A spectacular new fossil provides insight on the sex lives of pterosaurs, part I


We know all too little about the biology and behaviour of the pterosaurs, the amazing, often bizarre flying reptiles of the Mesozoic Era. Most of our ideas - about feeding behaviour, locomotion, physiology and social and sexual behaviour - are inferences based on bones, or inferences based on interpretations of the bones. A new paper - published just a few weeks ago by Junchang Lü, David Unwin and colleagues in the journal Science - describes what might be one of the most interesting pterosaur fossils yet discovered, and arguably one of the most significant in terms of what it tell us about pterosaur biology. [Exclusive image below, by Mark Witton, shows male and female Darwinopterus. You may note some slight dimorphism...].


It's an adult pterosaur skeleton, poised as if 'caught in the act' of laying an egg [the specimen is shown below: image courtesy of Junchang Lü, Institute of Geology, Beijing, used with permission]. As usual with fossil mothers preserved in the act of parturition, it seems likely that the egg was pushed out of the body during decomposition: the mother wasn't really laying an egg when she died (though this does, admittedly, remain a possibility).


The specimen concerned (nicknamed 'Mrs T') was discovered in the Middle or Upper Jurassic rocks of the Taojishan Formation of Liaoning Province, China, and has been identified as a new specimen of the recently named Darwinopterus.

As I explained in October 2009, Darwinopterus is already a remarkable pterosaur: it's a crucial 'transitional' animal that helps bridge the gap between the mostly long-tailed, archaic 'rhamphorhynchoid' pterosaurs and the generally shorter-tailed, longer-winged pterodactyloid pterosaurs (Lü et al. 2009). Unlike so many other remarkable fossil species, it was already known from several individuals when first described, not just one.

The global record of pterosaur eggs just improved by 25%!


The first thing to say about this new Darwinopterus specimen is that the discovery of a pterosaur egg is an extremely rare event. After decades of speculation about the missing eggs of pterosaurs*, their oviparous habits were confirmed in 2004 when the embryo of an ornithocheiroid, preserved within a partial egg, was reported from China's famous Yixian Formation (Wang & Zhou 2004). A second embryo-within-an-egg - this time belonging to the weird, filter-feeding pterosaur Pterodaustro - was reported from Argentina's Lagarcito Formation (Chiappe et al. 2004). And a third - another ornithocheiroid - was reported from the Yixian Formation (Ji et al. 2004) in the same journal issue as the Pterodaustro egg, just one page later [a reconstruction of the Pterodaustro baby within its egg is shown here: does anybody know who the artist is? It's nice, but they seem to have given it bird feet].

* By which I mean that people did sometimes ask such questions as "Where are the missing eggs of pterosaurs?".

While the idea that pterosaurs were viviparous has been mooted, these discoveries of pterosaur embryos preserved within eggshells demonstrate beyond doubt that pterosaurs were egg-layers. Darwinopterus is more basal within the pterosaur family tree that any of the other taxa for which eggs have been discovered, so it helps to push the undoubted presence of oviparity 'down' the pterosaur tree. Of course, we predict that all pterosaurs were oviparous anyway, given the presence of egg-laying habits in the taxa that bracket them phylogenetically (namely, crocodilians and birds).

The three eggs found prior to the new Darwinopterus specimen were extremely informative: they showed that at least some pterosaurs produced parchment-like, apparently highly permeable eggshells, and showed that pterosaurs hatched as precocial babies, able to fly (and presumably look after themselves) within days or even hours of hatching (Unwin & Deeming 2008). But those eggs were found on their own, in isolation. As such, they provide only a limited amount of information on the reproductive biology of pterosaurs as a whole. The discovery of a definite female preserved with one of her eggs might tell us much, much more.

Is it really an egg? Yes

Before concluding that the object is a definite egg, a few other ideas had to be considered and eliminated. Firstly, the possibility that Mrs T might be a fake or composite can definitely be discounted. This specimen wasn't on sale as 'pterosaur with egg' in a fossil shop or anything like that, but was prepared (under documented supervision) at the Chinese Academy of Geological Sciences, from its original entombed state. Could the object be something other than an egg, like a mollusc shell, a sedimentary concretion of some sort, a gigantic fish scale or an ossified tumour? [the egg is shown below: image courtesy of Junchang Lü, Institute of Geology, Beijing, used with permission].


Well, no. Not only does the egg have the microscopic surface features unique to eggshell (and unique to pterosaur eggshell at that), it's the right size and shape for an egg, it's exactly the right size for the pelvic canal of the associated skeleton, its style of preservation shows that it consisted of tough soft tissues rather than mineralised tissue, and it was discovered right next to the pelvis of an undisturbed, complete adult pterosaur skeleton. I'd say this is pretty convincing.

While the egg was crushed during burial (as was the accompanying skeleton), its dimensions don't appear to have been altered much, so its preserved size (28 mm long, 20 mm wide) is probably similar to its original one.

Searching for sexes in Pterosauria: a bit of background

Parsimony dictates that an animal preserved with an egg emerging from its cloacal region really must be female. So, for the first time ever, we have a pterosaur of confirmed sex. This is hugely significant, as various minor and major skeletal differences in pterosaur specimens have previously been suggested to evidence sexual dimorphism.

As is well known, pterosaurs were often spectacular, showy animals, sometimes decorated with (occasionally enormous) head crests. Our sample sizes are typically so low that we often can't make confident claims about sexual dimorphism. However, there are at least a few indications that the crests only appeared with the onset of maturity (Martill & Naish 2006), and that their form and/or size is correlated with sex. Consequently, the idea that sexual display was their primary function has been fairly popular (e.g., Wellnofer 1991, Naish & Martill 2003, Unwin 2005, Bennett 2006).

