Redrawing the Dinosaur Family Tree

So we did it! We finally published our radical new hypothesis! We ‘blew it up’ as Charlton Heston once put it! We are ‘manics!’

After three years of entering data in to our matrix, we, meaning Paul Barrett, Dave Norman and I, finally published the results of our big analysis of early dinosaurs in the journal Nature on March 23rd 2017 (doi:10.1038/nature21700) … arguably the best day of my life so far!!

So what on earth did we do that was so different? What did we see that others did not?

Well, nothing really…

What we did was just look, really, really carefully, at a LOT of early dinosaurs! We compared all the specimens that we chose to include in our study using a lot of anatomical features or characters and this revealed a long overlooked possibility – the way in which we have drawn the dinosaur family tree for the last 130 years could be very, very wrong

This change comes as the result of one thing and one thing only… objectivity…    oh, and size, size matter too, apparently… so two things… two

Most critically, we did not make the assumption that Ornithischia is just the weird outgroup of the other dinosaurs, as so many previous studies have done; we didn’t score the whole clade as a single operational taxonomic unit and we didn’t look at the ornithischian specimens for ornithischian and ancestral dinosaurian features only… in fact if  you look really closely, many ornithischians can actually be scored in the same way as theropod (meat eating) dinosaurs for a number of characters that have previously only been applied to theropods

By not starting our work assuming that Ornithischia and Saurischia (‘lizard-hipped’ dinosaurs) are the only two groups that we could possibly hope to find at a fundamental level, we were able to properly and objectively test the key interrelationships within Dinosauria for the very first time using a big datatset and modern computing power

We made sure that we included a good number of the all-too-often overlooked group Ornithischia (‘bird-hipped’ dinosaurs) and we didn’t dismiss the anatomical similarities between them and other dinosaurs as coincidences. We also made sure that checked other dinosaurs for things that we, under the old scheme, would only expect to find in ornithischians

We added new anatomical characters to the long list of ‘classic’ ones in order to better capture the many observed morphologies that are exporessed by early dinosaurs… we also made a number of observations of anatomical features that could potentially distinguish between or unite groups within the larger dinosaur family  — however, our results can be reproduced with all of our new characters turned off in TNT, just in case you were wondering

Then we come to size. It does matter, in this sense… more species included means more material for comparisons and more characters means more ways for comparisons to be be made – with  roughly 35,000 individual data points, our analysis is around four times bigger than its nearest competitor

We built it… it nearly killed me… but we built it

UK, South Africa, Berlin, USA, South Africa, USA, Argentina, USA, UK… to list the countries that I personally visited, in order of arrival… I actually had one month in 2015 in which I walked on four continents in under four weeks… and China, Paul Barrett has seen a lot of the species in China… so we had a very good global coverage

Then, thorough cross-examination of the many early dinosaurs that we had seen and read about, we were able to assess how they may related to each other and produced the shock result which has left a few people in our field dumbstruck

This is just a first quick post about it and I will be writing more in the coming months about how we assembled the data, what analyses we did, what results we got and how the world has reacted

A summary of the internet’s reaction is next on my agenda as it has been one of the most unexpected but enjoyable aspects of this whole experience

For now I will sign off as I need sleep… well earned, deep, long sleep… before I go back to writing up my thesis

I expect to be Dr Baron by summer… so wish me luck!


Matt Baron