Like modern birds, and like their close relatives among the theropod dinosaurs, the birds of the Mesozoic Era laid eggs and, we reasonably infer, made nests. But what else do we know about reproductive behaviour in Mesozoic birds? Essentially, we know very little, and by “very little” I actually mean “just about nothing”. A new paper just published in Naturwissenschaften by Gareth Dyke, Mátyás Vremir, Gary Kaiser and myself describes an assemblage of bird eggshell and bones that is amazing in its scale, and fascinating in terms of its behavioural implications (Dyke et al. 2012). It shows, for the first time, that some Mesozoic bird species - only distantly related to modern birds - formed enormous breeding colonies.

A few embryos and eggs belonging to the Mesozoic bird group known as Enantiornithines are already known (Elzanowski 1981, Mikhailov 1996, Grellet-Tinner & Norell 2002, Schweitzer et al. 2002, Zhou & Zhang 2004).

The most impressive of these fossils has to be the embryonic, partially feathered Lower Cretaceous Chinese enantiornithine [shown here], discovered within a near-complete egg, and described by Zhou & Zhang (2004). We can infer from the proportions and bone anatomy of this embryo and others that these bird species were precocial or even superprecocial: that is, the hatchlings were able to walk, fly and generally look after themselves within a day, or even within a few hours, of hatching.

But this doesn’t tell us anything about parental behaviour, since parents can still interact substantially with chicks in species where the chicks are precocial (though they don’t in all cases). And most of the other things you might ask about reproductive behaviour in Mesozoic birds remain unknown and often based on inference. Were egg clutches large or small? Where were nests located? What were the nests like? Did adults brood their eggs, or bury them? Did Mesozoic birds nest as lone individuals, in groups, or in colonies?

Given that Mesozoic birds were diverse in terms of body size and shape, and in ecology, they would presumably have been diverse in reproductive biology, just as modern birds are. Indeed, given that the Mesozoic members of Avialae (the bird lineage of Maniraptora) include early forms extremely similar to the little deinonychosaurs that were their close relatives, flightless walkers and runners, flightless, toothed seabirds, finch-, thrush-, crow- and vulture-sized terrestrial species, and early members of the ratite, duck, gamebird and neoavian lineages, it’s plausible that Mesozoic birds possessed a spectrum of reproductive behaviour, extending from ‘dinosaur-like’ at one end to ‘modern bird-like’ at the other.

Romania: fossil treasure trove

Our new research concerns a discovery made in the latest Cretaceous (Maastrichtian) Sebeş* Formation of Transylvania in western Romania. Already the Sebeş Formation (long erroneously considered Oligocene or Miocene in age!) is well known as a source of diverse Late Cretaceous fossils, including pleurodiran turtles, azhdarchid pterosaurs, ornithopods and the amazing dromaeosaurid theropod Balaur bondoc (Csiki et al. 2010, Vremir 2010).

* Pronounced something like “seb-esh”.

The sediments here were deposited by a large, meandering river system, surrounded by floodplains and forests with large trees. This was a continental environment*, with sedimentological evidence showing that long dry seasons were punctuated by short wet ones when extreme flooding sometimes occurred.

* This doesn’t mean 'on a continent’, since the region was actually a large island at the time. Rather, it means that the environment was well inland, and without any estuarine or marine influence.

It’s the detailed mapping and exploration of the Sebeş Formation by Mátyás Vremir of the Transylvanian Museum Society that has resulted in so many amazing Sebeş Formation discoveries. And it's one of these that forms the focus of our new paper. In sediments exposed on the banks of the Sebeş River, Mátyás discovered a large, lens-shaped chunk of calcareous mudstone packed full of thousands of broken eggshell fragments. Unfortunately (but – I hope – understandably), the concretion couldn’t be extracted as a single mass; rather, it broke into lots of separate chunks. When complete it was about 80 cm long, 50 cm wide and 20 cm thick. Its breakage isn’t a bad thing, since this allows the interior of the mass (rather than merely its edges) to be examined in detail. While virtually all of the eggshell within the accumulation is preserved as small, broken fragments, seven near-complete eggshells are included as well, as are more than 60 small bones. We term this mass of sediment, eggshell and bones the ‘Od accumulation’ after the specific outcrop where it was discovered (the Oarda de Jos site).

