A spectacular new fossil provides insight on the sex lives of pterosaurs, part II: what it all means for eggs, nests and the behaviour of babies

i-a555e57e17e96fb92080637688648539-Mrs-T-Lü-Junchang--resized-May-2011.jpg

ResearchBlogging.org

In January 2011, Junchang Lü, David Unwin, Charles Deeming and colleagues published their Science paper on the amazing discovery of an egg-adult association in the Jurassic pterosaur Darwinopterus (Lü et al. 2011) [the specimen is shown here: image courtesy of Junchang Lü, Institute of Geology, Beijing, used with permission]. Darwinopterus is the incredible 'transitional pterosaur', first unveiled to the world in October 2009 and rapidly becoming one of the most important pterosaurs of all in terms of what we're learning from it.

As I discussed in a Tet Zoo article published in February, Lü et al.'s egg-adult discovery not only adds another confirmed pterosaur egg to the rather paltry global record, it also provides new data on pterosaur sexual dimorphism and, by inference, behaviour and biology. However, that wasn't the whole story, and here I complete what I started a few months ago. Why the delay? All will be explained below...

The Darwinopterus egg is, like the three other confirmed pterosaur eggs, soft-shelled: that is, with a parchment-like shell that was almost certainly highly permeable. The fact that (at least some) pterosaurs produced soft-shelled eggs like most squamates and some turtles might appear surprising given that crocodilians and dinosaurs - the archosaurian reptiles generally hypothesised to bracket pterosaurs in phylogeny - are typically thought to produce hard-shelled ones [in the highly simplified cladogram shown below, pterosaurs are shown in their 'conventional' position as crown-archosaurs close to dinosaurs. A position outside crown-Archosauria, but closer to Archosauria than to protorosaurs or rhynchosaurs, has been suggested by some and can be called the stem-archosaur or archosauriform hypothesis, while a position close to protorosaurs has also been mooted on occasion].

i-5c0db8781017b67d81dfa5b37a501a51-archosauromorph-cladogram-May-2011-490-px.jpg

However, the general impression that there is a simple dichotomy between 'soft-shelled eggs' and 'hard-shelled eggs' is surely a simplification, and intermediate forms were likely present in the stem lineages of both the crocodilian and dinosaur lineages. It's even possible that soft-shelled eggs were ancestral for archosaurs, with the hard-shelled eggs of crown-crocodilians and most dinosaurs being independently evolved. Or maybe hard-shelled eggs were equally as common, or more common, than soft-shelled eggs within Pterosauria... it's not as if we have a compelling sample to say either way. Or maybe pterosaurs aren't crown-archosaurs but belong elsewhere within diapsid phylogeny. Of the four confirmed pterosaur eggs, the Pterodaustro egg does have a thin mineral layer but it's very thin: about 30 times thinner than alligator eggshell (Unwin & Deeming 2008).

Anyway, we should assume for now that thin, flexible, parchment-like eggshells were normal for pterosaurs. This has major implications for pterosaur breeding biology. It probably means that adults weren't sitting on the eggs as they were likely too soft and too fragile (Grellet-Tinner et al. 2007). Ok, various snakes (including pythons, mudsnakes, pitvipers and so on) and lizards (notably the Eumeces skinks) coil their bodies around their soft-shelled eggs* (in some cases generating heat and hence incubating - and not just guarding - their eggs), so you can still be in close contact with your eggs and not put your weight directly on them. But, needless to say, it isn't easy to imagine how a pterosaur might do this sort of thing.

* In some snakes that coil around their eggs, the shells are so thin as to be transparent (example: Okinawa pitviper Ovophis okinavensis).

Furthermore, eggs with soft shells are typically highly porous and hence exhibit high vapour conductance values. That is, they're prone to desiccation if left exposed and need to be kept in a humid chamber, or be in close contact with moist substrate (in the aforementioned lizards and snakes that coil around their eggs, high humidity is maintained within a nest chamber). Given what we know about pterosaur eggshell, it therefore seems safe to conclude that pterosaurs buried their eggs, either in sediment or beneath or within vegetation (Unwin 2005, Grellet-Tinner et al. 2007, Unwin & Deeming 2008, Lü et al. 2011).

i-aeec512af92a1a462d3d1b8b2c7e191e-Witton-azhdarchid-hatchlings-May-2011.jpg

Pterosaur parents might have abandoned the eggs entirely, but nest guarding is at least possible, especially if pterosaurs are indeed crown-archosaurs and hence bracketed by crocodilians and birds, both of which stay with their nests (though, again, the possibility exists that nest-guarding is convergent in crown-crocs and birds and wasn't ubiquitous across Archosauria: we don't have much data on stem-crocs, and nest guarding has not been well demonstrated in non-coelurosaurian dinosaurs). Maybe - speculation alert - it was only the big pterosaurs that stayed with their nests, or only those species that nested in 'safe' localities such as offshore islands or near soda lakes. The possibility of nest guarding is depicted in this brand new reconstruction shown above by Mark Witton, kindly provided with permission.

Superprecocial pterosaur babies?

i-213be0d05202c715c8cbccdc577b6853-Juvenile-Pterodactylus-wikipedia-May-2011.jpg

Based on the anatomical proportions of unhatched pterosaur embryos, and on the skeletons of the numerous juvenile pterosaurs discovered in the Solnhofen Limestone and elsewhere, it's looking likely that pterosaurs were superprecocial. That is, that they were able to fly and run around, and generally look after themselves, right from hatching (Unwin 2005). This proposal is explored in depth in Unwin's 2005 book The Pterosaurs from Deep Time where he draws in data from the proportions and anatomy of embryos and young and very young juveniles to make a pretty convincing case. A sort of superprecociality is typical for lizards, turtles and crocodilians but, among birds, it's best known for megapodes. On hatching, baby megapodes dig their way out of the surrounding vegetation or sediment and are immediately able to run and, sometimes, fly. In some species, chicks are known to have made non-stop flights of over 35 km when just a few days old (Healey 1994). Juvenile pterosaurs weren't necessarily on par with this, but they might have been [the adjacent historical reconstruction of a juvenile Pterodactylus was produced by Harry Seeley for his 1901 book Dragons of the Air. Baby Pterodactylus specimens like this were originally thought to represent the separate taxon Ptenodracon].

