Dinosaurs Without Bones

by Anthony J. Martin

  To further back up their claim, they pointed toward a possible perpetrator, Muttaburrasaurus langdoni. Skeletal remains of this ornithopod were discovered in 1963 near the small town of Muttaburra in central Queensland, northeast of Lark Quarry, and like many dinosaur finds in this part of Australia a rancher, Doug Langdon, spotted its bones. Still, it wasn’t named officially until 1981, when the paleontologists took care to honor Mr. Langdon’s discovery by naming the species after him. Based on its body parts, Muttaburrasaurus was a big dinosaur. Its femur alone was about a meter (3.3 ft) long, which meant its hip height would have been close to the calculated value of 2.5 meters for the Lark Quarry trackmaker. Hence, it seemed a perfectly reasonable dinosaur to pick as a possible large three-toed trackmaker for Early to mid-Cretaceous rocks in this part of Australia.

  In contrast, bones of only one large predatory theropod have been found in the same area and rocks of similar age: Australovenator wintonensis, lovingly nicknamed “Banjo” after Australian poet and writer of “Waltzing Matilda,” Andrew Barton “Banjo” Paterson. Despite the unabashed wickedness of this dinosaur—gorgeous big-clawed hands attached to long grasping arms, a mouth full of nasty pointed teeth, and a lightweight frame which suggested that it was a speedy predator—this theropod seemed a bit too small to match the trackmaker at Lark Quarry: more Allosaurus and less Tyrannosaurus. Otherwise, body parts of tyrannosaur-sized theropods are unknown in Queensland, but also in much of Australia with the exception of a few isolated bones in Victoria. Nearly all that is known about big Cretaceous theropods of this entire continent comes from their tracks, which Thulborn and others described from Western Australia in the 1990s, and a few that I and others found in Victoria in 2007–2008.

  When Australian paleontologists Tony Thulborn and Mary Wade proposed the “dinosaur stampede provoked by a large stalking theropod” idea in 1979, this was nine years before the publication of Moratalla and his colleagues’ study and thirty years before the world learned about Australovenator. So Thulborn and Wade can’t be faulted for what they didn’t yet know, and they were using the best science known then to interpret the tracks. Also, like I said before (and it bears repeating), they did fantastic work. I consider their 1979 paper a classic in dinosaur ichnology, which they also followed up in 1984 with a considerably more detailed 105-page report.

  Nonetheless, it’s certainly possible they made a mistake in identifying the maker of the large three-toed tracks. At the time, they tentatively identified these on the basis of their resemblance to large theropod tracks known elsewhere, which had been allied with tyrannosaur-like tracemakers. Interestingly, objections to this identification, voiced formally by Romilio and Salisbury, were not new, as a few dinosaur paleontologists questioned it soon after Thulborn and Wade’s second article came out in 1984. Romilio and Salisbury were just the first to rigorously test the original hypothesis using statistical methods as opposed to just the scientific equivalent of name-calling.

  Backtracking the Walking Dead

  So the hypothesis is dead; long live the new hypothesis! This means the popular story of Lark Quarry needs to be revised, and the tour guides there must alter their spiels or at least present “both sides” of the argument. Furthermore, the Gallimimus-panicked-by-Tyrannosaurus scene in the movie Jurassic Park needs to be redone with an alternate version, showing the Gallimimus running for some other, unspecified reason. And, without a doubt, a stage play is absolutely necessary: to do anything less would be irresponsible. (I say the last facetiously, but a small troupe of creative folks actually did write and perform a short musical number retelling the original story of the dinosaur stampede, shown at the 2012 Museum Australia National Association meeting in Adelaide. Could a Broadway production be far behind, but now with the anticlimactic ending inserted?)

  Not so fast. There’s one little problem with such simple and definitive declarations. What if Romilio and Salisbury were also wrong? What if they had neglected a few clues in their study, or carried out the research in a less-than-careful way? For instance, we’ve seemingly forgotten all about one of the cast of characters in this drama, which was the medium-sized ornithopod that first left tracks at the site, before the arrival of the terrorized flock and the big … well, whatever. Were its tracks studied in the same way, to test whether it was also an ornithopod or not? Also, how about revisiting Australovenator as a possible trackmaker?

