Robert Coram (UK)
The Mesozoic and Cenozoic deposits of Southern England have long been a rich source of fossil reptiles. Past finds of great historical importance include some of the earliest known examples of dinosaurs, ichthyosaurs and pterosaurs. Fossil material, including new species, continues to be revealed, mainly at rapidly eroding coastal sites. All these reptiles would have been active participants in their local ecosystems, whether on land or in the sea. Much information about the roles they played and their interactions with other organisms can be gleaned from their skeletal anatomy and from comparison with living relatives such as crocodiles.
What this article is concerned with, however, is evidence of specific incidents in the lives, and deaths, of individual reptiles; tiny snapshots of opportunities, mishaps and the daily drudge of staying alive. These add more detail and colour to our knowledge of the lifestyles of these long-vanished animals. This evidence will be provided by four selected terrestrial and marine deposits from southern England, spanning the last quarter of a billion years of Earth history (Fig. 1).
Trace fossils in a desert world – the Triassic Otter Sandstone
Rocks dating from the Triassic period, laid down between approximately 250 and 200 million years ago, are magnificently exposed along the coast of East Devon, representing the western part of the ‘Jurassic Coast’ World Heritage Site. These rocks are non-marine, having been laid down by wind and rivers in the arid interior of the supercontinent of Pangaea, which had formed from the melding together of all the Earth’s previously separated continents (Fig. 2). They are generally very poorly fossiliferous, with the exception of the Middle Triassic Otter Sandstone, seen between the coastal towns of Budleigh Salterton and Sidmouth. This was deposited by life-supporting braided rivers running through what was otherwise a barren desert environment.
The Otter Sandstone fossils are uncommon and generally fragmentary, but record the presence of plants, arthropods, fish and amphibians. In particular, diverse reptiles populated the local landscape. They comprised herbivores, such as the small lizard-like procolophonid in Fig. 3 (left), with stubby plant-crushing teeth, and pig-sized, and somewhat pig-shaped, rhynchosaurs (Fig. 2). Undoubtedly predatory archosaurs (relatives of crocodiles and dinosaurs), with razor-sharp serrated teeth, included the probably rat-sized creature in Fig. 3 (right) and no-nonsense killers several metres in length.
When these rocks were deposited about 245 million years ago, the Earth’s biota was recovering from the devastating end-Permian mass extinction, which took place less than ten million years earlier and was probably caused by volcanism. The Otter Sandstone reptiles lived several million years before the earliest mammals and dinosaurs, and almost a hundred million years before the first birds took to the air.
As well as bones, the Otter Sandstone yields trace fossils, and these provide additional insights into the lives of the Triassic reptiles. Trace fossils are remains not of the organisms themselves, but structures left in the sediment as a result of their activities while alive.
The walkway beneath Connaught Gardens along Sidmouth Seafront is backed by a vertical red sandstone cliff. Scattered among ‘regular’ bedding features in the cliff-face are some more intriguing structures. These have been interpreted by researcher Ramues Gallois as being fossil burrows in what would originally have been fairly soft sand, deposited by migrating river channels. Certainly some, such as that in Fig. 4, show what appears to be an inclined entrance tunnel leading to a larger chamber (oval in cross-section in the cliff) in which an animal could have sheltered from predators or the searing desert heat.
There is evidence from their bone structure that procolophonids, such as the one in Fig. 3 may have been burrowers, but these pet lizard-sized creatures would have been far too small to make the substantially-sized Sidmouth excavations. Fairly similar burrow-like structures from South Africa contain the skeletons of therapsids (‘mammal-like reptiles’; see Fig. 2), which were quite widespread during the Triassic.
These could, therefore, have constructed the Otter Sandstone burrows. The only problem is that no therapsid bones or teeth have ever been found in these deposits. They may turn up one day, or alternatively the burrows were constructed by some other creatures, such as the chunky, and evidently common, herbivorous rhynchosaurs, which had a stout ‘beak’ and claws suitable for digging up roots, and perhaps also digging holes.
Moving on to a different type of trace fossil, in the 1830s, some very unusual footprints were observed in Triassic rocks in Germany. They bore a rather spooky resemblance to human hand prints, so much so that, in 1835, they were christened ‘Chirotherium’ by Johann Jakob Kaup, from the Greek for ‘hand beast’. Similar prints of approximately the same age were subsequently recognised elsewhere in the world, including, recently, in the Otter Sandstone of Devon (Fig. 5).
What sort of creature was the ‘hand beast’? First off, it is evident that the resemblance of the prints to human hands is superficial only and the distinctive ‘thumb’ was a curved toe lying on the outside, rather than inside, of the foot. But that was comparatively little help in establishing the maker of the prints. Johann Kaup conjectured that it might have been an enormous marsupial. The famous British palaeontologist, Richard Owen, in 1842, thought instead that it was more likely to be a large amphibian, remains of which had at least been found in British Triassic rocks.