Based on the large number of juvenile pterosaurs present in such deposits as the Solnhofen limestone, you might speculate that pterosaurs were r-selected and that juvenile mortality was high (we don't know anything about clutch size so this is - yet again - very speculative but the substantial number of preserved juveniles provides support for the contention). The juveniles of small species were tiny animals (wingspans less than 20 cm) and could well have fallen prey to larger pterosaurs, to theropods large and small, to big fish, lizards and so on.

i-6d2fbe7c29dd8238a3a08b8ba9218ae7-Bennett-1996-year-classes-Solnhofen-pterosaurs-May-2011-crop.jpg

Whatever, it now seems that precocial juvenile pterosaurs - David Unwin (2005) calls them flaplings - made up a reasonable proportion of at least some pterosaur populations, and furthermore that they acted as separate 'ecological species' from adults, foraging in different environments and exploiting different prey from older, bigger morphs of the same species (Unwin 2005). The same seems to have also been true of non-avialian dinosaurs and perhaps early birds as well. It might explain why adult mini-pterosaurs (and mini-dinosaurs) were relatively rare: each one species acted as three, four or five 'ecospecies' during its lifetime [in the adjacent figure - from Bennett (1996) - note the major shape difference between the juvenile specimens shown in E and G and the adults shown in F and H. Bennett (1996) argued that E ('Pterodactylus elegans') morphed into F ('Ctenochasma gracile')* and that G ('Pterodactylus micronyx') morphed into H ('Gnathosaurus subulatus')**. Clearly, juveniles could exploit different resources from adults. The scale is 2.5 cm for E and G, but 5 cm for the others].

* The final name for the species is Ctenochasma gracile according to Bennett (1996). It's since been synonymised with C. elegans (Bennett 2007).
** The final name for this species is Gnathosaurus subulatus (Bennett 1996).

Juvenile dinosaurs belonging to several lineages seem to have been social (e.g., Varricchio 2011), and juvenile social behaviour - even with siblings apparently looking after each other - is known for some lizards (see Amazing social life of the Green iguana). Accordingly, it seems reasonable to imagine social behaviour of some sort being practised in juvenile pterosaurs, though yet again I'm speculating and evidence would be nice...

i-1271e3c9a16482b66c20b537e0813262-Pteranodon-feeding-baby-Zdenek-Burian-1960-May-2011.jpg

Anyway, those old pictures that show adult pterosaurs bringing mouthfuls of food to flightless babies, sat with sprawled limbs on cliff-top nests (like this one, by ZdenÄk Burian), now look mightily inaccurate in view of the evidence. Then again, we can't pretend to know the full life history of all pterosaurs, and you can still hold out hope that some pterosaurs were like this, if you want to.

Does this tell us anything about the metabolic status of pterosaurs?

The obvious indication that pterosaur eggs were buried led Lü et al. (2011) to note a possible incongruity with previous ideas on pterosaur physiology. If pterosaurs were burying their eggs (and their eggs were thus developing at ambient temperature), is their style of reproduction inconsistent with homeothermic endothermy? It might be, though note that egg burial ALONE doesn't falsify the hypothesis of homeothermic endothermy in pterosaurs, since we know of living birds that bury their eggs in sediment.

i-27a82518f15af548fdc6bb542684f1da-Megapodius-reinwardt-Toby Hudson-wikipedia-May-2011.jpg

Here we come to the reason for the long delay in my posting of this article. I actually wrote a letter to Science on this subject, comparing what we know of megapodes (Macrocephalon maleo and some species of Megapodius in particular) with the pterosaur data [Orange-footed scrubfowl Megapodius reinwardt shown here by Toby Hudson, from wikipedia. Why are they called 'megapodes' again?]. The authors responded. I don't feel at liberty to discuss any of what happened in the ensuing correspondence; anyway, in the end I was contesting a single sentence in their paper and Science decided not to run with it. That's understandable.

Other factors in addition to egg burial need to be considered if we're interested in making physiological inferences from what we know about pterosaur eggs. Both relative egg mass and yolk quantity both come to mind.

We don't know anything about relative yolk quantity in any of the reported pterosaur eggs, but what does egg size tell us? By comparing the estimated mass of the Darwinopterus egg (c. 6 g) with the estimated mass of the female (somewhere between 110-220 g), Lü et al. (2011) noted that parental investment in egg mass looks squamate-like, rather than bird-like. However, because the parchment-like shell of the pterosaur egg would have allowed water uptake during incubation, they estimated the final mass for the Darwinopterus egg to be about 11 g, which puts it in the range estimated for a bird of this size. It isn't clear to me what this might means for physiology and maternal provisioning, especially given (1) that we know nothing about yolk volume, and (2) some birds fall into the squamate range anyway [the graph below, from Lü et al. (2011), shows how the estimated initial egg mass (that is, the mass of the egg before it has absorbed water during incubation) of Darwinopterus (black) and Pterodaustro (red) compares with birds (blue), squamates (purple), crocodilians (dark brown) and turtles (green). My colour vision is poor, so graphs like this are a nightmare].

i-ba05510950074f649e6c7f73a3868160-Lu-et-al-2011-pterosaur-egg-mass-compared-to-birds-and-squamates-May-2011.jpg

I'm not sure where we can go from here, other than admitting - as usual - that more work will help shed new light. Pterosaur eggshell structure provides strong evidence for egg burial, and hence for incubation at ambient temperature. And pterosaur babies quite probably were superprecocial and formed a relatively large component of any pterosaur population (Unwin 2005). Whether we can draw from this the conclusion that pterosaurs were 'reptile-like' in physiology seems problematic, though Lü et al. (2011) have a point in bringing attention to the fact that pterosaurs seem to have employed a 'reptilian' strategy as goes eggs and nesting. It's all exciting stuff and there's more work that needs doing.

This wraps up my two-part take on Lü et al.'s excellent 2011 paper. Thanks indeed to Junchang Lü and David Unwin for their help with images and data, and also to Mark Witton for the use of another world exclusive.

For part I on the Darwinopterus egg-adult association, see...

And for previous Tet Zoo articles on pterosaurs see...

