Medullary Bone and the Dinosaur-Bird Link

One of the coolest lines of evidence that birds are extant dinosaurs is the presence of medullary bone in multiple dinosaur species. Medullary bone (hereafter referred to as MB, to avoid confusion with the medullary surface) is a temporary tissue that forms on the interior surfaces of the long bones of birds. MB is identified by unique collagen organization: it is both densely mineralized and strongly vascularized. This structure helps MB serve its purpose as a readily retrieved source of calcium for use in forming eggshells, and prevents incapacitating bone resorption during this process. Among extant animals, MB is only found in mature female birds in the process of producing eggs. Its creation is triggered by hormones during the onset of ovulation, and it disappears during the laying process. Among extant animals, MB is only known in birds. However, in 2005 Mary Schweitzer and colleagues reported their discovery of medullary bone in a Tyrannosaurus rex individual. Lee and Werning followed up on this research in 2007 by reporting MB in the theropod Allosaurus and the ornithopod Tenontosaurus.

Since MB is unique to reproductively active females, most popular coverage of dinosaur MB  has focused on its potential use for determining the sex and life stage of individual dinosaur specimens. We shouldn’t, however, lose sight of the fact that MB is an independent line of evidence supporting a close phylogenetic relationship between dinosaurs and birds. Nearly all paleontologists agree that the evidence that birds are dinosaurs is overwhelming, and MB is but a drop in the ocean of shared characters between birds and dinosaurs. Nevertheless, it is noteworthy that few authors have attempted to challenge Schweitzer’s initial publication.

The only work I have found  that disputes Schweitzer and colleagues is the dissertation of Dr. Devon Quick  (.pdf link), in which Dr. Quick and Dr. John Ruben investigated the reliability of the methods used to recognize MB in the fossil record using extant animals. This is not, incidentally, the only work by Quick and Ruben challenging the dinosaur-bird connection. As a doe-eyed student, I’d like to take a shot at reviewing this paper. And since I’m posting it publicly, I of course welcome anyone who’d be so kind as to call me out for being wrong.

Quick and Ruben looked at cross-sections of the femora and tibiotarsi of a crocodilian (Alligator mississippiensis) and several birds. Scanning electron microscopy revealed that the medullary surfaces of the tibiotarsi of  reproductively active birds displayed the highly contoured and floccular texture that is characteristic of MB. Likewise, the male and non-reproductively active female birds displayed smooth medullary surfaces. In this regard, Quick and Ruben are in agreement with previous work. However, the authors also reported that the medullary cavity of the alligator femur contained “material superficially similar to…avian medullary bone” (Quick and Ruben 2008). This material was limited to the immediate diaphyseal side of the metaphysis, making it much less extensive than what was observed in birds. Since the alligator individual used in the study was a juvenile male, it was almost certainly not producing reproductively-specific MB. From this observation, the authors conclude from these data that a floccular texture may indicate early-stage bone mineralization and is not a reliable indicator of MB.

Quick and Ruben’s results are unconvincing in part due to a weak experimental design. Their conclusions are dependent on observations gleaned from a single alligator specimen, which is not an adequate sample. The authors’ conclusions would carry more weight if they had looked at multiple individuals. It would also be beneficial to compare males, females, adults and juveniles. Ideally, additionally crocodilian species ought to be included in the study, as well. Schweitzer and colleagues carried out a similar investigation, in which they looked for evidence of MB in multiple alligators, including gravid females, males and juveniles (Schweitzer et al. 2007). Schweitzer and colleagues found no evidence of MB, even with estrogen stimulation, and their larger sample size allows their study to carry more weight than that of Quick and Ruben. Furthermore, although Quick and Ruben assert that that “histological aspects of Tyrannosaurus tissues that are supposedly consistent with an avian-style reproductive physiology were not analyzed carefully”, they did not look at the Tyrannosaurus material as part of their study. Accordingly, no evidence is provided that the structures the authors observed on their alligator were synonymous with those observed by Schweitzer and colleagues on Tyrannosaurus. Finally, Quick and Ruben’s observations are focused on the floccular texture used to identify MB, when in fact Schweitzer and colleagues used several other indicators, including extensive vascularization, to identify MB in Tyrannosaurus. It is notable that the structure, thickness and texture of MB in modern birds varies considerably based on the specifics of the animal’s reproductive biology and the size of the taxa. Given that Tyrannosaurus is several orders of magnitude larger than most extant birds, some structural difference is to be expected (wow, that sentence had some serious science snark).

Quick and Ruben suggest that the floccular texture on the alligator bone may be the result of early-stage mineralization, which would be consistent with the sub-adult status of the individual they used in the study. The authors go on to speculate that a similar explanation might account for the evidence of MB in Tyrannosaurus. Again, it would have been helpful if the authors had amassed more examples of sub-adult archosaurs undergoing skeletal mineralization, and compared them directly to the Tyrannosaurus material in question, rather than merely speculating. If the Tyrannosaurus was forming MB, this would be consistent with information from lines of arrested growth in Tyrannosaurus and other dinosaurs, which indicates that dinosaurs became reproductively active before reaching adult size.

Having reached the somewhat tenuous conclusion that texture is not a reliable indicator of MB, Quick and Ruben go on to argue that even if MB is present in dinosaurs, the fact that it has been reported in both saurischians and ornithiscians “offers no particular insight into the phylogenetic origins of birds.” On the contrary, MB is an independently observable feature that unites the crown group Dinosauria with Avialae, and therefore supports the consensus that Avialae is bracketed by Dinosauria. At the very least, MB suggests marked similarity in reproductive strategies employed by birds and dinosaurs. As demonstrated by Schweitzer and colleagues, MB is not known in crocodilians. Quick and Ruben freely admit this, which makes their statement that MB “may well be a plesiomorphic trait that first evolved in basal archosaurs” nonsensical (Quick and Ruben 2008). The authors could theoretically argue that MB production is primitive but was lost in modern crocodilians, but there is no evidence for this.

Overall, Quick and Ruben’s work is hindered by weak experimental design and vague, unsupported conclusions. Given that a similar but more rigorous study regarding MB in crocodilians has already been carried out by Schweitzer and colleagues, Quick and Ruben’s interpretations are not convincing. Even the broadest interpretation of the available evidence indicates that MB originated after the divergence of crocodilymorphs from the main archosaur line. The phylogeny postulated by Schweitzer and colleagues remains most tenable, in which MB originated in early dinosaurs, and was inherited by ornithiscians, tyrannosaurids and modern birds (Schweitzer et al. 2005).

References

Lee, A. H. and Werning, S. “Sexual maturity in growing dinosaurs does not fit reptilian growth models.” 2007. PNAS 105:2:582-587.

Quick, D. E. and Ruben, J. A. “Amniote bone structure and longbone histology in birds, alligators and the theropod Tyrannosaurus rex.” 2008. Oregon State University.

Schweitzer, M. H., Elsey, R. M., Dacke, C. G., Horner, J. R. and Lamm, E. T. “Do egg-laying crocodilian (Alligator mississippiensis) archosaurs form medullary bone?” 2007. Bone 40: 1152-1158.

Schweitzer, M. H., Wittmeyer, J. L. and Horner, J. R. “Gender-Specific Reproductive Tissue in Ratites and Tyrannosaurus rex.”2005. Science 308: 1456-1460.