When I worked at the Denver Museum of Nature and Science as a volunteer in the educations collection (if you ever get the chance to volunteer at a museum, do it, it was a lot of fun and very educational), I happened upon a man with a young boy by his side, whom I presume was his son, looking at an impressive skeleton. I thought how great it was the man took time to bring his son to the museum. But when the boy asked what the skeleton was and the man answered, “Allosaurus rex,” my opinion of the experience dropped. I’m still glad he brought the boy to the museum, but it could have been so much better.

Allosaurus rex? Actually, a blue whale. Definitely NOT a dinosaur. Also, not DMNS. Really, the London Natural History Museum. Photo by author. Click to enlarge.

So what was wrong with what he said? There is no such creature as an “Allosaurus rex.” There is an Allosaurus and there is a Tyrannosaurus rex, but not the two together. It also did not help that the man was standing directly in front of the plaque that read, “Fin whale.”

Much has been written on what a dinosaur is and what is not, but considering the extreme levels of confusion in the general populace, I thought it worth discussing it here. Not everything that is a big skeleton and/or extinct is a dinosaur. In addition to the above whale, I have heard dimetrodons, pterosaurs, saber-toothed cats, mammoths, giant rhinos, mosasaurs, modern elephant skeletons and many other things called dinosaurs, when in reality, NONE of the previous creatures qualify as dinosaurs.

So, if none of those are dinosaurs, what is? What are some of the things people think about when they try to define dinosaurs? The first thing that most people think about is a giant, scaly reptile. However, some dinosaurs were the size of chickens, so not all of them were big. Many also had feathers, so the scaly motif is not altogether correct either. They are reptiles, but unlike anything most people would consider a reptile today.

Quetzalcoatlus northropi. UT Museum. Not a dinosaur either. Photo by author.

Other traits people often use to distinguish dinosaurs are that none of them were aquatic and none of them flew. These aren’t really true either. Some will say they only lived in the Mesozoic Era and died out at the end of the Cretaceous Period  65 million years ago. Again, false. It is true that most animals considered dinosaurs did indeed live only in the Mesozoic Era, but not all died out at the famous K-T extinction event. Some of you may be thinking to yourselves the author has no idea what he is talking about. Nevertheless, as we will see, some dinosaurs are quite at home in both air and water, especially the ones that lived past the Mesozoic.

Even if all these things were true about dinosaurs, none of it really matters. These traits are all distinguishing characteristics that are generally true about most dinosaurs, but not necessarily true about all. What defines an organism is not the same as the diagnosis. Definitions are done by evolutionary relationships. Diagnoses tell us how we can recognize them, what distinguishing characteristics can be used. However, distinguishing characteristics do not necessarily describe all the members of the group. For instance, one might distinguish all members of a family by their last name, but if people get married and change their last name, they do not suddenly stop being a member of the family. A daughter, for example, is defined by who her parents are, not by what name she has. Likewise, her distinguishing characteristics tell us who she is as an individual and can give us clues to her family relationships, but her definition as a daughter depends solely on her relationship to her parents.

The discredited and completely unaccepted haemothermia hypothesis: Mammals and birds are both warm-blooded; therefore, they must be related. I.e., superficial similarities can be misleading. Image from wikipedia.

I should note here that it wasn’t always this way. Back in the old days, organisms were grouped by similarity, who looked like whom, so at that time, there was really no difference between definition and diagnosis. But as people studied organisms more closely, they found that a lot of times, superficial similarities masked deeper differences, which indicated they weren’t really related at all. So for most of the last century, scientists have tried to find real, evolutionarily related groupings to build something akin to a geneology of life. To do so, it became evident that definitions and diagnoses of groups had to be different, so for the last couple of decades, definitions have been based on relationships and diagnoses based on characteristics. This system of classification by defining groups according to their relationships is called phylogenetic systematics.