Refs - -

Bennett, S. C. 1996. Year-classes of pterosaurs from the Solnhofen Limestone of Germany: taxonomic and systematic implications. Journal of Vertebrate Paleontology 16, 432-444.

- . 2007. A review of the pterosaur Ctenochasma: taxonomy and ontogeny. Neues Jahrbuch fur Geologie und Paläontologie, Abhandlungen 245, 23-31

Grellet-Tinner, G., Wroe. S., Thomson, M. B. & Ji, Q. 2007. A note on pterosaur nesting behaviour. Historical Biology 19, 273-277.

Healey, C. 1994. Dispersal of newly hatched orange-footed scrubfowl Megapodius reinwardt. Emu 94, 220-221.

Lü J, Unwin DM, Deeming DC, Jin X, Liu Y, & Ji Q (2011). An egg-adult association, gender, and reproduction in pterosaurs. Science (New York, N.Y.), 331 (6015), 321-4 PMID: 21252343.

Unwin, D. M. 2005. The Pterosaurs from Deep Time. New York, Pi Press.

- . & Deeming, D. C. 2008. Pterosaur eggshell structure and its implications for pterosaur reproductive biology. Zitteliana B 28, 199-207.

Varricchio, D. J. 2011. A distinct dinosaur life history? Historical Biology 23, 91-107.

Tags
Categories

More like this

In some species, chicks are known to have made non-stop flights of over 35 km when just a few days old

Non-stop flights of over 35 km?! That sounds astonishing to say the least - most galliforms (migratory quail excluded) aren't exactly known for their long-distance flight proclivities.

A biological species occupying several different eco-species roles throughout life also makes sense in the light of newish evidence of how dinosaur skeletons could change shape through life (Triceratops grew up to be Torosaurus at final adult stage). One thing which seems to have puzzled people for a long time is why the apparently vestigial forelimbs of tyrannosaurs were so well muscled; vestigial organs ought to be, well, vestigial.

It now looks like a tyrannosaur chick when hatched, was a cute ball of fluff with proportionately long limbs and a proportionately smaller body than the adult. The forelimbs at this stage would possibly have been easily long enough to serve a role in prey capture; the chick would start life as an insectivore, move on to a role as a meso-predator most probably preying on other juvenile dinosaurs, then as it grew bigger the body would enlarge much faster than the forelimbs, and somewhat faster than the hindlimbs.

This seems to be indicated by a find of a sub-adult T. rex, which had teeth markedly unlike those of the adult. The sub-adult teeth were much more blade-like, much more like those of a predator than the adult teeth, which are those of a scavenger (one which doesn't always wait for the prey to die first). I would predict that newly hatched tyrannosaur chicks would be equipped with fine, spiky teeth which would be replaced by more blade-like ones as it grew older, finally to be replaced by the ultra-robust adult dentition.

The key thing to remember here is that dinosaurian ecosystems are not going to look like mammalian ones or like reptilian ones; the dinosaurian one will be more mammal-like in that herbivores will greatly out-number predators, but reptile-like in that juveniles will massively outnumber adults. The breeding population would also have been skewed, since dinosaurs seem to have become sexually mature long before hitting adult size; adults would simply produce more offspring than subadults by virtue of being able in the case of females to make more eggs, and by being able to dominate mating in the case of males. Quite what this would do to population dynamics, I don't know.

By Dan Holdsworth (not verified) on 05 May 2011 #permalink

I'll throw in an alternative theory, from Peter Ward's Out of Thin Air. His idea was that bird-style respiration evolved in a low-oxygen world, and he has some theories about egg shells. Not being a paleontologist, I don't know whether the evidence supports him, but they're fun ideas.

To drastically simplify, his idea is that thinner eggs shells work better in low oxygen conditions. Dinosaurs? According to him, almost all dinosaur eggs are known from the Cretaceous (when oxygen levels were rising). The question is whether the dearth of Jurassic or especially Triassic dino eggs is a result of fewer fossils overall, or whether they fossilized more poorly because their shells were thinner.

Thus, one could posit that pterosaur eggs are thin due to function, not phylogeny. The problem pterosaurs face as flyers is that they probably had high inherent respiration rates. Thin egg-shells would be useful in that they helped the developing embryo to maintain a high respiration rate and develop more quickly. The upshot of this idea is that we can't necessarily deduce life-style from egg shell, at least by comparing with extant animals. We have to look at the mesozoic environment as well, because oxygen levels did vary, so far as we can tell. Burying eggs works well only when the embryos don't suffocate.

By heteromeles (not verified) on 05 May 2011 #permalink

Dartian: yes, a flight of AT LEAST 35 km for a megapode chick just a few days old was reported by Healey (1994). The chick landed on a ship called the Cendana: the nearest land was between 35 and 45 km away. The paper itself is available for free here.

Dan: the 'Triceratops morphs into Torosaurus' idea is far from widely accepted, and indeed some/most horned dinosaur specialists disagree with it. Most recently, Andy Farke contested it (Farke 2011). As for tyrannosaur ontogeny, there is good evidence for marked transformation in species such as T. rex and - if the specimens have been correctly assigned to taxa (you may be aware of the debate surrounding the status of Nanotyrannus*) - there is good evidence for a major shift in tooth morphology, especially in tyrannosaurines. See Carr (1999), Currie (2003) and Carr & Williamson (2004). You note that forelimbs were proportionally longer in juveniles; minor negative allometry was reported in the radius and ulna of some species by Currie (2003) but the humerus changed isometrically during growth. You might be remembering discussions of forelimb length in non-tyrannosaurid tyrannosauroids perhaps. By the way, it isn't clear that the abbreviated forelimbs of giant adult tyrant dinosaurs were vestigial - there are indications from pathologies and modelling that they were used extensively.

* See the previous discussion in 100 years of Tyrannosaurus rex.

Refs - -

Carr, T. D. 1999. Craniofacial ontogeny in Tyrannosauridae (Dinosauria, Coelurosauria). Journal of Vertebrate Paleontology 19, 497-520.

- . & Williamson, T. E. 2004. Diversity of late Maastrichtian Tyrannosauridae (Dinosauria: Theropoda) from western North America. Zoological Journal of the Linnean Society 142, 479-523.

Currie, P. J. 2003. Allometric growth in tyrannosaurids (Dinosauria: Theropoda) from the Upper Cretaceous of North America and Asia. Canadian Journal of Earth Sciences 40, 651-665.