A whole lot of eggshell

The volume of eggshell in the Od accumulation is extraordinary. About 70-80% of the accumulation as a whole is formed by eggshell, meaning that this is an eggshell-supported rock, reminiscent of the Patagonian ‘egg beds’ that incorporate vast quantities of sauropod eggshell. Averaged out, most shell fragments in the Od accumulation are about 36 mm long. They’re so closely packed that, in one 26-cm-sq section of the accumulation, more than 150 eggshell fragments are present. The several complete eggs preserved in the accumulation are 40 mm long and 25 mm wide; based on these measurements, each 100 cubic cm of the accumulation contains eggshell equivalent to about 46 whole eggs (Dyke et al. 2012). As we say in the paper, “The quantity of eggshell fragments in the Od accumulation is astonishing and beyond normal palaeontological experience”.

The eggshell looks avialan, and the isolated bones jumbled up within the accumulation are definitely avialan. These bones appear to belong to both adult and juvenile birds: they reveal a number of characters that allow them to be identified as those of enantiornithines, in particular a taxon close to (or congeneric with) the Late Cretaceus Argentinean form Enantiornis (Dyke et al. 2012). SEM analysis of the eggshell fragments reveals two equally thick shell layers formed of calcite crystals, as well as an absence of marked surface ornamentation. This eggshell form is typical of maniraptorans that are closer to crown-birds than to deinonychosaurs but outside of Neornithes, the crown-bird clade (Grellet-Tinner et al. 2006). Several exclusively Mesozoic bird lineages, enantiornithines among them, fit within this phylogenetic bracket. The data on eggshell microstructure is thus consistent with the osteological data from the accumulation (Dyke et al. 2012). Before anyone else says it, I should note that the phylogenetic distribution of certain eggshell characters remains the topic of debate - there are even cases where eggs confidently identified as those of maniraptoran theropods (Buffetaut et al. 2005) have turned out to be from squamates.

A waterside enantiornithine colony

We therefore hypothesise that the accumulation preserves the remains of an enantiornithine nesting colony, with the numerous eggshell fragments, the several complete eggs, and the various bones all belonging to the same one species.

Given the volume of eggshell, the colony must have been large. Assuming clutches similar in size to those of modern birds, we’re talking about hundreds of nests. The stacked, packed, jumbled nature of the eggshell fragments removes the possibility of this being an in-situ shared ‘mega-nest’ by the way (a bizarre possibility mooted by one reviewer!). There are no fossils of any other sort within the accumulation. It seems that a flood event swept across the colony, carrying eggs and birds (or their remains) before dumping them in a hollow area a short distance from the main channel (Dyke et al. 2012). This transport can only have occurred across a short distance (as in, several metres), since the bones are unabraded and several eggs are intact.

The form of the eggs is typical of ground-nesters that produce precocial young (Dyke et al. 2012). The absence of vegetation in the accumulation implies that the birds formed scrapes in the sediment for their eggs, as do plovers and some other waterside-nesting modern birds. [Adjacent image of Ringed plover eggs by Arnoldius.]

Given the amount of eggshell, why aren’t there more bones in the accumulation? The inner surfaces of the eggshell fragments are dull and etched, “suggesting that calcium mobilisation was advanced and that the eggs had either hatched or were destroyed late in the second half of incubation” (Dyke et al. 2012). In other words, it’s likely that many or most of the eggs present in the colony at the time of the flooding event were already hatched and lying, empty and discarded, in or close to their nests. This probably wasn’t the case for all nests, since the presence of both juvenile and adult bones implies that these birds perished when the colony flooded. Then again, it’s also conceivable that these bones belong to individuals that were already dead before the colony was swamped – it’s common to see the remains of dead chicks and even dead adults in nesting colonies.

Assuming that we’re right in our overall interpretation, what does the discovery mean for enantiornithine nesting and breeding behaviour? A waterside nesting habit for an Enantiornis-like enantiornithine is not all that surprising, since there's already morphological and stomach-content data showing that some members of this group were aquatic foragers, preying on crustaceans and other such prey (Sanz & Buscalioni 1992, Sanz et al. 1996). The fact that such a large nesting colony of this one enantiornithine species occurred in a waterside habitat strongly suggests reliance on aquatic resources, so the species concerned might have been gull-like or plover-like in ecology. [Adjacent photo of gannet colony by Octagon.]

And, like gulls, waterfowl, flamingos and other waterside, colonial nesters, it seems that these enantiornithines were sometimes unfortunate, and that local flooding events swamped and drowned their nest colonies (Dyke et al. 2012). Flooding occurs quite regularly in some modern bird colonies and is an expected hazard of nesting so close to water. There are even cases where the same colony gets flooded repeatedly during the same one breeding season (e.g., Peresbarbosa & Mellink 2001), and yet still the birds continue to nest there. Why do they nest in such dangerous places? Because they are otherwise highly suitable; more so than adjacent, well-vegetated regions in providing barren, relatively predator-free areas that are suitable for nesting, and are relatively close to the aquatic environment.

