Many modern predatory birds have enlarged claws on their second toes, similar to those of their paravian dinosaur ancestors – with seriemas being a particularly good example.

Seriemas are part of a lineage known as cariamiformes, highly terrestrial birds that were widespread across most of the world but are today represented today by only two living species in South America. During the Cenozoic this group repeatedly evolved into large predatory flightless forms like the the phorusrhacids and bathornithids, and were probably the closest avians ever got to recreating the “carnivorous theropod” body plan and ecological niche.

And yet none of them ever seem to have experimented with more dromaeosaurid-like claws.

…With one known exception.

Qianshanornis rapax here lived in East China during the mid-Paleocene, about 63 million years ago. It was a small cariamiform, probably around 30cm tall (1″), and is only known from fragmentary fossil material – but part of those fragments was a fairly well-preserved foot. And the bones of its second toe were unlike any other known Cenozoic bird, shaped incredibly similarly to those of dromaeosaurids and suggesting it may have had the same sort of big hyperextendible “sickle claw”.

While it had sturdy legs and short wings, and probably spent a lot of time walking on the ground like other cariamiformes, it was probably also still a fairly strong flier based on the known anatomy of its arms and shoulders.

Unfortunately, though, its head and claws were entirely missing, so without more fossil discoveries it’s hard to say anything definite about its ecology. I’ve restored it here based on other predatory cariamiformes, but since it was also closely related to a herbivorous species it’s not clear whether Qianshanornis was truly a dromaeosaur-mimic or if something else was going on with that unique second toe.


In the early Cenozoic mammals were rapidly diversifying and evolving. And while it was the placental mammals that would end up being the most successful across much of the world, they weren’t the first mammal lineage to take advantage of all the ecological niches left vacant in the wake of the end-Cretaceous mass extinction.

The cimolestans were a group of non-placental eutherians – mammals closer related to modern placentals than to marsupials – that very quickly evolved into a wide range of niches during the Paleocene and Eocene, becoming some of the largest mammals of their time and producing forms as varied as squirrel-like, otter-like, ground sloth-like, and hippo-like.

But some of the weirdest of them all were the taeniodonts. Originating back in the late Cretaceous, these herbivorous cimolestans were characterized by short blunt snouts with large front teeth, and limbs with long claws.

Stylinodon mirus here was one of the largest taeniodonts, standing around 70cm tall at the shoulder (2’4″), and was also one of the last of its kind, living during the mid-Eocene about 50-40 million years ago in western North America.

It took the specializations of its lineage to the extreme, with a odd-looking boxy skull with enormous chisel-like ever-growing front teeth similar to those of a rodent – but derived from its canine teeth rather than its incisors.

Stylinodon skull | photograph by Yinan Chen | CC0

Its powerful front limbs and large claws were clearly specialized for digging, and for a long time it was thought to be obvious what its diet was – clearly it must have been unearthing roots and tubers from underground, right?

However, closer looks at its teeth raise a problem with that interpretation. That sort of food source should have left numerous telltale marks on the chewing surfaces of its teeth, scratches and gouges and abrasions from dirt and grit mixed in with the roots being eaten.

Yet Stylinodon barely shows any of those wear marks, suggesting that it rarely actually ate those food items. Its tooth surfaces were instead worn very smooth, indicating that it was eating something particularly tough that was constantly “polishing” them as it chewed — but what exactly that food source was is still unknown.

It may also have used its forelimbs to help pull down branches down towards its mouth, stripping off leaves and bark similar to ground sloths, chalicotheres, and therizinosaurs – but it probably did mostly use those big claws to actually dig, just perhaps mainly to construct large burrows rather than to find food.


Echinerpeton intermedium here was one of the earliest known members of the synapsids, the lineage that includes all mammals along with other “reptile-like” stem-mammals such as the famous sailbacked Dimetrodon.

Living during the Late Carboniferous in Nova Scotia, Canada, this 60-70cm long (2′-2’4″) distant cousin to modern mammals was previously known only from the fossilized remains of juveniles – with all known specimens showing slightly elongated spines on their vertebrae that gave it a sort of high-backed “proto-sail” appearance.

But a newly described fossil has completely changed what we know about this animal.

A single vertebrae identified as belonging to Echinerpeton shows a much much longer spine than anything we’ve ever seen before, and confirms that this species actually had a large elaborate true sailback – making it the earliest known tetrapod to experiment with this type of anatomy.

This individual seems to have been older than the other known specimens, but still not fully grown, leaving the possibility that fully mature Echinerpeton may have had even larger sails than this.