Farke, A. A. 2011. Anatomy and taxonomic status of the chasmosaurine ceratopsid Nedoceratops hatcheri from the Upper Cretaceous Lance Formation of Wyoming, U.S.A. PLoS ONE 6(1): e16196. doi:10.1371/journal.pone.0016196

Is it not possible to infer typical brood size from the relative sizes of the egg and the adult? Given that the mother presumably still has to be capable of flight, there must be a limit - in terms of both mass and volume - to how many of those things could be carried inside her?

Thanks for a fascinating post.
I'm extremely dubious of any physiological inferences from egg size and shell type. There are just so many variables and so few clear patterns in extant amniotes, and that's to say nothing of the many arguments from spurious correlation that always pervade the topic of paleophysiology.

Thank you for showing the egg-size graph from Liu et al. (I hope the data are archived in the Supplementary zone). It is very poorly predictive even for those of us with 'normal' color vision. We can conclude that this particular species of pterosaur laid smaller eggs than most extant birds, but that's about it.

Turtles bury both parchment-shelled and hard-shelled eggs. The parchment-shelled eggs can (and apparently normally do) absorb water, but they can also lose water, and in any case the effect on hatchling size of differential water exchange is minor compared to initial egg size. The hard-shelled eggs are truly "cleidoic" and exchange very little water in or out during development (in a reasonably humid environment). AFAIK crocodilian eggs are intermediate--they are relatively hard-shelled but still experience significant water exchange.

If pterosaurs had parchment-shelled eggs, they were almost certainly buried, regardless of atmospheric oxygen content (except in clays, interstitial soil air is not hypoxic). But of course burial does not mean incubation at ambient environmental temperature, since megapodes and some at least some crocs build heat-generating nests in which to bury eggs.

Also there's no reason why egg burial precludes extensive post-hatching parental care.

Herman Rahn made a career of measuring the permeability of eggshells to gases, but they are all pretty high. If embryonic growth rate is limited by oxygen supply, then a much more important source of variation is the degree of vascularization of the chorion.

And growth rate of embryos has little to do with endo- or ectothermy of adults; all embryos are ectothermic, it's their thermal environments that differ.

So obviously it's a very very cool fossil-with-egg, but all we really learned about pterosaurs is that one of them buried its eggs.

By Sven DiMilo (not verified) on 05 May 2011 #permalink

Soft x Hard-shelled clades? Need names? Call them Molliovata and Durovata.

By J.S. Lopes (not verified) on 05 May 2011 #permalink

With regard to keeping the soft shelled eggs warm, don't you think that a perosaur "wing" would be big enough and light enough to cover them without destroying them. ?

By Alan Javel (not verified) on 05 May 2011 #permalink

Mike Simpson: I thought most birds and suchlike laid eggs over a long period before starting to brood, so regardless of the size of the clutch, they're not laying more than one at a time (for instance, domestic chickens will brood up to a dozen or so successfully, laying for a fortnight before starting to sit - this also means you can send unbrooded eggs by post, put them under a fostermother, and they'll develop normally).

By stripey_cat (not verified) on 05 May 2011 #permalink

"The fact that (at least some) pterosaurs produced soft-shelled eggs like most squamates and some turtles might appear surprising given that crocodilians and dinosaurs - the archosaurian reptiles generally hypothesised to bracket pterosaurs in phylogeny - are typically thought to produce hard-shelled ones"

I don't wanna be a douchey McNitpick, and most of my education in this area comes from nature documentaries (back when Discovery channel still showed good nature documentaries), but AFAIK crocodilians don't lay hardshelled eggs at all. Every video of hatching crocodilians I've seen, whatever the species, has showed them squirting out of leathery shells, not cracking their way out of hard ones.

Seems like superprecociality would be an enormous advantage to any egg-laying animal. Why would the atricial strategy come about at all? Also, this is something I've never thought about before, but where did dinosaurs fall on this superprecocial-atricial line? I know Horner likes his atricial duckbill idea, but hasn't that been questioned?

Basal pterosaurs are sometimes thought to have a slow metabolism. Rhamphorynchus is still argued to have a crocodilian like metabolism. Darren, are you sure that crocodilians have hard shells. I was positive they were leathery like squamates.

Changing econiches of different growth stages are well represented among many modern animals. For example in Komodo dragons. Juveniles still have very "conservative" varanid proportions, live mainly on trees and feed on small animals. When they grow bigger proportions change comparably drastically (I don´t have to tell you how adult komodo dragons look like of course), and synchronously the diet, which shifts to macropredation. In some other varanids similar changes are known, but not as drastically as in komodo dragons. In crocodylians something similar happens, a shift from very small prey which consists to a big degree of invertebrates, later much fish, and especially in the large and broader snouted species macropredation in the adult stage. But there are also examples from other vertebrates, for example many fish. Great white sharks change also from fish prey to mainly mammalian prey during their growth, and juveniles and adults don´t only differ in absolute size, but also in body proportions and tooth shape.

the adult teeth, which are those of a scavenger

Those of a bone-cruncher, you mean. Like a dog compared to a cat, or a hyena compared to a cheetah.

(one which doesn't always wait for the prey to die first).

No extant big carnivores pass up on a free lunch. The options are: "pure scavenger" (like most vultures) or "not" (like everything else). Contrary to what Horner has been implying for far too long, "pure predator" is not an option.

the 'Triceratops morphs into Torosaurus' idea is far from widely accepted, and indeed some/most horned dinosaur specialists disagree with it.

It's controversial, yes, and I bet that's why neither Denver Fowler nor John Scanella got the Romer Prize (750 $) at last year's meeting of the Society of Vertebrate Paleontology, even though they presented not only histological data to support this hypothesis, but also high-resolution stratigraphic data. No, T. and T. are not an anagenetic series.

By David MarjanoviÄ (not verified) on 05 May 2011 #permalink

Wasn't it here, a long explanation how bird and crocodilian embryos draw calcium from the eggshell, therefore there is no viviparous birds? Seems against the thin-shelled eggs of pterosaurs phylogenetically bracketed between. Or I messed something.