  First of all, the tracks made by the medium-sized ornithopod at Lark Quarry present what scientists might call a “control” at this site. In other words, these could be compared to the larger three-toed tracks to see whether or not they match in form. If they did match, this would have strengthened Romilio and Salisbury’s assertion that the statistical technique they used could be applied to three-toed tracks of differing sizes, and that they could distinguish more than just the one trackmaker at the tracksite. Alas, they did not do this, or perhaps they did but it was not mentioned in their published paper.

  How about Australovenator as the new antagonist (or protagonist, depending on perspective) for the Lark Quarry tracks? This isn’t a completely crazy question to ask, because “Banjo” is currently the largest known theropod in that part of the world, and its bones come from rocks of about the same age as the tracksite. However, its feet, which are known from well-preserved foot bones, seem too small to fit the 60+ cm tracks, a sort of Cretaceous Cinderella story in which the slipper (footprint) does not fit.

  Unfortunately, a documentary film trumpeted the conclusion that Australovenator’s feet did fit the tracks, and before yet another scientific investigation had been completed. Imaginatively titled Dinosaur Stampede, the ABC (Australian Broadcasting Corporation) broadcast this made-for-TV film in 2011, adding further confusion and controversy to public perception of a fossil site that didn’t need more of either. This film, which used a combination of slick computer-generated graphics, dramatic music, lots of footage of paleontologists reexamining the tracksite with new technologies (lasers and computers and 3-D models—oh my!), and a fully fleshed-out recreation of an Australovenator leg and foot, culminated in one main conclusion: Yes, “Banjo” could have indeed been the culprit for the large tracks. Now there is a third possible solution to the mystery of Lark Quarry, and one that honors the original story but with a slightly modernized and updated perspective.

  But here’s the problem. Before allowing preliminary results out into the public realm, we scientists are supposed to do peer review. This means not just doing the initial research, but also scrutinizing the results, summarizing these in a formal article, submitting the article to a journal, having that article scrutinized by experts in the topic of your article, and in the end possibly being told, um, no, you got this all wrong, do it over. Is peer review a perfect process? No, and if for no other reason, because the scientists who review journal articles are doing it as unpaid volunteers, finding time to perform this important duty in between all of their other tasks such as teaching, grading, research, walking the dog, or (most heinous of all) sleeping. Still, more often than not, peer review provides an outside perspective that ultimately results in scientists honing their argument or even rejecting it outright. In the latter case, the research may never make it into print. All of that work for naught, so to speak.

  Because of this possibility of being wrong, paleontologists and other professional scientists are generally discouraged from announcing their preliminary results to the public, and especially through popular media. An exception would be if these scientists deliver a talk or present a poster at a professional meeting in front of their peers, where an accelerated and more direct in-your-face form of peer review can take place. But to have a documentary film made about your research before the actual paper is published? This cheeky behavior often results in raised eyebrows and much harrumphing, mostly prompted by honest skepticism but also perhaps a wee bit of jealousy.

  So as far as most paleontologists are concerned, a documentary, no matter how entertaining and informati
ve, is never on equal footing with a peer-reviewed paper. Now, what if the research results were reported in a reality-TV series in which all of the competing paleontologists are forced to live together in the same house and argue over their respective results? I’d watch that for sure, but it still wouldn’t count as real science either. In the end, we have to write it down, have it reviewed, and go through the formal process.

  However, once an article is accepted by a journal and published, this triumph does not imply that its methods and conclusions are inviolable. All published articles are subject to further peer review long afterwards, or at least they should be. Indeed, this is exactly what happened to Romilio and Salisbury. Just like a boomerang, a major challenge to the revised hypothesis came from an unexpected source, which was Dr. Thulborn himself. Sadly, Thulborn’s colleague on the original two milestone studies, Dr. Mary Wade, passed away in 2005, meaning her original impressions and insights on the tracksite died with her. Thus it was up to Thulborn alone to respond to the critique of their work.

