
OF ALL the “terrible lizards”, there was none quite so strange and imposing as Spinosaurus. This croc-snouted carnivore could reach 14 metres in length and weigh more than 7 tonnes. Those vital statistics put it in the running for the largest terrestrial carnivore of all time. The dinosaur even got a starring role as the central villain in the 2001 movie Jurassic Park III, spurring endless internet debates over .
In reality, such a fight would never have happened: Spinosaurus lived in northern Africa about 100 million years ago, while Tyrannosaurus stalked North America between about 66 and 68 million years ago. But that isn’t to say that confrontations between colossal carnivorous dinosaurs didn’t occur. Difficult as it may be to imagine multiple multi-tonne predators co-existing in the same habitat, such territory sharing was common during the Jurassic and Cretaceous geological periods. The question facing palaeontologists is this: how could multiple giant carnivores carve up a single ecosystem between them?
Advertisement
We are now closing in on an answer. Palaeontologists have spent decades piecing together the daily lives of dinosaurs like Spinosaurus and Tyrannosaurus, charting everything from diet to how these famous beasts grew up. As researchers put flesh on the bones of these huge carnivores and envisage the roles they played in their ancient environment, one thing is becoming clear: we may have fundamentally misunderstood the way they lived and hunted.
For more than a century after the word “dinosaur” was coined in 1842, no one really questioned why gigantic carnivorous versions of these animals – known as megatheropods – grew to such frightening sizes: they did so simply because the world of the dinosaurs was bigger. It was clear even in the 19th century that some herbivorous dinosaurs got mindbogglingly massive. Such big prey seemed to necessitate enormous predators. Megatheropods evolved to hunt megaherbivores.
But living large is rarely so simple. In recent years, an ancient ecosystem that existed in northern Africa about 100 million years ago has been held up by some as the most striking example of how strange dinosaur habitats could be. For years, the ecosystem unearthed in Morocco – known as the Kem Kem – has been revealing itself through a steady flow of fossils. In a 2020 publication, Nizar Ibrahim at the University of Portsmouth, UK, and his colleagues attempted to make sense of the evidence found there to date. .

Among the dinosaur predators, there was the 14-metre-long Spinosaurus, the 12-metre-long Carcharodontosaurus and the 8-metre-long Deltadromeus – along with an as-yet-unnamed megatheropod categorised as an abelisaur that was at least 5 metres long.
Big dinosaur predators probably had large appetites, no doubt to fuel their warm-blooded bodies. In fact, if modern predators are a guide, every megatheropod in the Kem Kem probably munched through several tonnes of food each year, and so the search for prey must have occupied much of their time – probably creating a landscape of fear for prey species. “,” said Ibrahim when the research was published.
Fish-eating dinosaur
Yet it also seems to have been one of the most puzzling places. That is because Ibrahim’s team concluded the Kem Kem contained very few large plant-eating dinosaurs for the megatheropods to eat. How did the predators not starve? For Ibrahim’s team, the answer is to be found in the many waterways that once criss-crossed the Kem Kem. The team proposed that it was fish, rather than large plant-eating dinosaurs, that fed the carnivores found there.
In the eyes of many researchers, Spinosaurus certainly does have the look of a fish-eater – although whether Carcharodontosaurus and Deltadromeus also caught fish has yet to be assessed. Also unclear is how this variety of gigantic predators might have found enough aquatic prey to eat, even allowing for the fact that some of the fish in the Kem Kem are known to have grown to 4 metres in length.

While Ibrahim and his colleagues continue to investigate these questions, others caution that we still can’t be sure that this proposed fish-fuelled, predator-heavy ecosystem existed at all.
It is important to appreciate that the sequence of rocks in which a prehistoric ecosystem is found often spans a vast geographical area and represents millions of years of time. Without detailed records of exactly where in that rock sequence the fossils of a particular dinosaur were found, it is difficult to know for sure which species it co-existed with. These scientific notes are often lacking for the dinosaurs of the Kem Kem.
Naturally, as they are partially located in the Sahara desert, the Kem Kem rocks can be difficult for researchers to work in. Indeed, many of the fossils from the area are unearthed by members of the local community. Some specimens end up fuelling the bustling domestic (and legal) fossil trade – others end up on the illegal black market. In both cases, the precise geological information that palaeontologists require to understand the ecology of the Kem Kem is often lost.
