Paleo Safaris: The Late Devonian Sea

Ohio, USA, 370 million years ago

Dawn rises over a sea 370 million years ago. This is the Devonian period, and what will become the American States of Ohio and New York are submerged beneath salty waves. There are no birds in the sky, and no whales, dolphins or seals cruising the waters. There are not even any turtles or sea snakes. In fact, the marine reptiles and pterosaurs that co-existed with the dinosaurs will not be seen for another 130 million years! As a result, the ecosystem of this time is quite unique and very different to today! So let us take a plunge, and see what creatures lurk beneath the waves!

Ctenacanthus, a “shark” that is distinguished by its comb spines
Image Credit: James St. John,

The first animal we spot is a fish, for even this far back fish still rule the waters. But not just any fish, but what at first glance seems to be a shark! Sharks are an incredibly successful group, unfairly known by far too many people as large, blood thirsty monsters, and have a hugely diverse and fascinating history. Today, they come in a wide variety of different species, from the slow cruising, long living Greenland Shark; to the peaceful bottom dwelling Nurse Shark; to powerful predators like the Great White Shark. But it was during the Devonian period that sharks and shark like fish first evolved. Our fishy friend is a Ctenacanthus, whose name means “comb spine” due to a set of spines that protrude from its fins. These spines possess tubercles on them, giving them a superficially comb-like appearance. The Ctenacanthus cruises through the sea, using its keen sense of smell to direct it towards its next meal, ignoring the weird suction feeling…..


A bite of between 6,000 to 7,400 newtons almost snaps the shark in two! From out of the depths, a mighty predator uses its jaws to manoeuvre the lifeless Ctenacanthus, and before you can process what has happened it swallows it whole in a big gulp. Satisfied with its meal, the giant male Dunkleosteus closes its jaws behind a thick pair of lips and swims away. It will be back.

Dunkleosteus. Our ambush hunter from the depths!
Image Credit: Entelognathus,

This Devonian leviathan is 8 metres long and weighs almost three and a half tonnes. The bite inflicted on the poor Ctenacanthus is among the most powerful of any animal that has ever lived, and certainly the strongest ever delivered by a fish! He is part of the Placoderm class of fishes, a now extinct group characterized by hard armor plating covering their head and upper body. Placoderms are even theorized by some palaeontologists to include the distant ancestors of the vertebrates that would first crawl onto land. Therefore, at a time even earlier than this your ancestors may have been smaller and more peaceful variants of Dunkleosteus! Dunkleosteus’ size is almost unmatched by any animal at this time in earth’s history, with only the filter feeding Placoderm Titanichthys potentially eclipsing it. Dunkleosteus is not only one of the largest animals of the Devonian period, but also one of the largest animals that has ever existed up to this point in earth’s history.

Another fish that can be found in these seas is Cladoselache. Just like Ctenacanthus, this man-sized fish was another early relative of the shark family. However, Cladoselache possesses a few features that are very un-shark like! For one it had a short, rounded snout, lacked claspers (which are important structures used in reproduction) and was not quite as “scaly” as other sharks, with scales being restricted to the edges of the eyes, mouth and fins. But just like sharks it possessed two dorsal fins (though with a spine in front of each fin), paired pectoral fins at the front of the body, powerful jaw muscles, and it would’ve eaten almost any animal it came across. Using its excellent vision, one Cladoselache pinpoints a golden opportunity appearing into view, a school of ray finned fish. Cladoselache is built for speed, and powerful strokes of its tail allow it to dart in and out of the school, snatching fish using jaws full of branch shaped cusped teeth and swallowing them whole. In defense the school sticks together, co-ordinating their movements and sheer numbers to avoid and confuse the shark. But more Cladoselache join the hunt, attracted by the bounty on offer. There is no real pack hunting strategy here, but the corralling of one shark pushes some of the fish into the oncoming opportunistic jaws of another. The action intensifies into a frenzy of activity, with multiple Cladoselache darting quickly in and out of the school, grabbing all the fish they can. After a few hours the school has diminished in size, and the Cladoselache have quenched their hunger. The sharks once again go their separate ways, returning to their never-ending patrolling the deep blue sea.