  Respond he did, and with ferocious intensity. Most devastating of all, his reaction to the restudy of the tracksite was in print and published as a peer-reviewed journal article and not aired as a scientific grievance through the more instant media of news stories or blogs. In this manuscript, which was posted online in the same journal (Cretaceous Research) that published the re-study, Thulborn pointed out several significant flaws in the rebooted version, effectively sowing doubt on its new conclusions. A minor tussle resulted, magnified by online discussions, thus provoking much head-shaking and sighing amongst other paleontologists and dinosaur-track enthusiasts who wondered how this would end.

  One potential source of error related by Thulborn was that the 1988 study on dinosaur tracks done by the Spanish paleontologists was performed on a mixture of only sixty-six ornithopod and theropod tracks, and only from one formation in Spain. In other words, it was likely never intended as a universally applicable, one-size-fits-all method that could be used on dinosaur tracks of varying geologic ages anywhere else in the world. A second possible flaw is that the tracksite, once uncovered in the 1970s, had been outdoors for several decades before having the protection of a building around it. (A roof covered the tracksite for a while, but this shady spot in the middle of the desert attracted kangaroos, which found it a most excellent spot to sleep and defecate. Trust me, this is not good for dinosaur tracks.) Their exposure meant that the tracks were weathered and hence altered since the original measurements were taken on them, decreasing the accuracy of any new numbers derived from them.

  A third problem was that some of the “new” measurements must have been taken from Thulborn’s original line drawings rather than directly from the tracks. This sort of extrapolation again results in diminished accuracy. A fourth problem—and in my mind, this is the biggest one—is that the foot structure of Muttaburrasaurus, the alternative trackmaker, is based on just five bones: three metatarsals (heel bones) and two phalanges (toe bones). Yes, that’s right: We don’t know the actual size of its foot or other details of its anatomy. It would be like saying a suspect in a crime was identified by his tracks, but the suspect only has two toes on one foot and lost the other foot in an unfortunate accident. A fifth (but not final) problem was an oversight, which was that a set of perfectly fine but smaller ornithopod tracks was only a few meters away at the same tracksite (remember the first actor from the stage play?). Yet these were neither studied nor compared to the large tracks to see whether they matched in their dimensions. In short, the re-study had been re-studied, but this iteration put everything back to where we started, with the original story of Lark Quarry mostly intact.

  One problem, though, was that Thulborn’s critique was acerbically worded and pointedly critical of the new results, including the inflammatory word “fabricated,” implying that some of the data may have been invented. Consequently, Romilio and Salisbury complained vehemently to the journal, and as a result an editor or employee of the publishing company that owned the journal decided the article was too hot to keep in print. Hence, it did what very few academic journals do, which was pull the article offline.

  A few people (including me) were lucky enough to have downloaded it before it vanished from the Web, which felt vaguely like owning a bootleg Bob Dylan album. (Although, based on the tone of the article, Iron Maiden might be a better analogy.) Suddenly, the former existence of the article or people possessing a copy of it did not matter, as no one can cite it in their own research. Its abrupt absence from the journal was thus the academic equivalent of “disappearing” Thulborn’s work and all it portended.

  Nevertheless, this was not the end of this scientific squabble. Thulborn submitted a revised version of his article to the Australian paleontology journal Alcheringa. It went through peer review again and was published in early 2013. (Full disclosure: I was one of the reviewers of this paper.) So once this critique was back in the public realm and available for the paleontological community to assess, the controversy over Lark Quarry and its mysterious dinosaur tracks resumed again, and probably not for the last time.

  Then, also early in 2013, the aquamusical version was unveiled. In another article by Romilio and Salisbury, joined by paleontologist Ryan Tucker, they proposed that the “dinosaur stampede” was actually made by small dinosaurs—either theropods or ornithopods—swimming downcurrent in a river, not a lakeshore, and over days, not in a moment of panic. In this paper, they explained that long shallow digit impressions left by the feet and the long distances between each impression were a result of buoyant (but not necessarily ebullient) dinosaurs. This was the most radical of all reinterpretations for Lark Quarry, serving up a one–two punch to the public consciousness of this world-famous tracksite: no giant carnivore, and no stampede.