In 1992, Chris Bennett noted that the giant marine pterosaur Pteranodon longiceps comes in two morphs: a big one with a long cranial crest and a proportionally narrow pelvic canal, and a smaller one, with a short cranial crest and a proportionally wide pelvic canal. Bennett proposed, not unreasonably, that the large, big-crested animals were males while the smaller, small-crested ones were females whose wider pelvic canals were there to allow the passage of eggs (Bennett 1992). [Image below, by Matt Martyniuk, from wikipedia, shows big-crested putative male and small-crested putative female pteranodontids. Pteranodontid taxonomy was recently revised (Kellner 2010), hence the use here of the names Dawndraco, Geosternbergia and G. maysei].


This all sounds very sensible. But - in the absence of better data - it isn't necessarily convincing. For one thing, it's almost never possible to prove that your two 'sexual dimorphs' really are 'sexual dimorphs': they could represent closely related species. And while the idea that the sexes might be distinguishable based on pelvic (or pelvic canal) width seems sensible, in living archosaurs it's pelvic depth rather than width that's important, and this is rarely preserved in fossils due to crushing. Furthermore, a substantial overlap of pelvic dimensions between the sexes means that any dimorphism is only demonstrable with huge sample sizes (Prieto-Marquez et al. 2007). Having said that, analysis of the pelvic dimorphism reported for Pteranodon by Bennett (1992) proved it to be sex-specific variation of the sort seen in living crocodilians (Prieto-Marquez et al. 2007).

The sexual dimorphism hypothesis confirmed --- or is it?

One of the most exciting things about the new Darwinopterus with the egg is that - unlike the originally reported individuals of this taxon - it lacks the bony skull crest normally seen in this pterosaur. Here, at last, is what appears to be excellent supporting evidence for the sexual dimorphism hypothesis: we know that this female lacked a crest, ergo it's assumed that all females lack crests, and ergo the crested individuals must be males [sexual dimorphism in Darwinopterus depicted below: image by Mark Witton].


When we look at the dimensions of the pelvis in the different Darwinopterus specimens, we find (as Bennett did for Pteranodon) that Mrs T and other crestless specimens suggested to be female have proportionally bigger pelvic girdles than the crested specimens suggested to be males (Lü et al. 2011). Furthermore, in the putative females the pelves are not fused together along the ventral midline, nor dorsally to the adjacent sacral vertebrae. In the putative males, both areas are fused up.

As Lü et al. (2011) note, the apparent dimorphism present in Darwinopterus seems to confirm Bennett's proposal about Pteranodon, and it might be that other crested and crestless pterosaur 'species' could also represent males and females, respectively. Ornithocheirids, in particular, come to mind: both crested and crestless 'species' have been named from such units as the Cambridge Greensand, and the idea that these might be males and females, and not separate species at all, has been mooted by some authors (it's also made its way into popular culture: see ep 4 - 'Giant of the Skies' - of Walking With Dinosaurs). Is this the end of it then - do we finally know that pterosaur cranial crests were exclusive to males? Well, not so fast.

Devil's advocacy # 1


Two alternative possibilities cast doubt on this otherwise fine hypothesis. Firstly, do we know that the crestless female Darwinopterus preserved with the egg definitely belongs to the same species as those crested individuals of D. modularis? It now seems that Darwinopterus isn't a phylogenetic singleton: according to recent work by Xiaolin Wang and colleagues, there's a whole host of these weird, long-tailed, long-necked 'modular' pterosaurs. They call them wukongopterids, and in addition to Darwinopterus modularis, they include D. linglongtaensis, Wukongopterus lii, Kunpengopterus sinensis [shown here, from Wang et al. (2010)] and Changchengopterus pani (Wang et al. 2009, 2010).

So far as I can tell, the putative female Darwinopterus described by Lü et al. (2011) does look more like a D. modularis than anything else. But can we be absolutely sure that it belongs to this species? It has to if we want the sexual dimorphism hypothesis to remain viable. Personally, I think it's most likely that it is a female D. modularis, but I have to note the alternative possibility that it could instead represent a new wukongopterid species characterised by crestlessness (and perhaps by a proportionally bigger, more flexible pelvis). Some other wukongopterids (like Kunpengopterus) seem to lack bony cranial crests (Wang et al. 2010), though - if we of course endorse the sexual dimorphism hypothesis - the described crestless holotype of this taxon could be a female, with crested males awaiting discovery. But read on, read on...

Devil's advocacy # 2

Secondly, do we know that the crestless Mrs T is definitely an adult? Recent studies on reproductive maturity and bone growth in dinosaurs have shown that members of several Mesozoic dinosaur lineages were able to reproduce before they reached full skeletal maturity (Lee & Werning 2008). It's possible that the same was true for pterosaurs though, so far as I know, we don't know that it was. Could it be, therefore, that Mrs T was sexually mature (mature enough to mate and lay eggs), but not mature enough to possess a bony crest? Rather than representing anything specific to gender, that unfused pelvis could be a consequence of age.

Incidentally, if you're wondering why I've switched from talking about 'crests' to 'bony crests' it's because at least some (maybe lots, maybe ALL) pterosaurs possessed soft tissue crests that frequently left no fossil evidence of their presence. One of the most remarkable recent discoveries made about pterosaurs is that some taxa long assumed to be completely crestless - like Pterodactylus - actually had large, soft-tissue crests. It's only thanks to exceptionally well preserved specimens that we know this.

