musculature). As others have noted, this can sometimes have the effect of 'reinforcing' a hypothesis - of making it look more plausible, or more widely accepted than it is, especially if the art is good. In reality, it's just art.

On that note, while you might think that the idea of sauropod viviparity is dead and now well off the radar, it kinda lives on. I say this because several artistic reconstructions - and a particularly well known North American museum mount - show an adult (presumably female) sauropod guarding or in some way interacting with a single, well-cared-for baby. Most viewers are going to assume from such works that sauropods exhibited elephant-like reproductive behaviour. Viviparity might not be assumed by a naïve observer, but it might. I would, anyway, go as far as saying that such reconstructions (which - don't get me wrong - I like very much) are flat-out misleading, and inaccurate. Even supposing that sauropods did practise extensive post-hatching parental care (note that there's no evidence for this), we know that they produced big clutches, so a dutiful parent wouldn't be looking after a single baby. Well, not unless all the other babies had been eaten.

And, yes, this whole article is indeed a spin-off of one brief aside in the April 1st mokele-mbembe article. I am amused to see that some cryptozoologists who support the existence of the mokele-mbembe are aware of the viviparity hypothesis and suggest that mokele-mbembe likely practises this behaviour too. Meh.