On the one hand, the formation of large, waterside nesting colonies in Mesozoic birds is not all that surprising, since this behaviour is common and widespread in modern birds and has clearly evolved many times independently. On the other hand, enantiornithines are not typically imagined as being all that similar to modern, colony-nesting birds like gulls, terns, gannets, penguins and so on, and I’m not sure that I’d have guessed a colony-nesting habit for any enantiornithine prior to the discovery of the Od accumulation. For Mesozoic seabirds, like hesperornithines and Ichthyornis, sure, but – for enantiornithines – it’s a novel idea.

So we can now say that the latest Cretaceous Transylvanian Basin fauna of Romania was inhabited by peculiar, island-endemic dromaeosaurs, titanosaurian sauropods, both rhabdodontid and hadrosaurid ornithopods, azhdarchid pterosaurs, eusuchian crocodyliforms, pleurodiran turtles, and enantiornithine birds that formed enormous, waterside nesting colonies. More exciting Romanian finds are due to be announced in the near future. Special thanks to the outstanding Julio Lacerda for the excellent artwork he produced for this project. Julio has a DeviantArt page here and blogs at The Casual Paleoartist.

For previous Tet Zoo articles on Mesozoic birds, see…

Refs - -

Buffetaut, E., Grellet-Tinner, G., Suteethorn, V., Cuny, G., Tong, H., Košir, A., Cavin, L., Chitsing, S., Griffiths, P. J., Tabouelle, J. & Le Loeuff, J. 2005. Minute theropod eggs and embryo from the Lower Cretaceous of Thailand and the dinosaur-bird transition. Naturwissenschaften 92, 477-482.

Csiki, Z., Vermir, M., Brusatte, S. L., Norell, M. A. 2010. An aberrant island-dwelling theropod dinosaur from the Late Cretaceous of Romania. Proceedings of the National Academy of Sciences 107, 15357-15361.

Dyke, G. Vremir, M. Kaiser, G. & Naish, D. 2012. A drowned Mesozoic bird breeding colony from the Late Cretaceous of Transylvania. Naturwissenschaften DOI:10.1007/s00114-012-0917-1

Elzanowski, A. 1981. Embryonic skeletons from the Late Cretaceous of Mongolia. Palaeontologica Polonica 42, 147-179.

Grellet-Tinner, G. & Norell, M. A. 2002. An avian egg from the Campanian of Bayn Dzak, Mongolia. Journal of Vertebrate Paleontology 22, 719-721.

-., Chiappe, L. M., Norell, M., Bottjer, D. 2006. Dinosaur eggs and nesting behaviours: A paleobiological study. Palaeogeography, Palaeoclimatology, Palaeoecology 232, 294-321.

Mikhailov. K. E. 1996. Bird eggs in the Upper Cretaceous of Mongolia. Palaeontology Journal 30, 114-116.

Peresbarbosa, E. & Mellink, E. 2001. Nesting waterbirds of Isla Montague, northern Gulf of California, México: loss of eggs due to predation and flooding, 1993-1994. The International Journal of Waterbird Biology 24, 265-271.

Sanz, J. L. & Buscalioni, A. D. 1992. A new bird from the Early Cretaceousof Las Hoyas, Spain, and the early radiation of birds. Palaeontology 35, 829-845.

- ., Chiappe, L. M., Pérez-Moreno, B., Buscalioni, A. D., Moratalla, J. J.,Ortega, F., Poyata-Ariza, F. J. 1996. An Early Cretaceous bird fromSpain and its implications for the evolution of avian flight. Nature 382, 442-445.

Schweitzer, M. H., Jackson, F. D., Chiappe, L. M., Schmitt, J. G., Calvo, J. O. & Rubilar, D. E. 2002. Late Cretaceous avian eggs with embryos from Argentina. Journal of Vertebrate Paleontology 22, 191-195.

Vremir, M. 2010. New faunal elements from the Late Cretaceous (Maastrichtian) continental deposits of Sebeş area (Transylvania). Acta Musei Sabesiensis 2, 635-684.

Zhou, Z. (2004). A Precocial Avian Embryo from the Lower Cretaceous of China Science, 306 (5696), 653-653 DOI: 10.1126/science.1100000