We also can’t know what the eyes themselves looked like, although they are major contributors to the appearance of a dinosaur face presented to the public. The slit pupils of reptiles and the round pupils of birds are good guesses, but artists have tended to go with slits only when they want the dinosaur to look mean. Brougham’s protoceratops has nondescript beads with round pupils.
The rest of the face of Brougham’s protoceratops is the same shape as the fossil skull. There are no extra flourishes, folds, or fringes except around the mouth, where he pulled the flesh over the sides of the jaw to make cowlike cheeks. Cheeks “look right,” he says, but there’s no conclusive evidence for them. The Disney dinosaurs also have cheeks, as well as fringes that look like eyebrows and lips for speaking English. Brougham put the nostrils on his protoceratops far forward, near its beak, following the example of most living creatures. Think of alligators, lizards, dogs, and donkeys: The nostrils go as close to the mouth as the nasal cavity will allow. Vertebrates like to lead with their noses. Disney’s dinosaurs follow a different example, quite popular in dinosaur art. Their nostrils are farther back on the snout, halfway between the eyes and the mouth.
The velociraptor that stalks the protoceratops in the diorama is also based on a series of educated guesses about skeletal and soft-tissue structure, with one added consideration: This velociraptor has feathers. A multicolored frill graces its pate, and the rest of its body is covered in a fluffy sheath of unusual feathers that dinosaur artists have come to call dinofuzz—definitely not fur, but not quite feathers either. The feathers were Norell’s idea. They’re not required by direct fossil evidence, but there is a debatable set of quill-like impressions around some fossil finds. To Norell, it’s important to read these as feathers all over the body because the major thrust of his work as a scientist is to show that dinosaurs are direct ancestors of modern birds. The small fossils that line the walls of the Fighting Dinosaurs exhibit have helped him further establish an evolutionary lineage from Tyrannosaurus rex and Velociraptor to the emu, the tit, and the vulture. But he cannot prove these relationships. Very little in paleontology is subject to direct proof because we cannot resurrect these creatures. Nevertheless, it is a legitimate working hypothesis, and ensconcing a hypothesis in artistic reconstruction is common practice in scientist-artist collaborations.
We often think of bones as a stiff scaffold that everything
else hangs on, but anatomists know that soft tissues
morph bones as an animal grows
So. The bones are evidence that’s interpretable. The muscles are guided guesswork. Everything else—soft tissues, skin, folds, frills, cheeks, lips, eyes, nostrils, pattern, color, covering—is artistic license, or, in the case of Norell’s feathers, scientific license. How far off could the dinosaur renderings be?
The most conservative answer, which many scientists give, is that we just don’t know. Paleontology is the study of fossils; fossils are limited in the information they provide, so we will always run the risk of drawing a titanic hamster instead of an elephant, and we’ll never know if we’re doing it. Artists give a curiously different answer. In a sense, they know better than the scientists how far off they might be because they are the ones who must constantly do the guesswork. Many artists agree we’ll never know whether a particular drawing accurately represents a creature that’s been dead for 80 million years, but they also say it is possible to get a sense of how far off dinosaur art could be by exposing the process of layered guesswork. Hallett puts it this way: If he took a set of scientific evidence about a dinosaur and allowed himself to explore all the possible muscle placements, soft-tissue structures, facial variations and skin tones and patterns and folds and frills, he says he could draw a series of solutions and “come up with very different-looking dinosaurs.” To scientists, of course, they’d all be the same, based on the same evidence. But to museum goers, they’d be different creatures. The Society for Vertebrate Paleontology’s Scott Sampson thinks such a varied lineup would be a great way for the science to peer through the art, because the only constant across the range of pictures would be hard facts: “Don’t just illustrate [a few options] with a given set of evidence, but do the whole range. Vary everything and get 17 different things. It underscores our ignorance about these animals.”
Museums and moviemakers seem not likely to offer that approach soon. You can’t have a movie dinosaur looking entirely different from one scene to the next. The art is supposed to convince the viewer of the image, not of the science behind it. “I want to suspend the disbelief as much as I can,” says Michael Skrepnick, a Canadian paleo-artist who has done dinosaur illustrations for National Geographic and various museums. “I want you to believe in the animals you’re looking at. I want you to think they’re real.” It is also uncommon for differing interpretations to crop up among illustrators. Because illustrators know how much hard work goes into reconstructing a dinosaur for the first time, they respect each other’s original interpretations. Once a species is painted or sculpted, later illustrators generally honor the speculations of the original artist.