It is now known that the chirothere printmakers were almost certainly archosaurian reptiles, belonging to a currently rather untidy taxonomic grab-bag known as ‘rauisuchians’. Rauisuchians, teeth and bones of which can be found in the Otter Sandstone, were terrestrial predators that were probably occasionally bipedal and would have looked superficially like theropod dinosaurs, although were more closely related to crocodiles. At up to five metres or so in length, they were certainly apex predators in the mid-Triassic landscape. They died out at the end of the Triassic period, as the true dinosaurs were in the ascendancy.
Footprints are found at several levels near the top of the Otter Sandstone, mostly in red mudstone bands that were probably deposited in wide and very shallow lakes. One such layer can be seen in favourable conditions on the foreshore both sides of Sidmouth, at sites two kilometres apart, so was of wide extent. Judging from the density of the exposed footprints (often several per square metre), there are probably millions of them just in this one thin layer around or beneath Sidmouth town.
How many animals or separate visits does this particular footprint layer represent? Since the prints show a range of sizes, they were not produced by a single very busy individual; rather, it seems that these creatures were gregarious. It is not known, however, what timespan the thin layer represents – it could have been merely days, a season or many years, so it is not possible to establish how many were present at any one particular time.
What brought these animals to these sites? Perhaps they gathered to drink, although if this was the main purpose, one might expect the footprints to be concentrated around the margins of the water bodies rather than throughout them. Another possibility is that they were there to feed. Occasionally found in close proximity to the footprints are small clusters of broken bone. These seem all to derive from temnospondyl amphibians, which would have resembled overgrown armour-plated newts, reaching several metres in length (Fig. 6).
On close examination, some of these bone fragments show grooves that could well have been produced by serrated archosaur teeth; others show an unusually finely pitted surface similar to the acid wear seen on bones that have passed through the digestive tracts of modern-day carnivores. It seems, therefore, that they could be the defecated or regurgitated remains of rauisuchian meals.
The problem with feeding as an explanation for gatherings of archosaurs is that the temnospondyl fragments evidently belonged to creatures consumed and digested elsewhere. Apart from occasional shed archosaur teeth that had become blunted from over-use, the lake sediments the prints are preserved in yield no other animal fossils, and certainly no traces of unconsumed temnospondyls. The immediate environment therefore apparently lacked any organisms that could plausibly have been prey. The water may have been rendered inhospitable due to elevated salinity or extreme temperature fluctuations. Rather than feeding or drinking, then, perhaps the animals took to the water as a means of cooling down from the effects of the ferocious desert sun or even removing ectoparasites.
There is evidence from chirotheriid track sites preserved in Germany that groups of these archosaurs undertook lengthy, and presumably seasonal, migrations along the margins of a long-vanished sea. It could be that the Sidmouth animals were doing the same, conceivably even to or from America or continental Europe, which were at that time all part of the same continuous landmass. Perhaps what is now the Devon coast represented a way-station on a gargantuan desert trek.
Caught in the act – the marine Lower Lias
As the Triassic period gave way to the succeeding Jurassic, around 200 million years ago, the supercontinent of Pangaea was beginning to fragment and Southern England became an island archipelago, mostly submerged beneath the sea. This sea deposited the richly fossiliferous grey mudstones and limestones of the Lower Lias, which are now finely exposed around Lyme Regis, on the ‘Jurassic Coast’, close to the Devon/Dorset border, and also the north Somerset coast, bordering the Bristol Channel.
Various reptiles returned to the sea in the Triassic period and, by the Early Jurassic, had diversified into groups such as the long-necked plesiosaurs and sleek, superficially fish-like ichthyosaurs, both of which are known from fabulously preserved skeletons from the Lower Lias. Ichthyosaurs, such as that in Fig. 7, gave birth to live young and remarkable fossils from the Lower Jurassic of Germany (and also recorded from southern England) preserve embryonic ichthyosaurs within, or emerging from, the bodies of their mothers. Such intimate associations between preserved organisms are a further type of evidence casting light on the lives and lifestyles of ancient animals, including reptiles. The organisms involved need not be the same species or even remotely related, but could be, for example, predator and prey.
Ichthyosaur jaws bristled with sharp pointed teeth indicating predatory habits, but the teeth alone can say little about specifically what was eaten. Commonly encountered in the Lower Lias are coprolites, or fossilised droppings, reasonably uninspiring to look at, brown-grey in colour, generally oval in shape and sometimes with a coiled structure. They often contain recognisable remnants of prey, principally fish scales and bones. The problem is precisely relating them to the animal that produced them. It is very likely that many were produced by ichthyosaurs, but others would have come from creatures such as sharks.
