Field of Science

Ceratopsids: A Cretaceous Flash in the Pan

Leptoceratops, one of the latest-surviving ceratopsians. Reconstruction from here.

After the previous post on ceratopsians, Zach Miller asked if I could follow up my basal-ceratopsian-focused post with one on the more famous ceratopsids, which for various reasons, most significantly time, I had rather neglected.

Sorry, Zach - this is not that post.

But what I thought I would elaborate on was something I referred to offhand in that post about the significance of the basal ceratopsians compared to the ceratopsids proper. I mentioned that the small bipedal ceratopsians, despite their relative obscurity, actually persisted in North America for just as long as the giant ceratopsids, and were with them 'til the end. I would like to add to this that as surprising as it may sound to any readers who are only familiar with popular presentations of evolution and their tragic tendency to fall into the "March of Evolution" trap, this actually wasn't much of an achievement. For even though ceratopsids are one of the iconic dinosaur groups, instantly recognisable by 95% of the developed world's population, they weren't actually around for very long.

Most of you will have probably heard of the Triassic, Jurassic and Cretaceous periods that together make up the Mesozoic era of earth's history. While these periods are quite sufficient for broad generalisations about the history of life, palaeontologists generally find that they need finer-scale divisions to refer to more specific time periods. I recommend going to if you want to see the subdivisions for the Jurassic and Cretaceous, because I'm going to have to refer to a few of them in the course of this post. I know I'll be checking back there regularly as I write this post, because personally I can never keep track of them all.

Yinlong downsi, the earliest known ceratopsian. Reconstruction by Andrey Atuchin.

I mentioned in the previous post that the earliest ceratopsians are known from the Late Jurassic. Specifically, Yinlong downsi comes from the Oxfordian, which started about 161 mya (million years ago). (Taxon ages for this post have been taken from Justin Tweet's Thescelosaurus! website.) In the rough grade system I used in the previous post, Yinlong would be a psittacosaur-grade ceratopsian*, but phylogenetically speaking it is the sister taxon to later ceratopsians. The earliest and most basal known protoceratopoid-grade ceratopsian, Liaoceratops yanzigouensis, comes from early in the Cretaceous, some time between the Valanginian (starting about 140 mya) and the early Barremian (130 mya). The earliest known Ceratopsidae, in contrast, didn't crash the party until the mid-Campanian (perhaps about 78 mya). Ceratopsians had been around for about 80 million years already by the time the ceratopsids appeared. Or to put it another way, less time separates us from the latest ceratopsids than separates the ceratopsids from the earliest ceratopsians! With their extinction at the end of the Cretaceous, about 65 million years ago, the Ceratopsidae are only known to have been around for about 15 million years - an impressively long time by human standards, but really not so very impressive when compared to the nearly 100 million years of ceratopsian history, or the more than 180 million years the Mesozoic lasted for in total.

*You may be wondering how the psittacosaur-grade ceratopsians fared in terms of longevity. Psittacosaurus is the latest well-established psittacosaur, and seems to have survived until about the end of the Early Cretaceous, about 100 mya. However, the analysis of Butler et al. (2008) suggests, albeit with rather low support, that the poorly-known (yet ambitiously-named) Micropachycephalosaurus hongtuyanensis might be a very basal ceratopsian. Despite this basal position, Micropachycephalosaurus actually dates from the Maastrichtian, the very last part of the Cretaceous. If Micropachycephalosaurus is indeed a ceratopsian (a proposition that should be taken very cautiously), then psittacosaur-grade ceratopsians survived at least relictually for almost the entirety of ceratopsian history.

True, the ceratopsids did attempt to cover up for lost time through rapid speciation, so that more ceratopsid taxa have been described from that last fifteen million years than all the non-ceratopsid ceratopsians over 100 million years, but this is like the rapid propagation of any fad - the flashy new designs may temporarily overshadow the old classics in the public eye, but the classics are still very much there.


Butler, R. J., P. Upchurch & D. B. Norman. 2008. The phylogeny of the ornithischian dinosaurs. Journal of Systematic Palaeontology 6 (1): 1-40.


  1. Yeah, it's always a surprise when I tell people that ceratopsid dinosaurs were the "last" to arise. Because Triceratops and its buddies are so iconic, the assumption is that they were around for the whole of the Mesozoic.

    Interestingly, ceratopsids and true duckbills (hadrosaurinae + lambeosaurinae) arose around the same time, and both groups florished briefly, but impressively, in their short life-spans. Gotta wonder what was happening at the end of the Cretaceous that was so helpful to big, showy herbivores.

  2. Personally, I was quite surprised to discover just how late the ceratopsids were. It's like that bit in The Restaurant at the End of the Universe were the Great Prophet Zarquon returns only to have his final message cut off by the end of time fifteen seconds later.

    In terms of why the ceratopsids and hadrosaurids were so successful in such a short time, I do wonder how much the "taxonomic osteology limitation" affects things. Because many of the features that distinguish living species would not be preserved, many fossil "species" could be species groups rather than individual reproductive species. Ceratopsids and lambeosaurines differ from other dinosaur groups in developing ornate osteological structures that are most likely related to display, so the "species" in these groups are more likely to represent the actual separate reproductive species. This can't be the only explanation - it doesn't work for hadrosaurines, for instance - but it would be a complicating factor.


Markup Key:
- <b>bold</b> = bold
- <i>italic</i> = italic
- <a href="">FoS</a> = FoS