Many small, precocious juveniles suggest R-based life history, but small abdomen and carrying extra load in flight speak against producing large numbers of eggs by pterosaurs (unlike big dinosaurs).

Soda lake environment would likely suck water from a thin-shelled egg, kill the embryo, as happens here: http://en.wikipedia.org/wiki/Century_egg

Change from small flapping flyer to large soarer needs very big change of wing shape, muscle structure, aerodynamics etc. Maybe that's why large pteranodontids are said to grow to adult size in less than a year, and I suggest large soaring pterosaurs in general fed their chicks.

PS. I would also be interested in this megapode dispersal.

Fascinating stuff, Darren, and presented with your usual flair. One thing that drew my attention, though:

"Given what we know about pterosaur eggshell, it therefore seems safe to conclude that pterosaurs buried their eggs, either in sediment or beneath or within vegetation"

Burying a soft-shelled egg under vegetation or sediment strikes me as A Very Bad Idea. Crocodiles do it, true, but croc eggs are (as noted) hard-shelled. Likewise for megapodes and other mound-nesting birds. Parchment-shelled eggs seem tailor-made for open-air brooding by the mother, as in many snakes.

I have this Image in my head of a female pterosaur hunkered down on her belly, with a heap of eggs along each side, and a wing covering each heap. Plenty of warmth from her body (and presumably a warm ambient temperature), plenty of oxygen and moisture from the air, very little weight to risk crushing the eggs.

By wolfwalker (not verified) on 05 May 2011 #permalink

I think a theory that explains everything is that this fossil is a female with an egg in a pouch (warm, moist environment perfect for a lightweight thin-skinned egg). Babies flew off immediately after hatching. This eliminates the need for adults to ever land, and solves the paradox of larger species apparently being too heavy to take off.
(Everyone is free to write it up and submit to Nature or Science as long as I am listed as the corresponding author).

Jura: I stand corrected. However, the crocodilian egg still doesn't look like what I picture when I think of a 'hard shelled' egg. Definitely harder than the snake egg, but it still didn't look as hard as a bird's.

Darren:

yes, a flight of AT LEAST 35 km for a megapode chick just a few days old was reported by Healey (1994). The chick landed on a ship called the Cendana: the nearest land was between 35 and 45 km away. The paper itself is available for free here.

Wow, amazing! Thanks for the reference. The long-distance dispersal potential of megapodes is something I should have known about (considering this group's distribution pattern). Well, I know now.

Jerzy:

I would also be interested in this megapode dispersal.

Apart from checking out the paper that Darren linked to, also look up:

Olson, S.L. 1980. The significance of the distribution of the Megapodiidae. Emu 80, 21-24.

This eliminates the need for adults to ever land, and solves the paradox of larger species apparently being too heavy to take off.

This supposed paradox lies behind the Burian picture, which shows the adults feeding their young from the air. However, recent work by Jim Cunningham and Mike Habib (the latter has published) has completely eliminated it. Pterosaurs, Quetzalcoatlus included, simply didn't take off the bird way; instead, they used their wings to jump off the ground and surpass stalling speed before the first wingstroke. This works in today's atmosphere and without any wind.

By David MarjanoviÄ (not verified) on 05 May 2011 #permalink

David M stated: "No extant big carnivores pass up on a free lunch. The options are: "pure scavenger" (like most vultures) or "not" (like everything else). Contrary to what Horner has been implying for far too long, "pure predator" is not an option." - while I guess it depends on what you mean by 'big carnivore', some large snakes (pythons, king cobras etc) tend to be fairly close to being pure predators, and most wild specimens will usually refuse to take anything other than live prey (I have been told that some wild caught specimens will starve to death before eating dead prey provided to them). Captive bred specimens, or those caught while young are acclimatised to eating prey they didn't kill themselves, wild caught mature specimens tend to prove more difficult.

Having said that I think it is probably irrelevant to the specific question David was addressing, which was about the diet of adult tyranosaurs.

By Mark Lees (not verified) on 05 May 2011 #permalink

Dartian: yes, a flight of AT LEAST 35 km for a megapode chick just a few days old was reported by Healey (1994). The chick landed on a ship called the Cendana: the nearest land was between 35 and 45 km away.

Couldn't this be a case of a chick carried by strong winds? The fact that something ends up 35km off-shore from the air doesn't mean it flew there in fully powered flight per se, I would think.

As for the soft-shelled eggs, has anyone presented the idea of ovoviviparity? Or does the lack of a fossil embryo in the Darwinopterus rules that out?

By Kilian Hekhuis (not verified) on 06 May 2011 #permalink

Kilian:

Couldn't this be a case of a chick carried by strong winds?

There is no mention in the paper of any kind of unusual weather conditions.

Good question, Mike S - I bet you always figure out whodunnit in detective thrillers.

Parchment-shelled eggs seem tailor-made for open-air brooding by the mother, as in many snakes.

No. Brooding is quite rare in snakes, and usually occurs within a nest of leaf litter or similarly confined space. Exposed to 'open air' (unless humidity's really high) parchment eggs just shrivel. Also in brooders (such as pythons) and some other snakes, the eggs end up stuck together into a compact clump, so the surface area is effectively much reduced.

By John Scanlon, FCD (not verified) on 06 May 2011 #permalink

Mark Lees -- my family had a pet fox snake years ago, and it flatly refused dead food. Many captive snakes are fed dead food for reasons of practicality, and the keepers often have to be adept at making it look alive so the snake will strike at it. I also remember having to give live food to our pet lionfish.

But among mammals, I think pure predation is rare to nonexistant, especially among the large predators like lions. If you're that picky, you'll probably starve. And as far as the "scavenger dentition" goes, it was long thought that hyenas were scavengers who rudely bullied their way into other animals' kills. Now we know differently; they can and do hunt live prey (and I seem to recall reading of lions stealing hyena kills, reversing the usual stereotype). Having bone-crushing teeth doesn't mean you don't kill your own food when you have the opportunity. It just means you can make use of more of the carcass. Hyenas will eat the entire animal, bones and all, since their jaws are strong enough to crush bones. Perhaps as T. rex got larger, it simply became more necessary to extract maximum caloric and nutritional content from the kill, whether T. rex did the actual killing or not.