More interesting and informative is finding the food before it has left the predator’s body. The ichthyosaur in Fig. 8 (top), from the Somerset coast, is incomplete (although would probably have been whole before erosion got to it), but what remains is very well preserved. Dark traces of skin drape some of the bones, but more importantly here, the contents of its stomach are visible behind the rib-cage. In close-up, they are seen to be made up of numerous hooklets from the arms of squid-like cephalopod molluscs, which swarmed in the Liassic seas alongside the well-known and reasonably closely related ammonites and belemnites (Fig. 8, bottom). These molluscs may well not be all that this ichthyosaur species ate, but here is unequivocal evidence that, on at least one occasion, this particular individual did. It was its final meal before death from causes unknown (presumably not indigestion).
Several other deposits in the Southern English Jurassic, primarily along the ‘Jurassic Coast’, are rich sources of fossil reptiles, including the marine Kimmeridge Clay (finely displayed in the Etches Collection at Kimmeridge), and the lagoonal Purbeck beds, which overlap into the Cretaceous. But it is to the Cretaceous that we now turn.
Clues from the sediments- the Cretaceous Wealden beds
Towards the end of the Jurassic, around 145 million years ago, the sea withdrew from Southern England and, not long after, in the Early Cretaceous, a thick series of multi-coloured mudstones and sandstones known as the Wealden beds was deposited by meandering rivers. There are spectacular exposures, totalling several kilometres of constantly-eroding boggy cliffs, along the southwest coast of the Isle of Wight. They have long been known as a source of diverse dinosaurs and other terrestrial organisms.
They also yield trace fossils capturing small episodes in the lives of individual dinosaurs. Footprints are fairly frequently encountered, recording animals wandering for reasons unknown as the chirothere printmakers did in Devon over a hundred million years earlier. The Wealden rocks can also be used to illustrate another type of evidence for incidents in the lives of extinct reptiles and this is provided by the character of the enclosing sediments themselves.
Fig. 9 shows some smooth round pebbles from the Wealden beds. They do not look particularly exciting and seem to have no connection whatsoever with the reptiles that lived at the time.
If they had been found in layers rich in similar pebbles, it would be most logical to assume that they had all been transported by the particularly energetic water flows that produced the other gritty and pebbly bands occasionally seen in the Wealden. But their sedimentary context is more unusual than this. They were all found on their own in otherwise fine-grained massive floodplain mudstones. They were also not local stones, but originated a distant 150km or so to the west in an area of uplands occupying much of what is now Devon and Cornwall.
So, how did these rogue stones end up where they did? There is a reasonably high chance that they were carried in the guts of dinosaurs. It is known that a number of herbivorous dinosaurs carried clusters of tough pebbles in their gizzards to help grind down tough plant material (marine reptiles sometimes did too to act as ballast). When these stones became too smooth and polished to do the job properly, they were regurgitated. So that might explain these stones: coughed up by plant-eating dinosaurs as they plodded across the floodplain.
Most of the Isle of Wight dinosaur fossils themselves are incomplete, their bodies clearly having remained unburied for a long period, their bones scattered by water movements, and sometimes scavenged. In common with most fossil reptile remains, it is not possible to ascertain how most of the Wealden dinosaurs actually died; but there is an exception.
There is a metre-thick layer of reddish-green sandy mudstone in Brighstone Bay, on the southwest coast of the island, known as the Hypsilophodon Bed – so named because it yields virtually no fossils except for well-preserved remains of a small (dog-sized, with a much longer tail), bipedal, plant-eating dinosaur called Hypsilophodon (Fig. 10). On close inspection, the Hypsilophodon Bed is actually seen to comprise two layers, the upper of which, about 50cm thick, seems to be the richest, and this is what will be referred to from now on (although the lower layer is likely to have formed in a similar fashion).
In the 150 years or so since the first example was described, remains of something like a hundred separate individuals have been collected from the Hypsilophodon Bed, and no doubt many more have been lost to erosion. Very unusually, it appears that all, or nearly all, of the Hypsilophodon (commonly abbreviated to ‘Hypsies’) were fossilised as whole skeletons, often with bones and even bony tendons in faithful life position. It seems quite evident that these dinosaurs were buried very rapidly. None of them show any signs of pre-burial scavenging by crocodiles or predatory dinosaurs.
Hypsi remains have been recovered from the entire kilometre length of the exposed Hypsilophodon Bed (which almost certainly has, or at least had, a greater extent than what happens to be visible now). It is a fair estimate that it originally contained thousands of skeletons. Since the bed is quite thin, and individual skeletons can occupy a reasonable proportion of its thickness, there is little reason to doubt that the dinosaurs were all buried at the same time. They were evidently members of a giant herd that experienced some sort of calamity (lots of them were clearly juveniles). What could have killed such huge numbers of these creatures?
