Dividing the dinosaurs

The standard dinosaur evolutionary tree has two key branches: the 'bird-hipped' Ornithischia and the 'reptile-hipped' Saurischia. A revised tree challenges many ideas about the relationships between dinosaur groups. See Article p.501

Every child obsessed with dinosaurs knows that these creatures are grouped into the 'bird-hipped' Ornithischia (which includes the herbivorous stegosaurs, ceratopsians and duckbills) and the 'reptile-hipped' Saurischia, which are subdivided into the carnivorous Theropoda (such as Tyrannosaurus rex and birds) and the herbivorous Sauropoda (think Brontosaurus)1. Although this dinosaur family-tree arrangement seems universally acknowledged by everyone from children to wizened palaeontologists, we might all have to rearrange our mental furniture. On page 501, Baron et al.2 present a revolutionary proposal that the relationships of the major groups in the dinosaur evolutionary tree should be radically reorganized.

The clade Dinosauria was named and described3 by the palaeontologist Richard Owen in 1842. However, in 1888, Harry Govier Seeley noticed such glaring differences between dinosaurs4 that he divided them into two groups named after differences in their pelvic bones, and, for a century, Dinosauria was not recognized as a natural group that evolved from a single common ancestor5. This view changed in the 1980s, when the palaeontologist Jacques Gauthier showed that dinosaurs form a single group, which collectively has specific diagnostic traits that set them apart from all other animals1. So both Owen and Seeley were right. And there it pretty much stood — until this issue of Nature.

For decades, the reconstruction of evolutionary trees in all branches of life has been standardized by the use of a phylogenetic systematic approach that insists on recognizing natural groups only by newly evolved traits that their members uniquely share6. Baron et al. use this analytical method, but reach different conclusions from those of previous studies by incorporating some different traits and reframing others. Because the authors followed standard methods, their results cannot be dismissed as simply a different opinion or speculation. Instead, the trait analyses they used will need to be scrutinized in minute detail by researchers. But this will also be true of the traits used for previous analyses. And that, in the end, can only be good for the field.

The detailed phylogenetic analysis conducted by Baron et al. builds on previous analyses by others, but is original in many respects. They arrived at a dinosaur evolutionary tree (Fig. 1) containing one main branch that subdivides into the groupings of Ornithischia and Theropoda, and a second main branch that contains the Sauropoda and another, more archaic group, the carnivorous Herrerasauridae (usually positioned as either basal theropods or basal Saurischia, or outside Dinosauria but close to it)1,5,7.

Figure 1: The dinosaur evolutionary tree.

a, A traditional dinosaur evolutionary tree places Sauropoda (such as Brontosaurus) and Theropoda (such as Tyrannosaurus rex) in the branch known as Saurischia. The other main branch of the Dinosauria tree is Ornithischia (which includes Stegosaurus). The group Herrerasauridae is potentially placed as basal theropods, basal saurischians or basal dinosaurs1,5,7, as indicated by dotted lines and question marks. b, Baron et al.2 propose a revised dinosaur evolutionary tree in which one main branch contains Herrerasauridae and Sauropoda, and the other main branch contains Theropoda and Ornithischia.

Baron and colleagues' evolutionary tree is unique, and here's why. In this dinosaur family-grouping game, the joker in the pack has always been Ornithischia. From their earliest appearance, they have been weird. They have a strange additional chin bone, their incisor teeth are smaller than those of other dinosaurs, their cheek teeth are regular and closely spaced like molars, they have beaks, and their hip bones are enigmatically organized7. Also, unlike nearly all the other dinosaurs except Sauropoda, they are clearly herbivores, as their teeth and jaws show.

But weirdest of all, Ornithischia don't begin to diversify substantially until the Early Jurassic, about 200 million years ago, or maybe by the 'last party weekend' of the preceding Triassic period. By contrast, the other dinosaurian groups already existed by at least the early Late Triassic, some 30 million years earlier. The single specimen of the small Pisanosaurus from the Triassic of Argentina is conventionally regarded as an ornithischian7, but it is incomplete and poorly preserved, and has been something of a Rorschach ink-blot interpretation test for decades. And although Ornithischia don't establish a firm presence until the Early Jurassic, they then seem to be everywhere at once8.