There is yet another take on dinosaur art, one that comes from a small but growing circle of young paleontologists who say that many of the decisions currently guessed at by artists might someday be based on real evidence from scientists. Larry Witmer, a professor of anatomy at Ohio University’s College of Osteopathic Medicine, maintains that scientists can increase the yield of data from the bones they’ve got. All they have to do is look closer. His idea sounds simple enough: Look hard at the bones of modern animals to study the tiny marks that soft tissues make on bones, and see if such subtle marks can be found on dinosaur fossils as well. This growing field of soft-tissue research is a radical departure for paleontologists, most of whom were trained in geology (fossils are rocks found in rocks). Their soft-tissue analysis of bones has been largely limited to the muscle scars that can be seen with the naked eye. “People thought that soft-tissue stuff was just truly out of reach—that there was no rigorous way to know, so they just let the illustrators do it,” Witmer says.
Larry Witmer (above) makes scientific inferences about the soft tissues of dinosaurs based on a comparative analysis of microstructures in dinosaur fossils and the bones of modern animals, including the white rhino, the manatee (the skull in the scale), and the hooded seal (CAT-scan images). Witmer also studies crocodiles and birds, which are the closest living relatives of Velociraptor (next to tools) and Deinonychus (in his hands). |
We often think of bones as a stiff scaffold that everything else hangs on, but anatomists know that soft tissues actually morph bones as an animal grows, and over the course of evolution. “Soft tissues evolve, bones respond” is Witmer’s working slogan. In the case of tapirs, Witmer found that their skulls have various characteristics such as nerve passages, blood ducts, and small muscle attachments that are caused by the snoutlike structure, and therefore necessary for the snout’s existence. When he searched for these characteristics on diplodocus skulls, he found none.
Witmer’s inference that Diplodocus did not have a tapirlike snout now has some scientific basis. He did a similar study of the mooselike jaw muscles that some illustrators put on duck-billed dinosaurs and found that duck-bills do not have the hefty bone structure moose need to support such prominent jowls. Witmer understands why the artists put them on in the first place. The big mammalian muscle “looks great in reconstruction.”
Most renderings of duck-billed dinosaurs have that same familiar look of Iguanodons—they are great lizards reminiscent of horses. “If you make a duck-billed dinosaur look like a horse, which so many illustrators do, well, it looks right. But we have reconstructed it in modern guise,” Witmer says. We live in an era of mammals, so dinosaur art is heavy with mammal flesh that is simply out of place. Dinosaurs are more closely related to snakes than they are to mammals, and the moose and the horse are no better guides to what dinosaurs looked like than manatees, humans, or voles.
Commenting on the series of choices illustrators have to make, from muscle reconstruction to soft tissues and facial features, Witmer suggests that any time an artist makes a decision because it “looks right,” he risks falling into the mammal trap. Artists could do a better job by stealing soft tissues from crocodiles and birds to flesh out their dinosaurs, he says, and science-minded artists like Hallett, Skrepnick, and Brougham have already begun to do so. Witmer does not expect the approach to satisfy our expectations. “If you go back and do it right, then it’ll look totally weird. But it’ll be right,” he says. “We’ve missed a lot of their bizarreness by using these false analogues.”
Witmer continues to dissect modern animals, looking for tiny, signature bone structures that they might share with fossil dinosaurs. He just obtained “a wonderful huge head of a rhino,” which required a bigger freezer in his lab. Many paleontologists have maintained that the horns of ceratopsians were rhinolike, and Witmer can now test that hypothesis.
He’s always on the lookout for crocodiles and birds, too, because they hold the best clues to dinosaur soft tissues. The ideal find for him is a set of soft-tissue marks on the bones of both crocodiles and birds. If he can find those same marks on dinosaur fossils, then he has very strong evidence that the dinosaur had the same soft tissue. He has already used this technique to establish that dinosaurs had an air sack housed in a skull cavity between the eye socket and the nasal opening. For more than a century, paleontologists kicked around hypotheses as to its function: It was a muscle, it was a gland for salt. Crocodiles and birds have air sacks in that cavity, and those air sacks leave the same characteristic microscopic marks Witmer found on fossil skulls.
This subcutaneous soft-tissue discovery doesn’t dramatically change the outside look of dinosaur art. And Witmer’s goal isn’t to improve dinosaur art. His purpose is to learn more about questions like: What organs did dinosaurs have and what were they used for? But Witmer’s work, and the work of a handful of other young paleontologists who are approaching the soft-tissue questions with scientific methods, may ultimately allow artists to flesh out dinosaurs with more hard data. “Anything that increases the chances of a paleo-artist [rendering] something more accurate would be welcomed with open arms,” says Hallett. He says he and his colleagues aren’t in this business to make guesses. They want to know the truth.
And what about all the paleontologists who say anyone trying to figure out what dinosaurs really looked like is asking the wrong question? Deep down inside, suspects Hallett, they want to know too.