Conclusion: sexual dimorphism most likely


Despite this devil's advocacy, I conclude that the sexual dimorphism hypothesis is the most sensible conclusion. Mrs T most probably is fully mature given that (1) she is similar in size to at least some putative male specimens of Darwinopterus, and (2) she exhibits extensive skeletal fusion throughout the non-pelvic part of the skeleton (Lü et al. 2011). It might be that the bony cranial crest grew really late in ontogeny (and a complication is that pterosaurs from this region of phylogeny probably had indeterminate growth), but the idea that it might only appear after full skeletal maturity had occurred seems like a bit of a stretch. It's also contradicted by evidence from other pterosaurs.

A few years ago Dave Martill and I described a juvenile thalassodromid pterosaur where the bony crest seemed about 'half grown' compared to that of an adult (Martill & Naish 2006), yet this individual was well away from adult size [the adjacent diagram, from Martill & Naish (2006), shows our hypothesised ontogenetic sequence for Thalassodromeus. Note that even juveniles like the one shown in B have half-grown bony cranial crests]. Juvenile specimens of other crested pterosaurs - Tapejara and Sinopterus among them - also have smaller versions of the bony crests developed to full size in adults. These data indicate that pterosaurs started to develop adult sexual paraphernalia during their 'teenage' years and were fully developed in terms of sexual ornament by the time they'd reached adult size. Mrs T's lack of a bony crest is, therefore, most likely the normal condition for skeletally mature individuals of her taxon.

Therefore, adult female specimens of D. modularis lacked bony crests, and it seems valid to now assume that crestless or small-crested pterosaurs really are the females of the big-crested versions. Can this be tested?

Medullary bone in pterosaurs? Dammit, no...

It's now quite well known that Mesozoic dinosaurs laid down a special of spongy bone tissue - medullary bone - when preparing their bodies for egg-laying. Consequently, some specimens can be unambiguously identified as female. The idea that female pterosaurs produced medullary bone, and hence that their fossils can also be sexed, is appealing. However, just because pterosaurs are close relatives of dinosaurs, it doesn't follow that they used medullary bone too. Within archosaurs, it may be a special feature unique to Dinosauria (it's apparently not present in crocodile-group archosaurs). Furthermore, the leathery eggshells of pterosaurs may not have required the production of medullary bone in the first place. Anyway, people have looked at pterosaur bone histology an awful lot already and have never found medullary bone, even when searching for it specifically (L. Steel, pers. comm.).

There's a lot to say about the Mrs T specimen - so much so that another article on the specimen (this time focusing on what the specimen means for such things as nesting behaviour and the biology of hatchlings) is to follow...

This article featured some new artwork by Mark Witton (the next one will too). Here's a good chance to remind you that Mark is currently working on a big, comprehensively illustrated book on pterosaurs for Princeton University Press. A sneak peek is here.

For previous Tet Zoo articles on pterosaurs see...

Refs - -

Bennett, S. C. 1992. Sexual dimorphism of Pteranodon and other pterosaurs, with comments on cranial crests. Journal of Vertebrate Paleontology 12, 422-434.

- . 2006. Juvenile specimens of the pterosaur Germanodactylus cristatus, with a review of the genus. Journal of Vertebrate Paleontology 26, 872-878.

Chiappe, L., Codorniú, L., Grellet-Tinner, G. & Rivarola, D. 2004. Argentinian unhatched pterosaur fossil. Nature 432, 571-572.

Ji, Q., Ji, S.-A., Cheng, Y.-N., You, H.-L., Lü, J.-C., Liu, Y.-Q. & Yuan, C.-X. 2004. Pterosaur egg with a leathery shell. Nature 432, 572.

Kellner, A. W. A. 2010. Comments on the Pteranodontidae (Pterosauria, Pterodactyloidea) with the description of two new species. Anais de Academia Brasileira de Ciências 82, 1063-1084.

Lee, A. H. & Werning, S. 2008. Sexual maturity in growing dinosaurs does not fit reptilian growth models. Proceedings of the National Academy of Sciences 105, 582-587.

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., Jin, X., Liu, Y. & Ji, Q. 2009. Evidence for modular evolution in a long-tailed pterosaur with a pterodactyloid skull. Proceedings of the Royal Society B 277, 383-389.

Martill, D. M. & Naish, D. 2006. Cranial crest development in the azhdarchoid pterosaur Tupuxuara, with a review of the genus and tapejarid monophyly. Palaeontology 49, 925-941.

Naish, D. & Martill, D. M. 2003. Pterosaurs - a successful invasion of prehistoric skies. Biologist 50 (5), 213-216.

Prieto-Marquez, A., Gignac, P. M. & Joshi, S. 2007. Neontological evaluation of pelvic skeletal attributes purported to reflect sex in extinct non-avian archosaurs. Journal of Vertebrate Paleontology 27, 603-609.

Unwin, D. M. 2005. 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.

Wang, X., Kellner, A. W. A., Jiang, S., Cheng, X., Meng, X. & Rodrigues, T. 2010. New long-tailed pterosaurs (Wukongopteridae) from western Liaoning, China. Anais de Academia Brasileira de Ciências 82, 1045-1062.

- ., Kellner, A. W. A., Jiang, S. & Meng, X. 2009. An unusual long-tailed pterosaur with elongated neck from western Liaoning of China. Anais da Academia Brasileira de Ciências 81, 793-812.