  What do I personally think happened? Based on the previous evidence, I used to have little doubt that a dinosaur stampede took place, and that it happened after the arrival of the big three-toed dinosaur. But now that I’ve also seen hundreds of dinosaur swim tracks from the Early Jurassic of Utah, I also wonder about this as a possibility too. As to whether the large dinosaur was a large ornithopod or theropod, I’ll continue to reserve judgment on that. A “submerged” scenario complicates any detailed analysis of the tracks, because tracks made underwater often become vaguer or distorted compared to those made out of the water.

  I’ll further state, somewhat impudently, that in the “stampede” hypothesis, the identity of the big trackmaker doesn’t matter. After all, a lumbering ornithopod also could have provoked panic in smaller dinosaurs, which may have been just sensibly avoiding being in the same spot as an ornithopod foot. Or a stampede could have been prompted from some other cause that happened a day after a large theropod or ornithopod walked through the area. The direction of movement also might have been an artifact of landscape, in which a nearby lake shoreline kept all of the dinosaurs moving in one of two directions. Thus I am left with more questions than answers, a common problem in science but one particularly exemplified by the dinosaur tracks of Lark Quarry.

  Regardless, this whole affair also serves as a reminder that almost no one gets the final word in such debates, and that some problems surrounding dinosaur trace fossils defy solving while still tantalizingly beckoning us to solve them. This is surely very good for paleontology, for long after I have written these words we will have learned much more about the dinosaur tracks of Lark Quarry and other known and still-undiscovered dinosaur tracksites.

  Sleeping with the Dinosaur Tracks

  Just how did I get trapped in the building housing these fabled and controversial dinosaur tracks? In July 2007, I was in Queensland teaching a study-abroad program through my university that was hosted by James Cook University in Townsville. One day, after looking at a map and realizing that Lark Quarry was a mere seven-hour drive from my location, I decided to make a pilgrimage there. So once I was done with teaching my part of the program, I rented a car and drove west from Town
sville to Winton. Once in Winton, I took a left, and drove another hour to Lark Quarry, most of it on a dirt road, braking and swerving madly whenever kangaroos bounded across the road. (Please don’t tell the car-rental agency.)

  Having successfully made it to the last afternoon tour of the day, many other visitors and I were met by a tour guide in the main lobby of the building. After a brief introduction, she unlocked a door and led us inside a vestibule. With one turn around a corner, we were in a very large room and on an elevated walkway, looking down on thousands of dinosaur tracks. Once there, our guide told the familiar, traditional yarn of Lark Quarry to a large and eager crowd of Australians ranging from children to grandparents. It was warmly reassuring for me to see so many people there, all admiring these trace fossils and marveling at the gripping narrative they inspired.

  During this time, I kept quiet, a stealth paleontologist not wanting to affect the tour or otherwise call attention to myself. Once the tour ended, I purposefully dawdled, waiting for other people in the tour to leave so I could soak in the beauty of these traces alone and in silence. I took many photographs and otherwise gazed longingly at the tracks, beheld at long last after reading about them for more than twenty years. It was a lovely moment, one that needed savoring.

  Nonetheless, while in this ichnologically induced nirvanic state of mind, I neglected to hear the clicking of the door—locked by the tour guide who had not bothered to check whether anyone was still inside with the tracks. The immoveable door invoked a panic akin to what might have been experienced by the small dinosaurs there about 98 million years ago. After all, I was about 110 km (60 miles) from the nearest town of Winton, without food, water, blanket, toilet, or other amenities, and would not be rescued until mid-morning the next day, and in front of an audience of about a hundred people. On the bright side of things, I was not locked in overnight with a voracious predator, although I was concerned about a fitful sleep filled with bad dreams taking me back in time to that Cretaceous lakeshore. It looked to be a long night.