So, how then do we define a dinosaur? The standard modern definition of a dinosaur has been stated as the most recent common ancestor of Triceratops and modern birds and all of its descendants (Padian, 1997). Note this means that all birds are, by definition, dinosaurs. So it is obvious that some dinosaurs are quite small (e.g. hummingbirds), flew (most birds), and could reasonably be called aquatic (e.g. penguins). The reason for this is because the earliest birds are clearly closely related to animals, such as Velociraptor and its close relatives, which are indisputably dinosaurs. If one took a picture of Archaeopteryx which had no feathers to a bunch of elementary school kids and asked them to identify it, the general answer is that it is obviously a dinosaur, but put feathers on it and they call it a bird. They are thus both correct answers. Dr. Thomas Holtz has proposed a slightly different definition: the most recent common ancestor of Megalosaurus and Iguanodon and all of its descendants. Both definitions encompass the same groups of animals as far as our understanding goes, but Holtz’s definition includes the dinosaurs that were first discovered, so is considered by many to be a superior definition.

Dinosaur family tree, by Nobu Tamura (paleocritti.com). Click to enlarge. Note on trees like this, the branches can be pivoted on their branch points without changing relationships (e.g.,it makes no difference whether you say “your brother and sister” or “your sister and brother”), so anything in Saurischia is equally related to anything in Ornithischia. Thus, it doesn’t really matter what animal in each group you choose as the representative of each group as long as the groups as a whole remain intact.

Unfortunately, scientists don’t always talk about these terms correctly either, making the whole process confusing. Michael Benton, a leading paleontology researcher with numerous well-respected publications, got it wrong when he defined dinosaurs according to synapomorphies in The Complete Dinosaur. While synapomorphies are used in developing modern classifications, they are used for diagnoses, not definitions of groups, which Dr. Benton assuredly knows (as evidenced by the fact he normally states it correctly in his other works), but when someone is as prolific as Dr. Benton, the occasional error is bound to slip in now and then.

Synapomorphies are very important in figuring out relationships. Unlike any old similarity, synapomorphies are shared, derived characteristics, meaning that the character is the same in the organisms being compared because they share a common ancestor, i.e. they are derived from the same source. Of course, because it requires knowledge of the common ancestor, synapomorphies can only be identified AFTER one has a hypothetical relationship. Thus, to discover strong relationships, scientists will create (mathematically, using a computer because it is way too complex to try doing by hand for anything beyond a few species), all the possible relationships between all the organisms under consideration and then map all the characteristics they can onto every possibility. The relationships requiring the fewest inconsistencies and providing the simplest explanation is considered the most likely. Inconsistencies can occur due to convergent evolution (organisms not closely related developing similar characteristics due to similarities in environmental constraints) or homoplasies (characteristics changing back to a previous form), but with examination of enough characteristics, good relationships usually appear. Of course, the more data you can put into the analysis, the better the results normally get, which is why we continue to study and try to find new fossils and collect more data (and who wouldn’t want more fossils?).

This is all well and good, but how then do we recognize a dinosaur when we see one? Dr. Benton provided a good list of skeletal characteristics, which really won’t mean a lot to people who are not very familiar with skeletons and scientific terminology (but a basic description can be found here and Wikipedia has a surprisingly in depth description). Nevertheless, there are some generalities we can make. If we exclude birds, we can say the non-avian dinosaurs (that we know of) died out at the end of the Cretaceous Period. They all carried their legs underneath their bodies like mammals (and birds of course) and were decidedly unlike other reptiles. None were fliers, although the most bird-like ones got close, and none were aquatic, although some did indeed at least go wading. But remember, these are generalities and if we find a dinosaur with flippers, as long as it otherwise appears related to other known dinosaurs, it will still be a dinosaur. But if it does not fall within the relationships defined above, it will not be a dinosaur, no matter how much it may look like one.

I don’t have an online source for the Padian article, so here is the reference.

Padian, K. 1997.  Dinosauria: Definition.  pp. 175-179.  In Currie, P.J. and K. Padian (eds.) Encyclopedia of Dinosaurs.  Academic Press.