By Calli Arcale (not verified) on 06 May 2011 #permalink

David: I am very skeptical about that theory. The largest existing animals able to take off by jumping up are large katydids, and even in them the largest species have difficulties taking off from horizontal surfaces. Besides, wasn't here a post sometimes ago pointing out that large pterosaurs were too heavy to take off in any manner?

Darren, re the thin parchment-like eggshell, I still remember a day on the farm when I unintentionally ran over a laying hen with my bicycle. A soft-shelled (certainly parchment-like) egg was squeezed out of the hen, but she appeared fine. Since so few (four) fossil pterosaur eggs have been found, is it not possible that they were preserved only because they were still inside the female at death, and protected by her body, while the normal process of egg-laying could possibly have resulted in a shell that was more mineralized and therefore not as vulnerable once in the nest? Thanks!

By Lela Criswell (not verified) on 06 May 2011 #permalink

whether T. rex did the actual killing or not

...and there was no other carnivore of the size range of an adult or even subadult T. rex in its ecosystem.

The largest existing animals able to take off by jumping up are large katydids

No. All birds take off by jumping or running, and in most of them, regardless of size, all of the energy needed for takeoff comes from jumping.

This has all been discussed for months on the Dinosaur Mailing List.

Besides, wasn't here a post sometimes ago pointing out that large pterosaurs were too heavy to take off in any manner?

No. If anything, the opposite.

By David MarjanoviÄ (not verified) on 06 May 2011 #permalink

I suppose it is possible that Darwinopteris was either ovoviviparous or nearly so, producing a single egg at a time which hatched on being laid or imediately after. The egg would be incubated in a warm moist location (inside the female) and the thin shell might allow some transfer of calcium to the embryo. There would be no need for incubation or any sort of nest. The female would simply need a safe site to park the precocial young until it could fly off - much like a bat but without the placenta.

I don't know what the other three pterosaur eggs tell us, but pterosaurs could have been flexible between ovoviparity or oviparity depending on circumstances, like living squamates.

By Matthew Chapman (not verified) on 07 May 2011 #permalink

"All birds take off by jumping or running"
DM, flying is just modified jumping, isn't it? How can you imagine that birds started flight as runners, since running requires alternating gait and flying requires united gait?

David: that's simply not true. Look closely next time you flush a pigeon (not to mention a floating duck or a nightjar which has nothing to jump with). There is virtually no jumping motion involved.
In case of pterosaurs, taking off in this matter would involve fully extending arms down, then raising them all the way up while unfolding the wings. By the time the wings are raised and ready for a downstroke, a heavy animal would already be lying on its belly.

flying is just modified jumping, isn't it?

Why?

How can you imagine that birds started flight as runners

I am not at all talking about the evolution of flight. I'm talking about how extant birds take off.

Look closely next time you flush a pigeon (not to mention a floating duck or a nightjar which has nothing to jump with). There is virtually no jumping motion involved.

It has all been measured. That you can't see it doesn't mean it's not there.

By the time the wings are raised and ready for a downstroke, a heavy animal would already be lying on its belly.

No, the math has been done. Find and read the papers.

Remember that the very powerful forelimbs do most of the jumping, not the hindlimbs.

By David MarjanoviÄ (not verified) on 07 May 2011 #permalink

David: could you please explain to me how does a nightjar jump before taking off? It's feet are less than an inch long, and very weak. How does a duck jump up from floating position?

If a paper tells me that a cow can fly by flapping its ears, I don't have to go through the math to know that the paper is flawed. For jumping, you need not just "powerful" forelimbs: you need totally different muscles from those used for flying, and different muscle type from that needed for sustained flight. In a vertebrate, you also have to have much of those "jumping" muscles inside the limb, and that means you have to move them up and down with every wing stroke.

Hai-Ren (#39)-- Thank you for the duck tape! The clip of a small bird taking off that Cameron (#35) linked to looked like a hind-limb-powered jump (with the wings deploying when already off the ground), but the duck take-off looks more like what has been proposed for pterosaurs: forelimb-powered jump, with the wing actually contacting the substrate (water, in this case). I wouldn't a thunk it!

By Allen Hazen (not verified) on 08 May 2011 #permalink

Allen: there is an important difference: ducks hit the water with the entire surface of the wing. They basically start flying in a normal pattern while still on the water surface. What is proposed for the pterasaurs is using just the basal part of the wing for jumping, with the flight finger still folded. And it would have to generate so much vertical speed that by the time of the first downward stroke with fully extended wings the animal would be a few feet above ground - or it would hit the ground really hard with its fragile wingtips.

"the duck take-off looks more like what has been proposed for pterosaurs:"

Not something I know anything about, but I'll note that you can't see what the duck's feet are doing. Do we know that it's not pushing off with its feet underwater?

Mike:

Could this 35km-flying megapode chick have actually travelled part of the way on another ship?

Well, I'm sure that most of us, if we put our minds to it, could come up with several more or less plausible alternative explanations to explain away Healy's observation. By itself, it is indeed just an isolated anecdote.

But there is the wider biogeographical picture to consider here too. As I alluded to above - by citing Storrs Olson's 1980 paper - megapodes have a global distribution that includes a large number of quite distant volcanic islands in the Indo-Pacific region. (In fact, the megapodes' current distribution is far smaller now than what it was only a few thousand years ago; they used to be present on, e.g., Fiji, Tonga, Samoa, and many other Pacific islands.) Both paleontological and molecular data suggest that megapodes dispersed to these islands long before there were any humans with ocean-going vessels to help them along. Obviously, then, they must have gotten there on their own. And if megapodes are indeed not only capable but willing to cross long stretches of open sea by flying, there is no mystery to their distribution pattern and no special pleading is required to explain it.

(Most phasianid galliforms, by contrast, seem to have a very strong aversion to flying across open water for long distances; they are also noticeably absent from oceanic islands or island continents - with pretty much the only significant exception being the Coturnix quails. Incidentally, Olson went so far in his 1980 paper as to suggest that competition with larger phasianids is what has kept megapodes from colonising the Asian mainland, and that megapodes are effectively to be considered avian versions of marsupials and monotremes - that is, 'relictual' taxa only capable of surviving today in isolated Australasia.)

In this light, I'm myself prepared to take Healy's report at face value - that is, as offering support to the hypothesis of considerable long-distance dispersal capabilities in at least some megapode species.