- . & Zhou, Z. 2004. Pterosaur embryo from the Early Cretaceous. Nature 429, 621.

Wellnhofer, P. 1991. The Illustrated Encyclopedia of Pterosaurs. Salamander Books Ltd, London.


More like this

nicknamed 'Mrs T'

I pity the fool who came up with that nickname. (Sorry, couldn't resist...)

The thing that's always bugged me about the crests on pterosaurs is that they aren't aerodynamically neutral. As such, they make great sexual displays, but they must have been trouble to maintain.

Because of this, I suspect that there would be even more sexual dimorphism than we're assuming here. After all, the males had to support permanent crests, unlike (some) of the females, so one might expect niche separation between sexes in the more dimorphic species. Their bodies will be different shapes away from the crest.

If you really want speculation: There might also be multiple male morphs in some species, perhaps sneaker males. I hope that, if there was a pterosaur equivalent of the ruff Philomachus pugnax, we find good fossils of all the morphs.

By heteromeles (not verified) on 10 Feb 2011 #permalink

Seeing the picture of the various pteranodontid crests has made me wonder about that new paper on pteranodontid taxonomy. Is there any strong morphological evidence for separating Dawndraco and Geosternbergia from Pteranodon, or is it just the crests?

By Anonymous (not verified) on 10 Feb 2011 #permalink

This egg is not strange for a Archosaurs?
(I think that all Archosaurs have eggs very stiff and hard shell).

This egg is no longer appropriate for a relative of Prolacertiformes or Drepanosauromorphs. (?!?)
Convergent evolution? Lightening?

Will have consequences in the debate on the origin of Pterosauria?

Medullary bone in pterosaurs? Dammitâ¦yes (possibly)

Another excellent blog Darren! Not much to add, except to note that Chinsamy et al. (2009) figure a cross-section of the femur of Pterodaustro (MHIN-UNSL-GEO V [Museo de Historia Natural, Universidad Nacional de San Luis, Departamento de Geologı´a, Vertebrados] 382) that contains a large amount of endosteal bone (op cit., fig. 5), interpreted by these authors as possibly medullary bone. If this interpretation is correct then we may have another independent line of evidence for assigning gender. Happily, as Junchang and I have recently been awarded some serious funding, and have also gained access to numerous examples of Darwinopterus we will have an opportunity to investigate this aspect of gender later this year. Concurrently, I will also be revising the systematics of pterosaurs from the Tiaojishan Formation and demonstrating that there are, at most, only two species of monofenestratan: Darwinopterus modularis and possibly one other. One final comment, Unwin et al. (2006) explored the distribution of oviparity in pterosaurs at some length, see for example our figures 10 and 11 which set this issue in a phylogenetic context.

Refs - -

Chinsamy, A., Codorniú, L. & Chiappe, L. 2009. Palaeobiological implications of the bone histology of Pterodaustro guinazui. Anat. Record 292, 1462-1477 (2009).

Unwin, D. M., Lü Junchang and Deeming, D. C. 2006. Were all pterosaurs oviparous? Pp. 143-169. In: Lü Junchang, Kobayashi, Y., Huang Dong and Lee, Y.-N. (eds) Papers from the 2005 Heyuan International Dinosaur Symposium. Geological Publishing House, Beijing.

By Dave Unwin (not verified) on 10 Feb 2011 #permalink

I can't wait for that book!

By Marco Tedesco (not verified) on 10 Feb 2011 #permalink

Nice, except it tells us nothing about "sex lives" of pterosaurs. Some bonew***ers need to be taught about the bird family which has bony crests. Hornbills are also remarkable in that most species are not only faithfully monogamous. In fact, males go to extreme parental investment by feeding both female and chicks until they reach flying ability.

BTW, interesting suggestion that pterosaurs were almost ovoviviparous in practice. Flying young hatched within hours from laying an egg. Reducing flightless period as much as possible is also an evolutionary trend in birds. Could it be why so few pterosaur eggs were found?

BTW, any chance that pterosaur chicks grabbed to their parents like young bats?

Jerzy- you cannot throw nasty insults at all vertebrate paleontologists and then expect them to provide nice answers to your questions. Actually does seem rude.

A mere pterosaur pfanboy popping in here to say how much he's looking forward to Mark's book and how inexpressibly cool it is to be reading this post and see a comment from Dave Unwin, author of the wonderful Pterosaurs: From Deep Time. A great example that book is for Mark to live up to.

By Mike from Ottawa (not verified) on 10 Feb 2011 #permalink

The things I miss when I lack access to Science...

a reconstruction of the Pterodaustro baby within its egg is shown here: does anybody know who the artist is? It's nice, but they seem to have given it bird feet


I see a bird (tarso)metatarsus followed by a pterosaur metatarsus followed by toes.

The scaled tail is bad; the feet are moronic.

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

It's rare to find suggestions that one or other pterosaur might have exhibited herd behavior, with an alpha male monopolizing (or anyway extracting rent for) access to multiple females. Does it show up anywhere in birds?

Is this a good place to bring up, again, hydrofoiling pterosaurs, that used their crest as a sail and their folded-under wingtips as foils? No, I thought not.

By Nathan Myers (not verified) on 10 Feb 2011 #permalink

Perhaps the first pterosaurs evolved with relatively high basal metabolic rates under relatively low environmental oxygen levels in the late Triassic. The pterosaurs either evolved or retained the thin egg shell because it allowed higher embryonic respiration rates and faster embryonic development. It's a less secure system, of course, and that makes for another interesting discussion of how they secured their nests.