Fossil footprints
This has led some palaeontologists to suspect that the abundance of carnivores in the Kem Kem has been overestimated. “Fossils from the Kem Kem do not reflect the actual full diversity of the area,” says Katlin Schroeder at the University of New Mexico. She thinks that describing what has been found so far as a complete picture of the ancient ecosystem “is no more accurate a representation of ecological diversity than a census at a steak house is an accurate representation of the dietary trends of an entire neighbourhood”.
Dinosaur footprints fossilised in the rocks also throw doubt on visions of a place with a surfeit of carnivores. While not as striking as the skeleton of a megatheropod, fossil tracks can be more ecologically informative. This is particularly the case since footprints stay put, whereas bones and teeth can be carried by rivers or floods and end up being fossilised a long way from where the animal they belonged to lived. In other words, says Matteo Belvedere at the University of Florence, Italy, footprints “are linked to the ecosystem”.
If this ancient environment was full of enormous carnivores stuffing themselves with fish, then that should be reflected in the fossil footprint record. But it isn’t. Most of the tracks found in the Kem Kem seem to have been made by medium-sized theropods, while .
“My personal opinion is that the Kem Kem bone record suffers some bias,” says Belvedere. The bone fossils might actually record a variety of different environments and communities of dinosaurs that existed in different parts of the landscape, he says. The bottom line: the most dangerous place on Earth may not have overflowed with hungry carnivores after all.
If that is the case, however, the Kem Kem might have been unusual – because we know for sure that other Jurassic and Cretaceous ecosystems did, in fact, boast several megatheropods living alongside one another. Moreover, through the study of these other ancient ecosystems we are closer to understanding how different sharp-toothed species could coexist.
Take the fern-covered floodplains of western North America in the Late Jurassic, about 150 million years ago. This environment is now preserved in a set of rocks known as the Morrison Formation – and for simplicity’s sake, the dinosaur locale of that time is given the same name. “There really is no place on Earth today that serves as a good ecological model for the Morrison Formation,” says Joseph Peterson at the University of Wisconsin Oshkosh.
On a safari through this corner of the Jurassic world, you might encounter plant-eating animals to rival the size of modern-day megafauna – the 3-metre-long armour-covered Mymoorapelta, for instance, or the 7-metre-long Stegosaurus. But these were far from the largest beasts you would come across. This is because the Morrison Formation was also home to a wide variety of truly enormous sauropods. These long-necked, plant-eating dinosaurs included Apatosaurus, Camarasaurus, Diplodocus and Brachiosaurus. Many of these had the potential to reach 20 metres or more in length, and some may have weighed 35 tonnes or more – as much as a humpback whale.

Then there were the megatheropods. The three-horned Ceratosaurus grew to lengths of over 7 metres. Allosaurus, meanwhile, averaged about 8 metres, but with some exceptional specimens growing a couple of metres longer still. Finally, there was the 10-metre-long Torvosaurus, a predator with a long skull and blade-like teeth. Together, these reptiles were among the first truly giant carnivores Earth had seen – and there is far more confidence that they co-existed because researchers have meticulous geological notes with which to reconstruct the ancient communities of the Morrison Formation.
Of the meat-eating menagerie, it is Allosaurus that palaeontologists encounter most often. Within the Morrison Formation is in central Utah, where Peterson conducts field research. The remains of at least 46 Allosaurus have been found there – with the total number probably reaching into the hundreds.
Even discounting the astonishing abundance of fossils in this quarry, the megatheropods of the Morrison Formation seem to have been common. An analysis of dinosaur ecosystems published earlier this year by James Farlow at Purdue University, Indiana, and his colleagues . So many, in fact, that it seemed like there wouldn’t be enough meat to go around unless the carnivores were almost entirely indiscriminate about what they ate. Yet subtle anatomical differences between the megatheropods point towards different prey preferences. The nature of those preferences is still up for debate, but researchers are beginning to address the broader question: how might each sharp-toothed hunter have found enough to eat here?
