Killing a Legend (cont'd)

A key link in Horner's scavenger hypothesis came in 1990, when amateur fossil hunter Kathy Wankel brought him several curious pieces of dinosaur bone. They turned out be parts of a T. rex arm, the first ever found for the species. Horner excavated the "Wankel T. rex," which turned out to be 90 percent complete — with both arms.

The arms turned out to be not just puny, but virtually useless. T. rex's upper arm, the humerus, is quite a bit longer than the forearm bones, the ulna and radius. This wasn't an arm designed for lifting things; in humans, both the upper and lower arm are roughly the same length.

The T. rex arm itself is very short in relation to the dinosaur's size, and much of that short length is buried in the animal's chest. What's left is a pair of spindly appendages that can't lift very much, can't reach the animal's mouth and can't even reach each other. These hands, Horner says, aren't good for much of anything but "scratching his belly."

The T. rex also reveals another feature that would create problems for a predatory creature: Its thighbone is longer than its shinbone. "If you look at any bipedal animal that runs fast, like an ostrich, you find a very short thighbone and a very long shinbone."

Consider the real predators of the Late Cretaceous, smaller and truly vicious bipedal dinosaurs such as Velociraptor, Deinonychus and Dromaeosaurus. Fossil skeletons (6 to 10 feet [2 to 3 meters] long) clearly depict agile creatures built for speed, with short, powerful thighs and long shins. Their arms were relatively long and strong, with vicious claws. One toe of each foot was armed with a long, sickle-like claw that could eviscerate its victims with a single swipe. Beneath large eyes was a mouth filled with sharp teeth serrated like steak knives. These were killing machines, and they probably hunted in packs like modern wolves.

Compare these consummate predators with Tyrannosaurus rex. "They are," Horner says, "just built completely different. When you add up all the features of a Velociraptor, you come to the conclusion that it had to be a predator. If you add up all the features of T. rex, it has to be a scavenger."

The trick, Horner says, is looking at this most popular of dinosaurs scientifically, to wipe away previous perceptions as though you had not been told since childhood that it was a terrifying killer. Then only the evidence can speak to you.

T. rex, he believes, came second to the kill, after the predators' sharp teeth had stripped away the prime flesh. An odious tyrannosaur — big, ugly and nasty — might then scare off the killers and hold them at bay long enough to use those enormous jaws and heavy teeth to chomp great chunks of bone, gristle and flesh from the huge carcass of, say, one of the very common Triceratops. Those are teeth that would have no trouble crushing big bones that the raptors could only gnaw.

Hints, if not proof, of that scenario come from a striking discovery of the Hell Creek Project. A sacrum (the bone at the bottom of the spine) from a Triceratops bears several deep, puncture marks. The holes are a good fit with T. rex teeth. It's very likely the bone represents scraps of a T. rex meal, although it offers precious little data about who killed the plant-eater. Nonetheless, the holes are on a part of the bone that would be inaccessible until the animal was torn apart.