This also makes me wonder if bird evolution was inhibited at first by their relatively thick egg shells.

By heteromeles (not verified) on 10 Feb 2011 #permalink

@ Anonymous
IMO Kellner's paper on _Pteranodon_ taxonomy makes a pretty weak case for separation. Separating _P. sternbergi_ from _P. longiceps_ is standard, and whether or not to place them in separate genera is purely an aesthetic choice. However, the main reason for splitting off _Dawndraco_ is that this particular specimen (which Bennett assigned to _P. sternbergi_) has a very long bill and a more upward-directed crest. But as the diagram Darren used shows there's no really good evidence that all _P.sternbergi_ didn't have a very large and long beak like that, and as Bennett suggested this could be a large individual. The only wild card is the apparent presence of such a narrow crest in what is (stratigraphically) a _P. sternbergi_. I wonder if it could be a traditional form, supporting bennet's idea that sternbergi->longiceps is a single, anagenetic, lineage.

As for _"G." maysei_ this is one of the only skulls known from the Pierre Shale, younger than even _P. longiceps_. It also has an odd, broad, upright crest that even though it's incomplete, looks fairly dissimilar from the _P. longiceps_ standard. Bennett reckoned this could be a very old male _P. longiceps_, which would make sense given the apparent "trend" of larger individuals crests being apparently more vertical. Many sources have treated the Pierre Shale pteranodonts as an indeterminate species, and given the crest, size, and stratigraphic position (basically the same qualifications used to distinguish _P. sternbergi_ I don't see why it couldn't represent a distinct species. But if it's really an anagenetic lineage, it couldn't be _Geosternbergia maysei_-it would need to be _Pteranodon maysei_ or a new genus.

For the record, I also produced a version of the profiles diagram illustrating Bennett's taxonomy, for comparison.

If three pterosaur eggs had been found previously, presumably this one increased the record by ~33%, not 25%?

Regarding oviparity, phylogenetic bracketing, etc, are there any secondarily oviparous amniotes? Viviparity seems intuitively like a "hard" change to undo.

Finally, I don't want to sound like a concerned citizen who is not a paleontologist, but Jerzy's insults are getting real old. Could he be put on moderation or something?

By Andreas Johansson (not verified) on 10 Feb 2011 #permalink


Some bonew***ers need to be taught about the bird family which has bony crests.

Did you not notice that Dave Unwin, one of the authors of the paper we're talking about, posted a comment just before you did? Or did you just not care?

Reducing flightless period as much as possible is also an evolutionary trend in birds.

Not really. Precocial young are characteristic of Ratites, Galliformes, and Anseriformes, the most basal bird groups. (Probably just like their dino ancestors.) All those blind, downy-to-almost-featherless baby birds belong to more derived groups.

@Dartian and others
Since the blog title begins with "sex life", then more colorful language is allowed, i suppose. If you puff with pride, what do you do in a discipline heavily involved in such non-serious topics as faeces, weird sex organs, rotting animals etc?

My comment was directed at confusing sexual dimorphism with reproductive strategy (nicely called "sex life" here), despite that in close living analogues the two are very weakly related. It appeared here somewhere between original article and blog post, and wouldn't provoke much interest except that I saw it before in popular paleontological literature.

Brian Unwin's comment I noticed now. But his comment about medullary bone doesn't relate directly to either reproductive strategy, hornbills, hatching time of egg nor parental care.

Many thanks for comments, especially Dave Unwin. On other matters...

Given that the topic being discussed above basically revolves around the apparent presence of SEXUAL DIMORPHISM, I think I'm justified in saying that this work informs us about the 'sex lives' of pterosaurs. On a rather more personal note: Jerzy, please, please, please give up on the arrogant and extremely annoying idea that you are cleverer than the palaeontologists and other zoologists who have actually worked on this stuff. Yes, some hornbills have bony crests (as do members of some other bird groups). Are you seriously using this one superficial similarity to say that the extremely peculiar nesting behaviour of (virtually all, but not all) hornbills was also practised by pterosaurs? Those of us who like to use evidence to formulate our hypotheses will confirm that this was (so far as we can tell) definitely not the case.

Incidentally, some similarities are apparent between hornbills and one specific group of pterodactyloid pterosaurs (the tapejarids), but the similarities concern ecomorphology, not reproductive behaviour. This has been discussed several times in the pterosaur literature already (starting with Wellnhofer & Kellner (1991)), and a lengthy in-press manuscript discusses comparisons between hornbills and tapejarids in depth. This makes what you say about 'bonewankers' being naive idiots both highly ironic and particularly frustrating.

Also in response to Jerzy: no, pterosaurs were not ovoviviparous, they were fully oviparous, merely producing leathery-shelled eggs (or, at least, eggs with extremely thin mineral layers). It's assumed that incubation periods were quite long, but the skeletons of both embryos and young juveniles show that pterosaurs were superprecocial. We don't know anything about parental care in pterosaurs, but superprecociality is contrary to the idea that they were carried around by their parents. In any case, pterosaurs were not bats â most had poor gripping/grasping abilities.

Ref - -

Wellnhofer, P. & Kellner, A. W. A. 1991. The skull of Tapejara wellnhoferi Kellner (Reptilia, Pterosauria) from the Lower Cretaceous Santana Formation of the Araripe Basin, northeastern Brazil. Mitteilungen der Bayerischen Staatssammlung für Paläontologie und Historische Geologie 31, 89-106.

