In The Blink Of An Eye (13 page)

This story illustrates the importance of understanding ancient environmental conditions before reconstructing ancient animals themselves. Suddenly all the fossil evidence of past climates and gaseous conditions is becoming relevant. But it is worthwhile also considering a different type of fossil evidence, one that can have equally important implications for fossil reconstructions. Ammonoids spent their life suspended in the water column. While alive they never set foot on the ground, their ground being the sea floor. Fortunately for palaeontologists, many animals did move on ground, and they left signs of movement in their wakes.

Sherlock Holmes, and indeed his creator, Sir Arthur Conan Doyle, had a keen eye for footprints. Holmes used the size and type of print as an identification tool, the orientation of the prints to deduce entry or exit, and the spacing of the prints to determine the impetuosity of the crime. Palaeontologists, it seems, have converged on this practice.

Dinosaurs left their footprints in mud that became hard-baked and preserved through time. Today the prints are known as trace fossils - not parts of the ancient animals themselves, but impressions made by their movements. Footprints have revealed many secrets of ancient movement, feeding and lifestyles, such as group behaviour. This is all old hat. Now the study of dinosaur footprints has advanced a stage further, following the recent discovery of 200-million-year-old, three-dimensionally preserved tracks in Greenland.

In 1998 an American scientific team set out to explore the tree-barren fields of east Greenland. The team, which included Stephen Gatesy, Kevin Middleton, Farish Jenkins Jr and Neil Shubin, had been lured by the Triassic (over 200-million-year-old) exposures and the prospect of discovering early mammals. But the bones and teeth of various ancient vertebrates were temporarily cast aside as the team's attention became drawn to strange trackways of indistinct footprints.

Figure 2.3
The footprints discovered in Greenland, made in both firm ground and sloppy mud.

There is a law among footprint workers: a trackway is not simply a record of anatomy. Rather it is a record of how a foot behaves under a particular pattern of movement as it makes contact with a particular type of ground. The varying conditions of ground can have a substantial effect on the features of the footprint - contrast a human print left in firm soil with one in wet mud. The Greenland tracks ranged from clear imprints to virtually indistinct traces, but they were made by the same species of theropod (carnivorous) dinosaur in, importantly, different types of ground. The ground varied from firm to sloppy, the range we find on a beach when we walk towards a fluctuating waterline. The prints made in the firm ground were run-of-the-mill, two-dimensional types as known from all corners of the prehistoric globe. As usual, they provided useful information about the owner of the prints and the precise form of the foot. It was the prints in the sloppy mud, however, that led to a breakthrough.

The sloppiness of the mud had preserved a three-dimensional footprint. It preserved the entry and exit ‘wounds' made by the foot. And following comparisons with living animals, it transpired that the deeper you sink, the more of the movement that usually takes place above ground can take place below it instead. This was an important finding. It indicated that the three-dimensionally preserved footprints, regardless of their futile patterns at the surface, could potentially provide data on the movement of dinosaur feet through the air. Quite amazing when you think about it. And the only way of extracting the informative data was to examine the footprints in cross section.

The American team cross-sectioned plaster casts of the fossil prints in abundance. Eventually they assembled complete three-dimensional images of the footprints on their computers. But at this stage the three-dimensional prints appeared just as puzzling as the surface patterns. To make sense of them, the team turned to biology and studied living guineafowl and turkey. Live birds were run through increasingly sloppy mud and it became apparent that they left very similar, three-dimensional footprints. But it was the way they made the prints that was interesting and led to a theory of how dinosaurs moved their feet in the air as well as on ground.

Live guineafowl and turkey placed their feet into the mud with toes
apart. But as they pulled their feet out of the mud their toes were brought together. When the birds walked on hard ground rather than soft, the same series of events took place, although this time they happened in the air. The same was concluded for dinosaurs - they opened their toes as their feet were placed on the ground, and closed them as their feet were lifted. Previously it was believed that some dinosaurs walked on the soles of their feet. But the sloppiness of the sediment revealed that in this theropod dinosaur the heel was carried the lowest, just a bit lower than in birds today. This in turn provided evidence that, compared to birds, the theropod stride was more strongly powered by the femur, while the lower leg and foot provided more of the power thrust.

The entire three-dimensional movement of a theropod foot through mud was modelled on the computer. This involved grafting the anatomy of a typical theropod foot on to the footfall pattern of a live bird. The images, and consequently the surface patterns made by the theropod, were self-explanatory (see Plate 7). It was nice to demonstrate that theropods walked in a similar fashion to birds, because the evidence from two-dimensional footprints and the bones themselves had been hinting at quite major differences in foot skeletons between the two groups. And of course this continued to feed the debate as to whether or not dinosaurs
were
‘birds'. Now it could be demonstrated that locomotion and limb function could have evolved gradually from theropods to birds, in common with many other features.

The precise relationship between dinosaurs and birds is a highly controversial issue. Signs of early feathers on a newly discovered Chinese dinosaur have been rejected by many, who prefer the interpretation that the downy outlines of the fossils are simply fibres from the skin that can fray when reptile skin surface is damaged. Ironically the specimen in question, a 120-million-year-old
Sinosauropteryx
, a theropod, has been brought to virtual life only to deliver a blow to its excavators, who sit within the ‘dinosaurs-are-birds' camp.