Thanks for comments here. On launch behaviour in birds and pterosaurs, I have to agree very strongly with David. Substantial work has shown has the majority of birds launch into flight by jumping (e.g., Heppner & Anderson 1985, Bonser & Rayner 1996, Earls 2000): in fact 80-90% of initial flight velocity is generated by leg thrust (that's paraphrased from Tobalske et al. (2004). This is even true of hummingbirds. That's right, their flight is hindlimb-initiated, though theyâre unusual in initiating the flapping stroke during the 'jump' phase (Tobalske et al. 2004). Groups with even more strongly reduced hindlimbs (like swifts) simply cannot take off from the ground and need to drop from height or, in the case of tropicbirds, launch from water. I donât know what nightjars (and other caprimulgiforms) do, but would be interested to find out. There are, of course, mostly aquatic bird species that have to use a running launch, but they are not typical, with posteriorly positioned limbs and other specialisations for swimming. Unsurprisingly in view of their hindlimb-initiated launching, hindlimb bone strength in birds is typically really high - even in specialised fliers (e.g., albatrosses), femoral strength is higher than humeral strength (Habib & Ruff 2008).

Pterosaurs have disproportionally high forelimb bone strength and were quadrupedal. They were altogether different from birds. As David has said, work has been done providing good evidence for a forelimb-dominated launch that involves generating thrust with the upper arm, and using the elastic, joint-linked properties of the wing-finger. One of the main guys behind this model is Mike Habib: he sometimes shows up here in the comments but Iâm not sure if heâs around at the moment. Anyway, this hypothesis has been published and generally well received â it isnât just an informal idea. See Habib (2008) and Witton & Habib (2010).

Refs - -

Bonser, R. H. C. & Rayner, J. V. M. 1996. Measuring leg thrust forces in the common starling. The Journal of Experimental Biology 199, 435-439.

Earls, K. D. 2000. Kinematics and mechanics of ground take-off in the starling Sturnis vulgaris and the quail Coturnix coturnix. The Journal of Experimental Biology 203, 725-739.

Habib, M. B. 2008. Comparative evidence for quadrupedal launch in pterosaurs. Zitteliana B28, 159-166.

- . & Ruff, C. B. The effects of locomotion on the structural characteristics of avian limb bones. Zoological Journal of the Linnean Society 153, 601-624.

Heppner, F. H. & Anderson J. G. T. 1985. Leg thrust important in ï¬ight take-off in the pigeon. The Journal of Experimental Biology 114, 285-288.

Tobalske, B. W., Altshuler, D. L. & Powers, D. R. 2004. Take-off mechanics in hummingbirds (Trochilidae). The Journal of Experimental Biology 207, 1345-1352.

Witton, M. P. & Habib, M. B. 2010. On the size and flight diversity of giant pterosaurs, the use of birds as pterosaur analogues and comments on pterosaur flightlessness. PLoS ONE 5(11): e13982. doi:10.1371/journal.pone.0013982

Vladimir:

There is virtually no jumping motion involved.

Why don't you do what David suggested and actually look up the relevant literature? You can start by reading:

Earls, K.D. 2000. Kinematics and mechanics of ground take-off in the starling Sturnus vulgaris and the quail Coturnix coturnix. The Journal of Experimental Biology 203, 725-739.

It's feet are less than an inch long, and very weak.

The nightjar's feet are relatively short indeed, but I suspect that their supposed weakness has been much exaggerated. As can be briefly seen in this video, an adult nightjar can stand up quite well (the video also shows that the nightjar's chicks are able to take walking steps).

Oops, cross-posting of the Earls reference. Sorry about that.

I've just checked the literature and nightjars can indeed jump off the ground. I'm going to go out on a limb and suggest that, like most other birds, their launch is hindlimb-initiated.

OK, I was wrong about bird takeoffs (although, just for the record, red-throated loons have legs positioned so far posteriorly that they can't possibly push upwards, still they can take off from the water without running). Thanks for the info.
The reason I didn't look into the sources is that I have a flight overseas in three hours ad still have to pack for a 3-month trip. But I did read Habib's paper, and:
1) The main argument (femoral vs. humeral muscles) is an excellent proof of my never-landing theory, showing that back legs were not used for taking off. The author even mentions similar ratios in frigatebirds and other bird species that have severely limited ability to take off from level surfaces.
2) The paper doesn't address the problem of the inavoidable delay between leaping up and being able to make a downward stroke with fully extended wings. It doesn't attempt to estimate the acceleration/speed/height of the leap that the muscles of given sizes could produce.
3)And, by the way, if I remember correctly, none of the (very few) known/proposed pterosaur trackways ends in takeoff pattern (unlike so many bird trackways), indicating that the tracks were probably left by downed individuals desperately trying to walk to a cliff face before being eaten :-)

Hi Vlad. Not sure I'm following the argument here, but if you're suggesting that some pterosaurs didn't land (sorry: is this what you're saying?) - forget it. Many exhibit strong adaptations for quadrupedality and walking, plus there are hundreds and hundreds of pterosaur footprints. Some ornithocheiroid pterosaurs (in particular nyctosaurs) were perhaps about as aerial as frigatebirds and many non-pterodactyloids probably did a lot of clinging and climbing, but the data for a substantial amount of terrestrial walking is good for ctenochasmatoids, dsungaripteroids and azhdarchoids at least.

Oh - and the maths has been done to show that quadrupedally-launching pterosaurs (as explained above: launch power comes from forelimbs, not hindlimbs) do indeed leap high enough to allow 'clearing' from the ground for the first downstroke. This is all in the Witton & Habib paper cited above.

I suspect we have a new David Peters now.

@56: since some caprimulgiformes are ground nesters (such as nighthawks), and other routinely perch on the ground, I'm pretty sure they don't have many problems taking off. In fact, I know from observation that they have no problem taking off at all.

By heteromeles (not verified) on 09 May 2011 #permalink

1) The main argument (femoral vs. humeral muscles) is an excellent proof of my never-landing theory, showing that back legs were not used for taking off.

The forelimbs were used for taking off.

The stresses of takeoff exceeded those of flight, and indeed the trabeculae in the forelimb bones are aligned for the stresses of takeoff.