It seems likely that the first flying birds were also superprecocial. Megapodes and Odontophorids can fly almost from the moment of hatching. One would expect birds to be under immense pressure to evolve vivipary and shorten the non-flying period to zero, but in reality vivipary has never evolved (even in species that lay one egg, and shouldn't mind carrying it inside for longer). I wonder why birds have never shifted to vivipary, while bats seem happy with it and Squamates have evolved it on multiple occasions. May be it has something to do with higher oxygen demands of bird embryos? But why not lay eggs with embryos at least partially developed?

Fossil evidence (both from close relatives of birds, and Mesozoic birds themselves) shows indeed that early birds were precocial or superprecocial. As for the absence of viviparity in birds, the classic source on this is...

Blackburn, D. G. & Evans, H. E. 1986. Why are there no viviparous birds? The American Naturalist 128, 165-190.

And below is some text I previously produced for Ask A Biologist...

It remains a good and frequently asked question: why haven't birds (and/or crocodilians and/or turtles) evolved viviparity? (viviparity = the ability to retain the egg within the reproductive tract until embryonic development is complete). Viviparity has evolved in just about everything else, in fact in fishes, amphibians and reptiles it has evolved at least 150 times! Several possible answers have been put forward (see Blackburn & Evans 1986): most recently, Andrews & Mathies (2000) argued that the eggshells and oviducts of turtles and birds - unlike those of other reptiles - do not allow much oxygen exchange to occur. They proposed that this might be a constraint that has prevented extended embryonic development in utero, and hence also prevented the evolution of viviparity. Their argument is actually rather more complex than that, but I hope that's an adequate summary.

Andrews, R. M. & Mathies, T. 2000. Natural history of reptilian development: constraints on the evolution of viviparity. BioScience 50, 227-238.

Semi-random thought, but a developing fetus shift's the mother's center of gravity, and this can cause issues for a flying animal (or a bipedal animal, if it comes to that). Obviously bats have solved the problem, but I do wonder how pregnancy constrains morphological diversity in bats.

By heteromeles (not verified) on 12 Feb 2011 #permalink

Holy crap! Pterosaur eggs have actually been known since 2004 and nobody bothered to inform me about this momentous occasion? Why am I always the last to know? :(

"Having said that, analysis of the pelvic dimorphism reported for Pteranodon by Bennett (1992) proved it to be sex-specific variation of the sort seen in living crocodilians (Prieto-Marquez et al. 2007)."

Whew! Good to know I won't have to burn/toss out my copy of In The Presence of Dinosaurs for having inaccurate paintings. ;)

Excited for the next portion of pterosaur goodness!

By Nathan Hofstad (not verified) on 12 Feb 2011 #permalink


The pterosaur eggs were reported in Nature*... it was pretty well publicized.

Ji Q, Ji S-A, C Y-N, You H-L, Lü J-C, Liu Y-Q & Yuan C-X (2004). Pterosaur egg with a leathery shell. Nature 432: 572.

Re #22

I don't think changes in the center of gravity are critical for a flying animal. They are for aircraft because of stability requirements. For an aircraft to be dynamically stable the center of gravity must be in front of the center of lift, because then disturbances in pitch and yaw will be damped naturally. If the center of gravity is behind the center of lift, then such disturbances will increase instead of being damped.
Now dynamic instability is not an inherently bad thing. An unstable aircraft (or animal) is much more maneuvrable than a stable one, and will also have slightly better fuel economy. The requirement for dynamic stability is due to our limitations as non-flying animals, we simply don't have the capability to continually correct small departures in pitch and yaw. Modern fly-by-wire control systems can do this, by simple negative feedback, and modern military aircraft are consequently built dynamically unstable. I think it is extremely likely that flying animals can do this too, after all they manage to do without any fixed control surfaces.

By Tommy Tyrberg (not verified) on 13 Feb 2011 #permalink

Thinking a bit further, at least hummingbirds and some insects can definitely fly in an unstable condition, since they can fly both forwards and backwards and can hardly avoid being unstable in one direction.

By Tommy Tyrberg (not verified) on 13 Feb 2011 #permalink

Oh, I agree that most flying organisms appear to be dynamically unstable. To my (limited) knowledge, though, the center of gravity is still pretty close to the center of the body. To use an extreme example, if an animal's center of gravity is well behind its wings, it's basically going to be flying like a helicopter. Its butt (at center of mass) will be swinging like a pendulum, while the wings push air straight down.

The issue for any pregnant flier is that it faces two simultaneous problems. First, she has to increase her metabolic rate and nutrient intake to feed the fetus, and second, she has to continuously adapt her flying style, as fetal weight pulls her center of gravity back. Additionally, the mother has to increase her respiratory rate to supply the fetus with oxygen.

As bats show, this isn't an impossible task. However, it sets a fairly stringent and interesting set of constraints on any viviparous flying species. I think it's reasonable to ask whether, with flying species, there is a tradeoff: are viviparous flying species more strongly morphologically constrained than oviparous species? While bats are quite diverse, they are morphologically conservative compared to both birds and pterosaurs. There are no flightless bats, bipedal bats, bats with extremely long tails, and so forth.

By heteromeles (not verified) on 13 Feb 2011 #permalink

I just want to say how much I enjoy that this blog has scholarly bibliographies in the freakin' comments.

By Squiddhartha (not verified) on 13 Feb 2011 #permalink

It is textbook wisdom that long-tailed birds and long-tailed pterosaurs are interpreted as dynamically stable fliers and short-tailed ones as dynamically unstable ones.