The fine silt from an ancient lake had preserved the soft structures of
Sinosauropteryx
, including a clear silhouette of the lungs. John Ruben, a respiratory expert from Oregon State University, took one look at the ‘lungs' and knew what he was dealing with. He had seen this lung arrangement before - in crocodiles. Immediately he constructed his virtual, living dinosaur, with the same compartmentalisation of lungs, liver and intestines that one would find in a crocodile, and not in a bird. This virtual dinosaur was incapable of the high rates of gas exchange needed for warm-bloodedness. So it contained cold blood, like the crocodile. Also, its bellows-like lungs could not have conceivably evolved into the high-performance lungs of modern birds. But still this evidence, that birds were not descendants of dinosaurs, is far from conclusive. As new fossils are unearthed and analysed with the lives of modern animals in mind, the building of a virtual dinosaur continues.

A study of vocal cavities and the surrounding bones has revealed the range of sounds once made by dinosaurs, from the high-timbred, lion-like roar of
T. rex
to the bellowing
Diplodocus
, with a voice reminiscent of air being forced from a hydraulic piston the size of a drainpipe. The nostrils of
T. rex
have been shifted further forward in its head to take a new position just above its mouth. Now
T. rex
has a much larger area of nasal tissue, fully laden with the capacity for a considerable sense of smell. This puts virtual prey in increasing danger, although as palaeontology becomes increasingly refined, maybe they too will become adapted, in this case to control their scents.

We identify the food of dinosaurs via the dentition of their jaws, the often fateful teeth marks left behind in bones, and their dung. But dinosaur dung has provided further information on ancient lifestyles and evolution - that of dung beetles. Radiating clusters of burrows have been found in Cretaceous dung that precisely match those made by dung beetles in elephant excrement today. These burrows indicate that dung beetles evolved with herbivorous dinosaurs, rather than with later occurring grassland mammals as previously thought. And here we have returned to the subject of trace fossils, which have breathed so much life into our models of extinct forms, right back to the Precambrian.

So dinosaurs are now running, breathing, smelling, roaring and
excreting on our computer screens. The famous
T. rex
, whose skeleton was once constructed upright with tail on the floor, in the style of Godzilla, now lives its virtual life in horizontal stature - perfectly balanced with legs acting as a fulcrum. Similarly,
Diplodocus
no longer scrapes its belly on the floor. And if the makers of those first dinosaur reconstructions had taken note of the trace fossils, or consulted Sherlock Holmes, they would have noticed bold footprints but not a trace of lagging tails or hauling bellies in sight. Importantly and necessarily, dinosaur studies have led palaeontology well into the computer age.

More recently, the idea of producing three-dimensional models has been applied to fossils themselves. Travelling back some 400 million years, to pre-dinosaur times, certain marine organisms living in the shallow waters of the Earth were also preserved exceptionally well. Algae from these waters can now be found in New York State, some which have been replaced with pyrite but others which have been chemically unaltered and still contain their original organic material, like the flies mummified in amber. But more mysterious life forms of the era have been found. The exceptional preservation of these invertebrates has given rise to an unusual property of their fossils - they are three-dimensional.

The British team that recently discovered and began work on these fossils comprised David and Derek Siveter and Derek Briggs. The discovery itself was perhaps lacking in the romance of some better known examples. I pictured this research team flying within the Grand Canyon in a 1920s biplane, but my dream was shattered when I asked David Siveter about the locality of the fossils. He pointed to a large mound of earth visible from his office window even on a grey, rainy day. However, the ingenuity and excitement of this project lay with its methods.

During one decisive meeting, the research team examined the diversity of their fossils and realised that classification would be
problematic. A view of only one surface or plane of a fossil, a view that fossils typically present, provided inconclusive evidence in this case, even at high magnifications. The three-dimensional preservation resulted in a limited view of the fossil, whose exposed parts lay flush with the rock. Imagine a golf ball embedded in a sand bunker with just one dimple exposed. The team knew there was more to these fossils than first meets the eye. To extract the maximum information, an unusual preservation called for unusual methods. In fact they chose to pioneer a new method for fossils. That method was risky - in the process of examination the valuable fossils would be completely destroyed. The gamble, however, paid off.

Today engineers employ computer-aided-design, or CAD, to construct and view car designs in three dimensions. Compared to pen and paper, CAD provides the advantage of enabling an object to be viewed in three dimensions and from all angles, as the object can be rotated on the computer monitor around any axis. The palaeontological team on this case wondered about the possibilities of introducing CAD to their analyses, and they soon enrolled a postdoctoral worker, Mark Sutton, with computer programming talents. But a hurdle lay ahead - the tiny fossils, perhaps only a couple of millimetres wide, required separation from the rock. Basically this was not possible for such a preservation type. So how could they determine the structure of all sides of the fossil - the food for the CAD-style program? This was the risky part - the fossils were to be ground away, a hair's width by a hair's width.

After each serial grind, a photograph was taken of the newly exposed section of the fossil. The palaeontologists were interested only in the surfaces of the fossils, since the innards had not been preserved. Although each grind revealed a redundant cross section, the photographs were fed, in order, into the computer, and the computer did the rest. The results were staggeringly good. In this chapter I have attempted to describe how fossils can be brought to life by piecing together one small fragment of evidence after another. Bit by bit fossils can grow virtual outer skins, fill with virtual blood, and walk across the computer monitor in search of specific virtual food. But in this case, what had involved years of work for other fossils happened in an instant. A complete animal, more than 400 million years old, came to
virtual life on the computer monitor with one press of a button. The worm-like, armoured forms of early molluscs and segmented worms, some the earliest known representatives of their kind along with ancient arthropods, appeared exactly as they would have when they originally roamed the reefs. There were no fragments of anything, just the entire animals. And the 3D images could be rotated on the computer monitor revealing views from above and below, from the front and the back . . . from any angle one desired. Amazing! It is to be hoped that this CAD-style methodology will enjoy a happy future in palaeontology.