Although much less famous than their larger horned and frilled relatives, the leptoceratopsids were a widespread and successful group of ceratopsian dinosaurs during the Late Cretaceous, with fossils known from North America, Asia, and Europe (and, dubiously, Australia).

They were fairly small stocky quadrupedal dinosaurs, sort of pig-like, with short deep jaws and powerful beaks adapted for eating fibrous low-level plants like ferns and cycads – and to process such tough food they even evolved a chewing style similar to mammals like rodents.

Prenoceratops pieganensis here is known from the Two Medicine Formation bone beds in Montana, USA, dating to about 74 million years ago. Around 1.5-2m long (~5′-6’6″), it was very similar to its later relative Leptoceratops, but had a slightly lower, more sloping shape to its skull.

Temnospondyl Toes

The evolutionary origins of modern amphibians are still a bit murky, but one of the most likely possibilities is that they evolved from a group of temnospondyls known as amphibamiformes. (Or, at least, that frogs-and-salamanders evolved from them. Caecilians might be a different type of temnospondyl.)

And a recent discovery adds a little bit more evidence to that hypothesis.

A new specimen from the 309-million-year-old Late Carboniferous Mazon Creek fossil deposits in Illinois, USA, shows some soft-tissue impressions around the body of a terrestrial amphibamiform* — most notably showing its toes, with chunky rounded fleshy pads at the end like those seen in many modern amphibians.

Fossil trackways already suggested that some terrestrial temnospondyls had chunky toes, but up until now all known soft-tissue impressions only showed the slender tapering toes of aquatic forms. This is the first direct fossil evidence of toe pads, and hints that a lot of modern amphibians’ soft-tissue features may have actually had a very ancient origin.

(*A more precise identification couldn’t be made, but it shows some similarities to both Doleserpeton and Pasawioops.)


Since the last couple of weeks have featured marine mammals, let’s have one more! This time not a cetacean but a member of the other group of fully aquatic mammals still alive today: the sirenians.

Although commonly known as “sea-cows” due to their herbivorous grazing habits, sirenians’ closest living relatives are actually modern elephants. They’re thought to have originated in Africa over 50 million years ago, starting off as pig-like or hippo-like semi-aquatic animals — but they must have been good swimmers capable of crossing oceans very early in their evolutionary history, since some of the earliest known sirenian fossils actually come from the other side of the Atlantic on the Caribbean island of Jamaica.

Sobrarbesiren cardieli here extends some of our knowledge of early four-legged sirenians to Europe, dating to the mid-Eocene about 42 million years ago. Hundreds of bones were found in Northeastern Spain, representing at least six different individuals and giving us a fairly complete idea of this species’ anatomy.

It was smaller than modern sea-cows, reaching about 2m long (6’6″), and seems to represent a transitional point between the semi-aquatic ancestral sirenians and fully aquatic later forms. It had a head very similar to its modern relatives, and probably a tail fin, but also still retained small functional hind limbs.

It was initially thought to still be somewhat semi-aquatic and capable of quadrupedal locomotion on land, but a later analysis of its hind limb bones suggests that it may actually have been much more aquatic than that. Its hind legs had a wide range of motion and were probably used for otter-like swimming, undulating the body while paddling, but might not have been capable of supporting its weight on land. So if Sobrarbesiren did still haul out of the water, it may have had to move more like a seal.


Last week’s weird-snouted Furcacetus wasn’t the only recently-discovered ancient platanistoid dolphin that deserves some attention.

Ensidelphis riveroi was described in the same paper, and also lived in the coastal waters around Peru during the early Miocene, about 19 million years ago. It was a little less closely related to its modern river-dwelling cousins than Furcacetus, and was slightly larger, estimated to have measured about 3m long (9’10”).

But what made it weird was its incredibly long snout, lined with around 256 tiny sharp teeth, which also curved markedly to the right side along its 55cm (1’10”) length.

Expectation vs reality

With only one known skull of Ensidelphis it’s impossible to tell if this was a natural condition for the species or if it was some sort of anomalous individual. It doesn’t seem to be a deformation of the fossil, at least.

Similar unusual right-side bending has been seen in the skulls of a few individuals of modern South Asian river dolphins, franciscanas, and Amazon river dolphins, possibly caused by injuries at a young age being exaggerated as the animals grew. However, many other platanistoid dolphins (especially squalodelphinids) are known to have naturally had similar bends in their snouts – but always to the opposite side, curving to the left instead of the right.

But naturally bent or not, what might Ensidelphis have been doing with that incredibly lengthy snoot?

Its long slender jaws would have had a fairly weak bite, so it probably wasn’t able to catch large prey, and it had a very flexible neck. Possibly it swam along near the seafloor using its snout to probe and sweep around in the sediment for buried small prey.