Cladoselache. It hunting shoals of fish may be speculative, but it certainly would’ve been a fast hunter of the Late Devonian Sea!
Image Credit: Nobu Tamura,

The Dunkleosteus is not interested in mere fish schools, and is content to cruise the water column, looking for a rocky haven to rest in for the night. All in all, he’s had a very good day; he has successfully hunted and fed himself whilst avoiding serious injury. This sounds like a simple task, but predators of all shapes and sizes often fail in their hunts, and accidents can happen! As a result, it is great to not waste precious energy on an unsuccessful one. On a different day the Ctenacanthus might have been more aware of the danger and could have narrowly escaped the Dunkleosteus’ mighty jaws. Despite this, the Dunkleosteus feels weirdly uneasy. Maybe it’s the strange suction feeling he’s only just noticed…


From out of the depths another Dunkleosteus plunges its sharp plated teeth into the soft tail of the big male. The attack devastates it, scything through muscle and blood vessels, and cripples the males swimming capabilities. Slowed down, and loosing strength rapidly, he is helpless as the newcomer rounds for another attack. The newcomer targets the body, and his 2nd bite cleaves through the male, stabbing into his internal organs. The sudden violent trauma, blood loss and punctured organs are too much for the male and his day, and life, are ended in the worst possible way. Dunkleosteus are opportunistic and eat anything they can sink their teeth into, including members of their own species. Even our male has gone after smaller Dunkleosteus on more than one occasion. After the second Dunkleosteus is done with his meal. Ctenacanthus gather to pick what is left before the body drifts down to the oxygen poor sea floor. Now that the scary predator is just a carcass, he is a lot more inviting for these opportunistic shark relatives! Eventually the Ctenacanthus go as far down as they dare, and after they leave the rest of the male comes to a rest on the muddy bottom. The oxygen quantity is poor here, so the Ctenacanthus don’t follow, and over time the armored head is buried by the movement of murky sediment which, in conjunction with the lack of oxygen, will shield it from further breaking and decomposition. 370 million years from now future vertebrates, in the form of humanity, will unearth the males giant head, where through careful study, they visualize how he looked and may have lived his life!

With the males passing, we end our journey through the Late Devonian Sea.

References/Further Reading

• The inspiration for this Paleo Safari came from rereading “Evolution: The Story of Life” by Douglas Palmer & Peter Barrett. In particular this story first came to mind from reading pages 80-81; “A Giant of the Devonian Deep” with the beautiful illustration of Dunkleosteus and Cladoselache done by Peter Barrett.

Palmer, Douglas, Barrett, Peter, “A Giant of the Devonian Deep”, “Evolution: The Story of Life”, octopus publishing group, 2009, pg 80-81

Carr 2010, a paper about the Paleoecology of Dunkleosteus, showing is preferred habitat and effect on the ecosystem both in life and in death

Carr, R. K. (2010). Paleoecology of Dunkleosteus terrelli (Placodermi: Arthrodira). Kirtlandia, 57, 36-45.

Anderson & Westneat 2009 paper examining the feeding of Dunkleosteus using a biomechanical model. This allowed them to calculate its staggeringly powerful bite force

Anderson, P., & Westneat, M. (2009). A biomechanical model of feeding kinematics for Dunkleosteus terrelli (Arthrodira, Placodermi). Paleobiology, 35(2), 251-269. doi:10.1666/08011.1

A Cleveland Museum of Natural History article on Cladoselache fossils found in the area, their anatomy, lifestyle and how they were preserved.