The suspicion has long been that ornithischians might have evolved from another pre-existing dinosaur group. But if so, which one? Sauropods have been proposed because both groups are herbivorous. However, a comparison of their skulls, teeth and skeletal structures suggests that sauropods and ornithischians evolved their modes of herbivory quite differently. Baron et al. find that theropods and ornithischians form a natural group, which could imply that if the impoverished Triassic record of ornithischians reflects a true absence, ornithischians might have evolved from theropods in the Late Triassic (although the authors do not go as far as to make this inference).

Also puzzling is Baron and colleagues' finding that the primitive-looking herrerasaurids, from the South American Triassic, are the sister group to the sauropods. This link is not strongly supported, but it is intriguing. Herrerasaurids were carnivores, and they are usually linked to or included with the carnivorous theropods7. Baron et al. suggest that this is an example of the independent evolution of the same trait, known as convergence, and certainly, the 'hypercarnivorous' features of large skulls and teeth evolved independently in the two groups.

However, a more moderate set of carnivorous or even omnivorous dental features might have been the ancestral condition for all of these groups. Of the groups closely related to but outside Dinosauria that were used for comparison, most of these, including most archosaurs, were carnivorous9. It seems probable that the herbivory of the classic ornithischians and sauropods evolved independently, a conclusion that also applies to the traditional dinosaur groupings. Baron et al. conclude conservatively that omnivory might be the basal dinosaurian condition because the teeth in the first members of these groups have different shapes (heterodont). This idea seems reasonable because many small animals with heterodont teeth are omnivorous; but some small mammals that are herbivorous or carnivorous also have heterodont teeth, so the assumption of omnivory might not be the final word on this matter.

There are other implications of Baron and colleagues' analysis. The authors find, consistent with most other studies, that early dinosaurs, like their closest non-dinosaurian relatives (including pterosaurs), were small, bipedal and had grasping hands10. They suggest that dinosaurs might have originated in the Northern Hemisphere, because most of their basal members, as well as their close relatives, known as outgroups, are found there. This is striking because, for decades, South America was thought to be the cradle of dinosaur evolution, thanks to the presence of herrerasaurids and other close dinosaurian outgroups11. The authors suggest that more such basal relatives will be found in northern continents and, to some extent, their predictions have already been validated by fossil discoveries12. It will be interesting to see how palaeontologists receive this original and provocative reassessment of dinosaur origins and relationships. Footnote 1


  1. 1.

    See all news & views


  1. 1

    Gauthier, J. in The Origin of Birds and the Evolution of Flight (ed. Padian, K.) 1–55 (California Acad. Sci., 1986).

  2. 2

    Baron, M. G., Norman, D. B. & Barrett, P. M. Nature 543, 501–506 (2017).

  3. 3

    Owen, R. in Report of the Eleventh Meeting of the British Association for the Advancement of Science 60–204 (Murray, 1842); see go.nature/2nwklvb

  4. 4

    Seeley, H. G. Proc. R. Soc. Lond. 43, 165–171 (1887–88).

  5. 5

    Padian, K. Earth Environ. Sci. 103, 423–442 (2013).

  6. 6

    Baum, D. (2008).

  7. 7

    Weishampel, D. B., Dodson, P. & Osmolska, H. (eds) The Dinosauria 2nd edn (Univ. California Press, 2007).

  8. 8

    Irmis, R. B., Parker, W. G., Nesbitt, S. J. & Lui, J. Hist. Biol. 19, 3–22 (2007).

  9. 9

    Nesbitt, S. J. et al. Nature 464, 95–98 (2010).

  10. 10

    Padian, K. Zitteliana 28B, 21–28 (2008).

  11. 11

    Ezcurra, M. D. J. Syst. Palaeontol. 8, 371–425 (2010).

  12. 12

    Irmis, R. B. et al. Science 317, 358–361 (2007).

Download references

Author information

Correspondence to Kevin Padian.

Related audio

Reporter Shamini Bundell finds out about the changes to the dinosaur family tree.

Related links

Related links

Related links in Nature Research

Palaeontology: Dinosaur up in the air

Palaeontology: Mystery of the horrible hands solved

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Padian, K. Dividing the dinosaurs. Nature 543, 494–495 (2017).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.