3)And, by the way, if I remember correctly, none of the (very few) known/proposed pterosaur trackways ends in takeoff pattern (unlike so many bird trackways), indicating that the tracks were probably left by downed individuals desperately trying to walk to a cliff face before being eaten :-)

Cliff face? Tracks are made on flat beaches, not on rocks.

A track of a landing pterosaur has been published. I've seen it myself (it's in Crayssac). The pterosaur touched down with the hindlimbs once, bounced back, landed again on all fours and immediately walked away at the usual high speed. This was clearly a routine maneuver, not a crash landing.

"Very few known/proposed pterosaur trackways"??? You seem to have missed the last ten years entirely.

By David MarjanoviÄ (not verified) on 09 May 2011 #permalink

"mostly aquatic bird species that have to use a running launch" DN, this is clearly derived, due to conspecific competitor evasion. Jump/fall-glide flight preceded, IMO.

re. 54: that is, "competitive conspecific and predator evasion" (eg. basilisk lizard swims but when hurried runs bipedally with no wing development)

Regarding ducks not having anything to jump off of with their hindlimbs (though the video made it pretty clear they can take off by pushing the water with their forelimbs), this is just not true. They have webbed feet. Water doesn't offer as much resistance as solid ground, but there's some, and it would matter.

An example of this is the loon, which has great difficulty transitioning between flight and swimming. It has to struggle mightily to take off, and may make several touch-and-goes before it can gain enough altitude to clear the treeline, especially if the wind isn't cooperative. It absolutely must paddle along furiously, "running" on the water while it flaps for all it's worth, or it has no hope of getting up above stall speed. It probably wishes it could jump like a pterosaur.

By Calli Arcale (not verified) on 10 May 2011 #permalink

OK, OK, at least I provoked an interesting discussion!
Of course, the reason there are so many pterosaur tracks is because they were downed en masse by severe weather effects; beaches are often bordered by cliffs; quadripedal adaptations show that many of downed ones actually did get to cliffs and survive; landing is not takeoff... but there is also a little chance that I am wrong... can't be right 100% of time.
Calli: all loons except the red-throated. That one CAN take off without running, that's why in areas where other loon species are common it is mostly found on small lakes and ponds.

Still regarding megapodes and their flight capabilities... A recent paper by Dial & Jackson (2011) shows that juvenile megapodes are indeed considerably better fliers than the adults are. (The authors also suggest that most long-distance dispersal - including over-water dispersal - in these birds is by juveniles.)

Reference:

Dial, K.P. & Jackson, B.E. 2011. When hatchlings outperform adults: locomotor development in Australian brush turkeys (Alectura lathami, Galliformes). Proceedings of the Royal Society B, 278, 1610-1616.

Of course, the reason there are so many pterosaur tracks is because they were downed en masse by severe weather effects;

...which would not be conducive to the preservation of footprints at all.

beaches are often bordered by cliffs

We know the geology of Crayssac and many other sites. No cliffs.

Dial, K.P. & Jackson, B.E. 2011. When hatchlings outperform adults:

*pretending to be able to raise one eyebrow* Fascinating.

By David MarjanoviÄ (not verified) on 10 May 2011 #permalink

David:

Fascinating.

It is indeed. In the paper, Dial & Jackson also speculate whether similar within-species differences in flight capabilities were also present in certain Mesozoic dinosaurs. Flying Velociraptor hatchlings, anyone?

Like I said, another David Peters. Reality denying? Check. Bizarre illogical proposals? Check. Ignorance towards the scientific discoveries about pterosaurs made in the last ten years? Very much check!

I know *Sylviornis* is not a megapode, but...is it possible that the adults were flightless whereas the chicks could still fly? Are any *Sylviornis* chicks known?

@Dartian
I suggested several years ago that flight appeared first in juvenile dinosaurs, that some forms had flying juveniles and flightless adults, and that first fully flying forms evolved by neoteny. I posted that on some internet forums, in fact.
The reasoning was:
- Early birds and related dinosaurs developed from juveniles to adults without much change of limb proportions (unlike modern birds). Body-mass ratio makes such juveniles automatically stronger and more capable of flying than adults.
- First flying species were much smaller than contemporary flightless dinosaurs, suggesting neoteny.
- Megapode chicks were not researched then. But other modern birds have flying juveniles, and heavy adults are flightless (steamerducks, giant coot) or incapable of taking off level ground (snowcocks).

I also wondered if feathers might evolve first for display in adults, and be later co-opted for flight in juveniles? Or if ecological force driving flight was juveniles escaping predation on trees?

I think nobody pointed before that pterosaur eggs are found singly, but modern reptiles which lay thin-shelled eggs generally lay whole clutch en masse.

It would be difficult for a pterosaur to revisit the place and bury many eggs without attracting predators, or waste time to protect just one egg at a time.

Carlos: the purpose of writing/commenting scientific blogs is exchanging knowledge and ideas. I learned a lot from this discussion, and a few other people too, apparently. So I think my little "theory" was worth trying out, even if it was a wrong one. Sorry you didn't get the humor in my last comment.

I enjoyed reading those 65 comments! I would rather choose not to comment on this but I would like to give all of you a clap. Your exchanging of ideas gave me additional knowledge and info.
to Vladimir. your right :)

[from Darren: peculiar and irrelevant url deleted]

By herky stubby (not verified) on 11 May 2011 #permalink

Flying Velociraptor hatchlings, anyone?

That would be... so... awesome...

And indeed, if we compare Microraptor, Bambiraptor and Velociraptor, I wonder.

By David MarjanoviÄ (not verified) on 12 May 2011 #permalink

No one took note of Calli Arcale's idea earlier of the fossil being an "undercooked" egg. Seems possible, unless we're both missing something. That there was a mineral layer at all, though thin, and the species is phylogenetically bracketed by hard-shelled species, makes me think it's secondarily thin-shelled. If it is actually thin-shelled at all.

By CS Shelton (not verified) on 15 May 2011 #permalink

But that idea ignores the fact that this is but one of four known pterosaur eggs, only one of which preserves a mineral layer (and even then it's extremely thin, as noted above). Furthermore, it's the only one preserved in association with an adult - the others were isolated. The 'undercooked' hypothesis isn't impossible, but it's unlikely based on the evidence we have at the moment.