Cleveland Museum of Natural History, “INTRODUCING CLEVELAND’S TOOTHIEST SHARK”, Cleveland Museum of Natural History,,

Acanthostega: The shape of things to come

File:Acanthostega MLCS.JPG - Wikipedia
Acanthostega wondering what you’re looking at!
Image Credit: Conty,

Throughout earth’s history there have been many major leaps in evolution; the evolution of eyes, the first multicellular animals and (from our point of view) the first time our hominid ancestors walked upright. However one that sticks out in a few people’s minds is when vertebrates first hauled themselves out of the water and started walking on land. To illustrate just how big this step was, imagine an alternate reality where it never happened. This parallel world would contain no reptiles, birds or mammals, and human civilisation would’ve never emerged. The vertebrates in this world are comprised of a wide variety of fish species swimming in seas, lakes and rivers across the world alongside a range of molluscs, crustaceans and corals (to name a few). On land the world is still covered in thousands of plant species but the only animals are invertebrates. Beetles, arachnids, and ants of all possible sizes scuttle along the ground. Dragonflies, wasps and flies buzz and dance through the air and worms bury through the soil keeping the ecosystem together. All in all, it is a world radically different to what we know.

As a result documenting how and why this important moment in life on earth occurred is key to understanding the world around us. One animal that has helped palaeontologists to do this is a 60 cm long stem-tetrapod that swam the rivers of Greenland during the Devonian Period (360 million years ago). Its name was Acanthostega gunnari, meaning “Gunnars spiny roof”.

File:Acanthostega model.jpg
A model of a swimming Acanthostega
Image Credit: Dr. Günter Bechly,

While fossils of Acanthostega were first discovered in 1933 (and described in 1952 by Gunnar Säve-Söderbergh, who the species is named after, and Erik Jarvik) the majority of what we know about it comes from a magnificent bone bed, part of the Celsius Bjerg Group rock sequence found in East Greenland, that was discovered in 1987 by a team led by Palaeontologist Jenny Clack. These beds contains the remains of multiple Acanthostega buried and preserved with their skeletons almost completely intact. A paper released in 2016 (Sanchez et. al. 2016) detailed another interesting observation about these fossils. Micro CT synchrotron scans of the interior of the arm bones showed that the bones were still reasonably cartilaginous and had yet to fully ossify (i.e. harden into fully formed bones). This ossification happens as animals mature, so it was deduced that all of the 1987 fossils were of juvenile Acanthostega (roughly around 6 years old) who seem to have been living together. The ossification process also seemed to have progressed further in some individuals than in others, suggesting that there was size variation between members of the group, either through genetic variation, sexual dimorphism or even both. Tragically for this ragtag group of youngsters, it seems that they all died together. It is thought that a flash flood might have washed all of them into a small pool of water. This then dried up after the flood receded leaving them stranded and exposed to the elements, away from the water that kept their skin from drying out.

Acanthostega is a great example of a transitional fossil. Its anatomy is comprised of both basal fish-like features (e.g. internal gills, fish-like teeth, fleshy tail fins and a lateral line system) and derived tetrapod-like features (e.g. simple lungs and limbs tipped with digits). Curiously all these features would have made Acanthostega perfectly suited for its river home. It used its fleshy tail to power itself through its river home, snapping at any fish that wondered too close, and to help locate its prey and navigate through its watery environment it used a lateral line system to sense movement and pressure gradient changes. These features (along with its internal gills) meant it stayed underwater for long stretches of time, though its simple lungs enabled it to take breathes of air if required. What surprised palaeontologists the most about Acanthostega was the structure and function of its limbs and the number of digits on each limb. The limbs were not large or robust enough to bear Acanthostega’s weight for long, meaning it would only rarely spend time on land (if at all). Instead the limbs acted as paddles, aiding with swimming and manoeuvring underwater. This is important because it showed that the early tetrapods didn’t evolve limbs when they started walking on land, but instead first evolved them to better aid them underwater. Then later in time they would adapt this pre-existing feature to use for walking on land. The story is the same for its digits. Each of Acanthostega’s limbs were tipped with 8 digits. This showed that the number of digits on stem-tetrapod limbs wasn’t restricted to a set number (originally thought to have been 5). These early digits would have had webbed and made the early limb a more effective paddle. Then later in evolutionary time digits (like limbs) evolved to help bear and spread out the vertebrate’s weight when it was on land.