Last year, a pair of biologists suggested a partial answer. They used a computer simulation of the ecosystem to explore the Morrison Formation, and calculated that the natural death rate of gigantic sauropods should have been high enough to create a near-constant supply of rotting meat across the landscape. They suggested that at least one megatheropod, perhaps Allosaurus, – like a gigantic flightless vulture moving between free meals. This would have reduced competition.
In addition, assuming that megatheropods only had eyes and stomachs for the largest animals in the landscape ignores something fundamentally important about dinosaur ecosystems. Every dinosaur – even a 25-metre-long Diplodocus – hatched out of an egg that, in most cases, was no larger than a grapefruit.

Although we can be reasonably confident that all the herbivorous dinosaur species were having large clutches of tiny offspring, palaeontologists rarely find them in the fossil record. There are a few potential explanations for this. Perhaps the small and delicate bones of hatchlings were far less likely to fossilise than the larger and sturdier bones of adults. Or perhaps palaeontologists digging for fossils are more likely to overlook such tiny remains.
There is another possibility: young dinosaurs might have been a target – maybe even a preferred meal – of some species of megatheropod. This makes sense, given that young animals would have been easier to catch and kill than the strong adults, which could seriously injure megatheropods with the thrashing of their limbs and tails. To a growing Allosaurus, a .
Voracious megatherapods
Farlow’s team suggested some large carnivores of the Morrison Formation might even have ignored large plant-eaters and their young to some extent. Rather than hunting megaherbivores, the megatheropods might instead have focused on smaller species of dinosaur – mirroring the way that the multi-tonne orcas in today’s oceans will often hunt salmon and seals rather than large sharks and whales. And if even that wasn’t enough? Adult Allosaurus, fossils from a Colorado site document.
In fact, the appetites of these megatheropods might have been so voracious that smaller theropod species couldn’t compete. That has everything to do with how dinosaurs grow up and how Jurassic and Cretaceous ecosystems differed from anything we know today.
“Most modern carnivore [communities] contain small, medium and large carnivores,” says Schroeder. But that diversity is lacking from the ancient environments where megatheropods roamed, according to an analysis Schroeder and her colleagues published last year.
Of course, “large” is a relative term here – in the Serengeti in Tanzania, for example, the largest modern carnivore is the lion, with males measuring about 2 metres in length and weighing about 250 kilograms. But alongside the lions there are medium-sized leopards, cheetahs and spotted hyenas – all in the 60 to 90 kg range – and smaller African wild dogs that weigh no more than about 35 kg.
Compare that with a megatheropod ecosystem. “Dinosaur communities are different than modern communities in that medium-sized carnivorous dinosaurs are suspiciously absent or rare,” says Schroeder. In the so-called Dinosaur Park Formation – an ecosystem preserved in 75-million-year-old rocks of that name in Canada – the largest predators were tyrannosaurs like Gorgosaurus that could weigh more than 2 tonnes. According to the fossil record, the next largest carnivores weighed just 200 kg, leaving a massive carnivore “gap”. In our modern world, this would be the equivalent of there being no carnivores on the Serengeti between the lion and the bat-eared fox – a 5 kg predator that feasts on termites.
The biology of the megatheropods probably explains this gap, says Schroeder. Just like the herbivorous dinosaurs, megatheropods entered the world as tiny hatchlings. Adults laid such large clutches that young megatheropods must have made up a substantial portion of the total biomass of the species – perhaps about 50 per cent. Even though relatively few of those hatchlings would make it to adulthood, the little nipping hordes nevertheless required a great deal of food. Schroeder and her colleagues concluded that the young megatheropods were abundant enough to outcompete other small to medium-sized carnivores by eating small prey, explaining the carnivore gap.
Palaeontologists have moved on from the assumption that gigantic predatory dinosaurs spent their days wrestling gigantic herbivores to the ground. While it is probably true that megatheropods did sometimes attack megaherbivores, we are beginning to build a more complete picture of their diets and behaviour. Some of these enormous predators probably picked off hatchlings and other small animals. Others busted up festering carcasses and yet more probably plucked fish from ancient waterways. In other words, there was no single way to be a megatheropod. The only common denominator may be this: where there was a maw, there was a way.