There are no flightless bats, bipedal bats, bats with extremely long tails, and so forth.

There are bats -- Rhinopoma -- with extremely long but also extremely thin tails. I don't know if such a tail weighs enough to matter.

By David MarjanoviÄ (not verified) on 13 Feb 2011 #permalink


There are bats -- Rhinopoma -- with extremely long but also extremely thin tails.

Many early fossil bats (e.g., Icaronycteris) also had long, thin tails.

I don't know if such a tail weighs enough to matter.

That's a good question. Does anyone know of studies on the aerodynamics of flight in Rhinopoma?

The comment about long tails is interesting. As a non-specialist, I have to say: are we talking peacocks, widowbirds, tropic birds or frigate birds? Or macaws? Or are we talking about pheasants? Which ones are dynamically stable? Note that while Rhinopoma has a comparatively long tail, there aren't bat analogs for most of the above.

That's the issue I'm pointing to: bat morphological diversity is far less than that of birds. It's also far less than pterosaur morphological diversity, and I suspect everyone here would agree that pterosaurs are a somewhat better analog for bats than birds are.

Is it just that birds are best, and pterosaurs were diverse because they got into the air before birds did? Or is there some other factor (such as oviparity) which freed both birds and pterosaurs to explore their morphological limits more than bats can?

By heteromeles (not verified) on 14 Feb 2011 #permalink

"There are no flightless bats, bipedal bats, bats with extremely long tails, and so forth."
No really big bats, either.

I've wondered that myself. Bats, however, evolved in a world already overtaken by birds, so they had to fly under the radar, as it were, to avoid direct competition. The only "large" bats (flying foxes) are freed up from size restrictions because of their unique diet (frugivory).

I suspect that if birds ever disappeared, bats would go through a startling evolutionary explosion.

@34: As for bat size, I'm willing to bet that it has to do with the limits of mammalian lung design. I don't know what the upper limit is, but I'll bet that bats come close. Combine a relatively large brain for signal processing and powered flight, and that's a pretty big respiration demand.

@35: I was looking up the evolution of owls and swifts vs. bats, and it looks like owls might have shown up first. Or not. There aren't many fossils of either group. I'd love to make a case for birds suppressing bats or vice versa. All I can say is that bat flight appears to have evolved prior to echolocation (based on the cochlea of Onychonycteris), and some birds have primitive echolocation, so there was a basis for competition, and the bats won their niches. Whether the birds kept bats out of other niches is an interesting question.

By heteromeles (not verified) on 26 Feb 2011 #permalink

On the evolutionary pressure possibly exerted by predatory birds on bats, see...

Rydell, J. & Speakman, J. R. 1995. Evolution of nocturnality in bats: potential competitors and predators during their early history. Biological Journal of the Linnean Society 54, 183-191.

It's rare to find suggestions that one or other pterosaur might have exhibited herd behavior, with an alpha male monopolizing (or anyway extracting rent for) access to multiple females. Does it show up anywhere in birds?


By khazar-khum (not verified) on 26 Feb 2011 #permalink

Darren (re: #37)--
Thank you for the reference to the Rydell & Speakman paper!
---Relevant bit of (alas!) anecdotal evidence: some time in the past quarter century I attended a lecture on something else in Melbourne, and in an offhand remark the speaker said there was an island -- somewhere in the south-western Pacific -- with an avifauna depauperate in predators (i.e. NO HAWKS). And the local species of fruit bat had switched to a largely diurnal lifestyle.
For the sake of discussion, let's assume that's true and pretend it's documented. It would seem to be strong evidence agains one of the hypotheses (flying in the daytime with mammalian physiology would lead to overheating(*)) as we have an example of a fairly large bat flying around in daylight in a fairly warm climate. It also seems to support the hypothesis that the threat of hawk predation is the important determinant of chiropteran nocturnality.
I was taken by the discussion of Pterosaurs: exact timing of the evolution of bats is pretty unconstrained (pity we don't have an early-Paleocene equivalent of Messel!), but I have the sense that these days, a decade and a half after the Rydell & Speakman paper, most workers in the field assume it was most probably Paleocene, not Cretaceous.
(*) Hmm... If I were asked to DESIGN a mammal that could dump heat readily to avoid overheating, I'd think about something to increase surface area. Like elephant ears. Or -- I've got it now! -- patagia!

By Allen Hazen (not verified) on 27 Feb 2011 #permalink


Relevant bit of (alas!) anecdotal evidence: some time in the past quarter century I attended a lecture on something else in Melbourne, and in an offhand remark the speaker said there was an island -- somewhere in the south-western Pacific -- with an avifauna depauperate in predators (i.e. NO HAWKS). And the local species of fruit bat had switched to a largely diurnal lifestyle.
For the sake of discussion, let's assume that's true and pretend it's documented.

No need to pretend; it is documented. ;) The diurnal fruit bat in question is the Samoan flying fox Pteropus samoensis (the peregrine falcon Falco peregrinus, which predates fruit bats on other Pacific islands, is not found in the Samoan archipelago). See:

Cox, P.A. 1983. Observations on the natural history of Samoan bats. Mammalia 47, 519-523.

Thomson, S.C., Brooke, A.P. & Speakman, J.R. 1998. Diurnal activity in the Samoan flying fox, Pteropus samoensis. Philosophical Transactions of the Royal Society of London B 353, 1595-1606.

Thomson, S.C., Brooke, A.P. & Speakman, J.R. 2002. Soaring behaviour in the Samoan flying fox (Pteropus samoensis). Journal of Zoology, London 256, 55-62.