File:Acanthostega gunnari.jpg - Wikimedia Commons
A skeletal of reconstruction of Acanthostega. Note its 8 digits, flat skull and paddle-tail.
Image Credit: Ryan Somma,

Looking at all of its features its certain that Acanthostega would have actually spent almost all of its time in water, patrolling the waterways and hunting for small fish and arthropods. Its fish like skull features and relatively weak bite force (adapted more for gripping prey) compared to later tetrapods were perfectly adapted for catching slippery aquatic prey, meaning it didn’t hunt terrestrial animals. Like modern day amphibians Acanthostega would have laid its eggs in water as the eggs lacked a hard watertight casing. Throughout its life Acanthostega would also have had to watch its back! Multiple species of large freshwater fish were alive during the Late Devonian and many of them would have seen Acanthostega as a tasty meal.

So while Acanthostega wouldn’t have been much of a “land lubber”, it was a shape of things to come. This small river dweller helped palaeontologists to figure out the early evolutionary history of the stem-tetrapods and showed that limbs and digits, those features that you use every day, were first developed for underwater use, and only later on evolved for use on land.

All that we know about Acanthostega, the evolution of limbs and digits and how vertebrates first ventured out of the water, couldn’t have been possible without the hard work and dedication of Jenny Clack. Before her work this evolutionary transition period wasn’t particularly well understood. However her meticulous research on every facet of Acanthostega (whose fossils she sometimes gave nicknames to, such as “Boris”, “Rosie” and “Grace”) and its relatives, revolutionised our understanding of this key period of vertebrate evolution. She was one of the world’s leading experts on stem-tetrapods and Acanthostega in particular. This is clear to see as almost every scientific paper released about Acanthostega over the last three decades has carried her name either as a researcher or as a source. Sadly Jenny Clack passed away in March of this year (at time of writing). She will be greatly missed by her friends, family and the wider scientific community. With her passing, the world has lost one of the great palaeontologists.

References/Further Reading

Sanchez et. al. 2016 paper detailing the growth and life history of Acanthostega

Sanchez, S., Tafforeau, P., Clack, J. et al. Life history of the stem tetrapod Acanthostega revealed by synchrotron microtomography. Nature 537, 408–411 (2016).

Clack 2002 paper on the skull roof of Acanthostega

Clack, J. (2002). The dermal skull roof of Acanthostega gunnari, an early tetrapod from the Late Devonian. Transactions of the Royal Society of Edinburgh: Earth Sciences, 93(1), 17-33. doi:10.1017/S0263593300000304

Neenan et. al. 2014 paper on the feeding biomechanics of Acanthostega

James M. Neenan, Marcello Ruta, Jennifer A. Clack and Emily J. Rayfield (2014) Feeding biomechanics in Acanthostega and across the fish–tetrapod transition, Proc. R. Soc. B.28120132689,

Porro, Rayfield & Clack 2015 paper on a 3d reconstruction of an Acanthostega skull. This allowed the trio to infer how Acanthostega caught prey.

Porro, Laura B et al. (2015) “Descriptive anatomy and three-dimensional reconstruction of the skull of the early tetrapod Acanthostega gunnari Jarvik, 1952.” PloS one vol. 10,3 e0118882, doi:10.1371/journal.pone.0118882

Tree of Life web project section on Acanthostega, written by the Late Jenny Clack

Clack, Jennifer A. 2006. Acanthostega. Acanthostega gunnari. Version 13 June 2006. in The Tree of Life Web Project,

Another Tree of Life project section written by Jenny Clack on the definition of “Tetrapod” and how it is debated

Clack, Jennifer A. 1997. The Definition of the Taxon Tetrapoda, 1997, in The Tree of Life Web Project,

The University of Cambridge Department of Zoology news article on the passing of Professor Jenny Clack

Aucott, Rachael, “Professor Jenny Clack, FRS, 1947-2020”, University of Cambridge, 26th March, 2020,

A Science Direct web page about lateral line systems

Science Direct, “Lateral Line System”, Science Direct,

Clack & Neininger 2000 paper on the Celsius Bjerg Group, a rock sequence that Acanthostega fossils have been found in

Clack, J. A. and S. L. Neininger (2000). “Fossils from the Celsius Bjerg Group, Late Devonian sequence, East Greenland; significance and sedimentological distribution.” Geological Society, London, Special Publications 180(1): 557-566.