Modern South Asian river dolphins swim along on their sides while doing this – almost always on their right sides, interestingly enough – and if Ensidelphis did the same sort of thing then a snout bent in that direction might have been an advantage.


The two living subspecies of the South Asian river dolphin are the last surviving members of a lineage known as the Platanistoidea, an early evolutionary branch of the toothed whales. This group was once much more diverse and widespread than their modern representatives, found in oceanic habitats around the world from the Oligocene to the mid-Miocene.

Many of them had forward-pointing protruding teeth at the tips of their snouts, resembling those of some plesiosaurs or pterosaurs, suggesting they were a convergent adaptation used for snagging hold of slippery soft-bodied aquatic prey.

Furcacetus flexirostrum is one the newest additions to this group, named and described in late March 2020. It lived in Pacific coastal waters around Peru during the early Miocene, about 19-18 million years ago, and was about the same size as modern South Asian river dolphins at around 2.3m long (7’7″).

And it had a uniquely-shaped snout for a cetacean, curving upwards for most of its length but then turning downwards right at the tip, which along with large forward-pointing teeth gave its jaws a vaguely crocodilian appearance.

A closeup view of the jaws of Furcacetus.

Much like slender-snouted crocodilians and spinosaurids, this arrangement would have allowed Furcacetus to make quick bites at small-fast-moving prey like fish and crustaceans.

Eons Roundup 6

Time for some more recent commissions from PBS Eons!

The hyainailourids Megistotherium osteothastes and Hyainailouros napakensis, from “When Giant Hypercarnivores Prowled Africa

The bear-dogs Daphoenus demilo and Amphicyon giganteus, from “The Forgotten Story of the Beardogs

The early panda Ailuropoda microta, from “The Fuzzy Origins of the Giant Panda

Weird Heads Month #31: What Even Is This Fish

For the final entry in this series, let’s take a look at a modern weird-headed species – and where better to find some of the strangest and most unique-looking animals alive today than the deep sea?

Malacosteus, also known as the stoplight loosejaw, is a 25cm long (10″) genus of dragonfish found at depths of over 500m (1640′) in oceans all around the world, with the exception of the Mediterranean and polar waters. Two different species are currently recognized, with Malacosteus niger here known from just below the Arctic Circle down to the southern reaches of the subtropics, and Malacosteus australis ranging from there to around 45°S, and up towards the equator in the Indian Ocean.

And there’s a lot to unpack here with the anatomy of this one.

First of all, there’s the fact that its entire head can hinge away from its body, gaping enormous jaws with long fang-like teeth.

The bottom of its lower jaw has no skin membrane connecting the two sides, attached to the rest of its bizarre head only by the hinges and a single exposed muscle, reducing water resistance so it can shoot its trap-jaws out extra fast to snare prey.

Diagram showing how the stoplight loosejaw's jaw parts articulate.
From Kenaley, C. P. (2012). Exploring feeding behaviour in deep-sea dragonfishes (Teleostei: Stomiidae): jaw biomechanics and functional significance of a loosejaw. Biological Journal of the Linnean Society, 106(1), 224-240. doi.org/10.1111/j.1095-8312.2012.01854.x

Once it catches something it retracts its head, and several sets of pharyngeal teeth further back grab hold of its prey and direct it down its throat.

(Let me remind you that this isn’t an early April Fools joke. This thing is completely real.)

In addition to all that anatomical weirdness, it’s also one of the only deep-sea fish that can both see and produce red-colored light. Most creatures living at that depth have lost the ability to see red since that frequency doesn’t penetrate so far down through water, but the stoplight loosejaw has evolved to take advantage of that by using bioluminescent red light as its own personal night vision goggles.

Using large red photophores under each eye, it can shine a spotlight out ahead of itself and see other deep-sea animals all clearly lit up, while remaining completely invisible to both them and any nearby larger predators. It’s able to perceive the color red thanks to a pigment in its eyes modified from chlorophyll, a visual setup unique to this fish and not known from any other vertebrate.

It also has a smaller green photophore further down on its head – inspiring its common name thanks to the resemblance to traffic lights – and many smaller blue and white ones over its head and body.

So, with its highly specialized jaws and ability to see things other deep-sea animals can’t, the stoplight loosejaw must be hunting something pretty impressive, right?

And as it turns out, it eats… plankton.

The vast majority of its diet appears to be copepods, small zooplanktoic crustaceans that are incredibly common in the waters the loosejaw inhabits. It may simply be “snacking” on such a convenient food source in-between rare encounters with larger prey – but it may also be getting the chlorophyll-based pigment needed for its night vision from eating them.