The Hypsilophodon Bed is of wide extent with a more-or-less flat base, so it evidently formed on the floodplain rather than within the narrow confines of a river channel, meaning that mud-laden river water must have overtopped the bank. Similar modern rivers (like the Mississippi in the USA) commonly have naturally raised banks known as levees. Exceptionally strong floods, however, can burst through these levees, allowing water to rush over a large area of the surrounding floodplain, depositing sediment as it does so. Heavier sand grains are deposited close to the break in the levee and, as the water spreads out and slows down, finer mud is laid down. This process forms a sedimentary structure known as a crevasse splay.
And such a structure can be seen in the exposed Hypsilophodon Bed section. For one short stretch of the cliff-face, the bed is pure sandstone, containing distinctive ‘climbing ripple’ marks that geologists know from recent observations in field and laboratory are associated with very speedy sedimentation, and which are often seen in modern crevasse splays (Fig. 10, bottom left).
As the sandstone is followed in both directions along the cliff-face, it is gradually replaced by finer sand and clay, until it is just a thin layer at the base of the bed, then just a seam of sand, then absent. This is a ‘fossil’ crevasse splay. It thus seems likely that a major breach of an ancient river bank overwhelmed a herd of Hypsies, which were perhaps browsing on fresh new plant growth nearby, and buried them almost instantaneously in the sediment carried through the breach and deposited on the floodplain. The Isle of Wight rocks have captured and preserved a calamitous event that took place on a single day about 125 million years ago.
Broken bones – the Paleogene Hamstead beds
Shortly after deposition of the Hypsilophodon Bed, the sea returned and remained for many millions of years until the end of the Cretaceous period, laying down a great thickness of sediments, including the Chalk that now forms the ‘backbone’ of the Isle of Wight. Succeeding the Chalk, and the death of the non-avian dinosaurs some 66 million years ago, is a sequence of Paleogene deposits, exposed along the northern coast of the island. Included in these are the clay-dominated Hamstead beds, laid down in lake and lagoonal settings about 35 million years ago. These yield the remains of diverse mammals, and also turtles and crocodylians. Unlike the examples discussed so far, these are reptiles quite closely related to species that are alive today, making their lifestyles much easier to interpret.
And there is indeed evidence that the Hamstead crocodiles (more technically, alligator relatives) were doing what crocodiles do best, and this evidence is directly impressed on fossil bones. Fig. 11 (left) shows a fragment of turtle carapace, which was broken prior to fossilisation. In the centre of the piece is a somewhat diamond-shaped hole that matches exactly the outline of a crocodile tooth, indicating beyond reasonable doubt that this turtle, alive or already dead, was the victim of a crocodile attack.
Having just above discussed the mass death of Hypsilophodon dinosaurs, this article can end on a slightly more cheerful note with a reptile that had a narrow scrape, but lived to tell the tale… at least for a while. Fig. 11 (centre) is the femur (thigh bone) of a crocodile, which had been seriously damaged, quite possibly in a tussle with another croc. The wound wasn’t fatal, however. The bony overgrowth at the broken end of the bone shows that the crippled crocodile soldiered on for some time afterwards, before eventually making its contribution to the fossil record.
About the author
Dr Robert Coram is a director of the rock and mineral gift wholesale company British Fossils (www.britishfossils.co.uk) and a Research Associate at the University of Bristol. His research interests include fossil reptiles, fossil insects and Mesozoic non-marine palaeoenvironments.
Coram, R. A. and Radley, J. D. 2015. Chirothere footprint sites from the Otter Sandstone Formation (Middle Triassic, late Anisian) of Devon, UK. Ichnos 22, 29-42.
Coram. R. A., Radley, J. D. and Benton, M. J. 2017. The Middle Triassic (Anisian) Otter Sandstone biota (Devon, UK): review, recent discoveries and ways ahead. Proceedings of the Geologists’ Association. In Press.
Coram. R. A., Radley, J. D. and Martill, D. M. 2017. A Cretaceous calamity? The Hypsilophodon Bed of the Isle of Wight, southern England. Geology Today 33, 66-70.
Gallois, R. W. 2013. Possible tetrapod burrows in the Mid Triassic Otter Sandstone Formation at Sidmouth, Devon, UK. Geoscience in South-West England 13, 208–215.
Lord, A.R. and Davis, P.G. (eds) 2010. Fossils from the Lower Lias of the Dorset coast. Field Guide to Fossils: Number 13. The Palaeontological Association, London. 436 pp.
Martill, D. M. and Naish, D. (eds) 2001. Dinosaurs of the Isle of Wight. Field Guide to Fossils: Number 10. The Palaeontological Association, London. 433 pp.
Pollard, J. E. 1968. The gastric contents of an ichthyosaur from the Lower Lias of Lyme Regis, Dorset. Palaeontology 11, 376-88.