And it isn't just fruit bats that may switch to diurnal habits in the absence of avian predators, either. The Azores noctule Nyctalus azoreum in the Azores islands in the Atlantic is also mostly diurnal. See:

Moore, N.W. 1975. The diurnal flight of the Azorean bat (Nyctalus azoreum) and the avifauna of the Azores. Journal of Zoology, London 177, 483-486.

The Azores noctule Nyctalus azoreum in the Azores islands in the Atlantic is also mostly diurnal.

Slight correction to the above: saying that the Azores noctule is 'mostly' diurnal is probably to put it too strongly. This species is also (and perhaps primarily) active by night (Speakman & Webb, 1993).


Speakman, J.R. & Webb, P.I. 1993. Taxonomy, status and distribution of the Azorean bat (Nyctalus azoreum). Journal of Zoology, London 231, 27-38.

There seems to be pretty good evidence that bats avoid competing with birds. In latitudes with 24 hour summer daylight, bats deliberately forage when the local insectivorous birds are resting - despite the fact that this is when insects are least active, and despite the fact that the local raptors are as active at this time as they are at any other (Moore 2000, Speakman et al. 2000).

Indeed, the otherwise fine hypothesis that bats evolved nocturnality to avoid predation seems flawed, since bats are heavily predated upon by owls and also by falcons and hawks that continue to hunt at night. You might still argue, however, that the risk of predation at night is less than that incurred during the day, and perhaps this was true early on in bat history (though there were lots of owls in the Paleocene and Eocene). I think the real reason that bats are mostly nocturnal is that they evolved from nocturnal ancestors, started their flying at night, and became more 'behaviourally keyed in' to nocturnality more than diurnality.

Refs - -

Moore, P. D. 2000. Bats about the Arctic. Nature 404, 446-447.

Speakman, J. R., Rydell, J., Webb, P. I., Hayes, J. P., Hays, G. C., Hulbert, I. A. R. & McDevitt, R. M. 2000. Activity patterns of insectivorous bats and birds in
northern Scandinavia (69°N), during continuous midsummer daylight. Oikos 88, 75â86.

falcons and hawks that continue to hunt at night

The things I learn on Tet Zoo...

By David MarjanoviÄ (not verified) on 27 Feb 2011 #permalink

I should probably do an article on nocturnality in hawks, eagles and falcons. Virtually all of the data is pretty new - as in, from the 1990s at the earliest. Bald eagles hunt auks in the darkness (sometimes stalking them on foot, as if they're flightless), Common kestrels hunt mice by moonlight, urban peregrines kill birds at night (first reported in the USA in 1990, and soon reported from peregrines worldwide), and so on and on... It's not necessarily 'new' behaviour, it's just stuff that we weren't able to observe before.

The things I learn on Tet Zoo...

Tsk tsk. Your job here on Tet Zoo, Dr. MarjanoviÄ, is to teach, not to learn.

On the subject of nocturnal raptors, the Letter-winged Kite (Elanus scriptus) of Australia is mostly nocturnal.

By Lars Dietz (not verified) on 28 Feb 2011 #permalink

Thanks for the Speakman et al. references.

Now I'm wondering how much avian nocturnality and bat diurnality could be laid down to local culture (ahem!) rather than genetics.

I had a pet pigeon who was willing to be active at night even though he couldn't see, and he taught me that they aren't machines, turning themselves off or on when the lights go out. That (indoor pet) bird would even "fly by braille," putting his beak against the wall and flying up to a perch with his beak on the wall the whole way.

By heteromeles (not verified) on 28 Feb 2011 #permalink

Exploring whether features such as crests represent sexual dimorphism or seperate species was interesting, but I was also drawn to the part about precocial hatchlings, because I have had an emotional prejudice about this theory, possibly because the picture of baby pterosaurs in their nests, attended to by devoted parents, was just too charming. Oh well, I suppose I'll have to adjust my thinking on this matter in light of the new evidence! And on the subject of bats, I wonder if anyone has any ideas as to why bats never reached the extreme proportions of birds and pterosaurs, could their nocturnal habits have played a part in this?

I've written a second article, all about pterosaur nesting behaviour, the ecology of juveniles etc. I can't publish it yet, but hope to do so soon. Watch this space :)

"... was discovered right next to the pelvis of an undisturbed, complete adult pterosaur skeleton. I'd say this is pretty convincing."

It is pretty impressive - like seeing Marsh Warbler's eggs right next to a complete female Cuckoo and being convinced that Cuckoos laid the Marsh Warbler's eggs. Or finding a weasel preserved with the nest it was raiding and being convinced that the weasel laid the eggs.

I'm speaking slightly tongue in cheek, of course, but we only have four specimens. Plenty of animals have connections to the eggs of other species - including raiding nests for food. In the history of fossils we are going to get fooled a few times by things that are pretty convincing but turn out to be wrong !

On a serious note I can't quite understand this part: "While the egg was crushed during burial (as was the accompanying skeleton), its dimensions don't appear to have been altered much". Why not? Surely we'd expect crushing to squash it - squeezing it to be wider to make up for being crushed thinner?

Thanks for the fascinating post !


Mac: parsimony. The Darwinopterus egg might not have an embryo inside (unlike the other known pterosaur eggs), but it matches other specimens in microscopic eggshell texture and so on. Ergo, it is simplest to conclude that it belongs to the pelvis it seems to be emerging from. As for dimensions, it seems to have collapsed fairly neatly upon itself. The possibility of some shape/size change has been admitted.