What, if anything, is a reptile?

In a comment on a recent post to my blog, I remarked that "there's no such thing as a reptile any more."  Steven J. Gould once wrote an essay entitled "What, if anything, is a zebra?" in which he considered similar claims about zebras.  The issue is one of taxonomy.

Modern taxonomists try to put organisms into what are called monophyletic groups.  A monophyletic group is one that contains all descendants of a single common ancestor, and nothing else.  Mammals, for instance, are considered to be monophyletic.   All modern mammals are thought to be descended from a single ancestor, probably Morganucodon or something closely related to it; from this ancestor they inherit some unique traits, such as three ear bones.  Those bones, which develop in mammalian embryos from tissue attached to the lower jaw, are found only in mammals (as ear bones; in other tetrapods they form part of the lower jaw in adults), and in no other organisms.  Such a peculiar design is so clearly arbitrary -- obviously sounds could be conducted via any number of bones, and there's no design justification at all for making them out of jaw tissue to start with -- that the only reasonable conclusion is that it must have been inherited from a common ancestor.  Separately evolving groups would hardly have arrived at this particular odd design independently. 

If you look up a mammal in the National Center for Biotechnology Information (NCBI) Taxonomy Browser (enter the name in the box; if you use a common name, select "token set" instead of "complete name"), you'll get something like this:

Bos taurus

Taxonomy ID: 9913
Genbank common name: cattle
Rank: species
Genetic code: Translation table 1 (Standard)
Mitochondrial genetic code: Translation table 2 (Vertebrate Mitochondrial)
Other names:
 
synonym: Bos primigenius taurus
synonym: Bos bovis
common name: cow
common name: bovine
common name: domestic cattle
common name: domestic cow
Lineage( full )
cellular organisms; Eukaryota; Fungi/Metazoa group; Metazoa; Eumetazoa; Bilateria; Coelomata; Deuterostomia; Chordata; Craniata; Vertebrata; Gnathostomata; Teleostomi; Euteleostomi; Sarcopterygii; Tetrapoda; Amniota; Mammalia; Theria; Eutheria; Laurasiatheria; Cetartiodactyla; Ruminantia; Pecora; Bovidae; Bovinae; Bos

In the lineage above, you can hover the cursor over any of the names, and a box will pop up telling you the rank.  So if you put the cursor on "Bos", you get a box that says "genus", for instance.  As you move up the hierarchy, you'll find that the name "Mammalia" refers to a class.  That's all the organisms with three ear bones and a few other uniquely inherited features. 

So what's up with the reptiles?  Try the same thing with a reptile, and you get a different result.  Take the box turtle, Terrapene carolina.   Here's its lineage:

Lineage( full )
cellular organisms; Eukaryota; Fungi/Metazoa group; Metazoa; Eumetazoa; Bilateria; Coelomata; Deuterostomia; Chordata; Craniata; Vertebrata; Gnathostomata; Teleostomi; Euteleostomi; Sarcopterygii; Tetrapoda; Amniota; Sauropsida; Testudines; Cryptodira; Testudinoidea; Emydidae; Terrapene

What you'll find is that, although this turtle is in the order Testudines and the superclass Gnathostomata, it has no class. 

Why not?  If you look around, you'll find plenty of people who will tell you that turtles are in the class Reptilia, because they're reptiles.  Along with what?  Snakes, lizards, and crocodilians.   And what's the problem with that?  Well, for reptiles to be a monophyletic group, they would have to include all descendants of a single common ancestor.  Certainly there was an ancestor shared by snakes, turtles, lizards, and crocs; but that ancestor was also the ancestor of birds.  In fact, it appears that crocodilians are more closely related to birds than they are to turtles, for instance.   And that means that if you want reptiles to be a monophyletic group, you have to count birds as reptiles.  Since birds are already a class (Aves), you can't have one class inside another -- and that means there is no class that includes turtles, snakes, lizards, and crocs.  In fact, it's misleading to think of these as a "group" at all, evolutionarily -- it suggests they're more closely related to each other than they are to other organisms outside that group, which isn't true.    We say that "reptiles" are a paraphyletic group -- the group includes some, but not all, descendants of a common ancestor.   Here's a nice image from Wikipedia showing these distinctions:

What's to be done?  1) Abolish "Aves" and put birds in the class reptilia along with the "reptiles".  2) Abolish "Reptilia" and come up with some (probably 3) new classes for turtles (Testudines), lizards/snakes (Lepidosauria), and crocs.  3)  Stop talking about "Reptilia" and just muddle along until somebody comes up with a new idea.