Not all Tyrannosaurus rex
skulls are equal. Some get pretty banged up in the process of arriving from birth to death, and from death through burial to a fossil, and then even then can get damaged when uncovered and prepared. So its no surprise that when some bones are found, they're not in the best of conditions. This entire process is called Taphonomy, and is a critical science to learning how to interpret fossils. The problem comes from assuming the "found" condition of the bones was the "live" condition. So here is a comparison of three T. rex
snouts, cut through the middle of the snout. Black represents bone and grey represents nonbone. Bones are labeled with three letter designations: NAS = nasal, MAX = maxilla, VOM = vomer, DEN = dentary. Tyrannosaur roots run deep, so they are shown in found positions and the sockets are preserved. There's some black on the teeth themselves that represent how much of the tooth shown is enamel.
As I discuss on my blog, qilong.wordpress.com/2014/03/1…
, there are many things going on in this image.
On the far left is a typical snout, representing the "classic" mount found at the American Museum of Natural History, known generally as 5027. The infamous Sue is in the middle, but the skull was strongly bent, crushed downward and skewed to the side, so what I'm shoing is only partially what was found, and you can see not just a mandible pushed up into the upper jaw, but also the other side of it skewed to the side. Sue's head was found lying on top of one of her mandibles. Finally, we have "Stan," who was a relatively smaller tyrant than Sue, but had a head about the same size. When it was found, Stan's head was very loosely attached, and the parts could be disassembled easily. What was found with this skull, though, were many defects of the internal regions of the snout, but there are also holes in the mandibles where there aren't normally. Stan also has abnormally short jaws, where 5027 and Sue have "normal" length jaws; but more than that, Stan's right jaw was shorter than the left! which certain makes for awkward eating.
Taken together, it is hard to herald the second or third as perfect, but the first shows fewer defects and we can place it as a "icon" of the species. These are only some of the reasons why using just one or two particulars of a broad collection of particulars is problematic, an informal fallacy known as generalizing the specific. Taken in full, tyran skulls are fairly complex, but ultimately prone to deformation that means we have to look at more than one thing.