Numerous groups of reptiles have “returned to the water” and become aquatic over the last three hundred million years, but tracing their direct ancestry can be surprisingly difficult. Highly modified and specialized anatomy, lack of transitional forms, and similar features convergently evolving multiple times can all obscure relationships, making it hard to properly classify them.
We’re only just starting to figure out the true origin of turtles (they’re probably archosauromorphs), and they’re a marine reptile group with living members.
Some of the completely extinct ones are even more uncertain. For example: mosasaurs. (Represented here by the eponymous Mosasaurus.)
While some semi-aquatic early mosasaurs are known, and they seem to be closely related to aigialosaurs and dolichosaurs, their exact placement within the squamates is a lot less clear. Traditionally they were regarded as the sister group to snakes, but some studies have found them to be closer to monitor lizards instead, and others have even placed them as much more basal scleroglossans. Their classification in phylogenetic analyses is “highly unstable”, changing depending on what other reptile groups are included, so there’s no real current consensus.
(And even if they are most closely related to snakes, that doesn’t necessarily help much – the exact origin and evolution of snakes is still very poorly known, too!)
Horse evolution is often represented as a simple progression from Eohippus* to modern Equus, but it was actually a lot more complicated than that – and some ancient horses had some very odd things going with their snouts…
(* For a long time Eohippus was considered synonymous with Hyracotherium, but more recently has been split back off as its own genus again.)
And the purpose of these holes is still unknown. Although superficially similar depressions are seen in various other perissodactyl groups, they vary in position and structure and probably weren’t all homologous.
Ideas have included resonating chambers, some sort of glands, inflatable sacs, or attachment sites for complex lip musculature.
Meanwhile Hippidion from the Pleistocene of South America (2 million – 10,000 years ago) had especially long and domed nasal bones. This must have supported an enormous nasal area – possibly giving it a saiga-like air-conditioning system, a highly sensitive sense of smell, or perhaps even some sort of prehensile proboscis-like snout.
Unless we find some exceptional soft-tissue preservation, the facial anatomy of these equids is going to remain enigmatic.
Amiskwia was a tiny soft-bodied creature from the Middle Cambrian, known from a fairly small number of fossils – about 18 specimens from the Burgess Shale in Canada (505 mya) and an additional one from the Maotianshan Shales in China (515 mya).
Despite only measuring about 2.5cm long (1”), it was one of the larger animals alive at the time. Its body features a head with two tentacles and a small mouth, a pair of stubby fins, and a flattened paddle-shaped tail, suggesting it was an active swimmer. Its internal anatomy has been well-preserved in some specimens, revealing a brain, gut, and traces of what may be blood vessels and a nerve cord.
But we don’t know what type of animal it is. At all.
It was initially thought to be an early arrow worm. However, fossils of Cambrian representatives of that group have since been found, and Amiskwia lacks their characteristic spines and teeth. A relationship to ribbon worms or molluscs has also been suggested, but these hypotheses have the same problems with missing key features.
So, for now, Amiskwia remains one of the “weird wonders” of the Cambrian Explosion with no obvious affinities to any other known group.
In 1923, paleontologist Charles Camp recorded the discovery of an unusual-looking skull from the Late Triassic (~220 mya) of Arizona, USA. He made a field sketch before attempting to remove the fossil from the surrounding rock – only for it to completely fall apart, leaving just a couple of intact fragments covered in odd bony knobs.
At first it was classified as a proterochampsid, a group of archosauriforms known from South America. But this classification was based on some additional skeletal remains that were thought to belong to it, which were later split off and named as the semi-aquatic Vancleavea instead. It’s also been compared to Doswellia and the pseudosuchian Revueltosaurus.
The material is just far too fragmentary to make a confident identification, and the original sketch is anatomically unclear. At best we can say that Acallosuchus was an “indeterminate diapsid” – some sort of reptile, but for now nobody knows what.
I’ve restored it here based mainly on proterochampsids, but any interpretation of this animal is going to be highly speculative until more fossil material is found.
A truly massive amount of biodiversity was lost in this event, with 96% of marine species and 70% of terrestrial species disappearing. Some marine ecosystems seemed to rebound fairly quickly, but overall it may have taken at least 5-10 million years for anything close to full recovery. Terrestrial vertebrates may even have taken up to 30 million years to regain previous levels of diversity.
Or it might have been a result of multiple causes at once, events that wouldn’t have been so severe individually but became disastrous in combination. This is known as the “Murder on the Orient Express Model”: maybe they all did it.
But there’s also a secondary element to today’s mystery. In the aftermath of the Great Dying, a small dicynodont synapsid briefly took over the world. For the first few million years of the Triassic, around 95% of the Earth’s population of terrestrial vertebrates were all Lystrosaurus – no other genus or species of animal has ever dominated to such a degree.
Why did these squat little dog-sized animals survive and thrive when everything else was struggling? They might have been opportunistic generalists able to deal with changing conditions better than other groups, the extinction of most large predators may have allowed their population to explode, or it might simply have been a matter of luck.
Vetulicolians were a group of small marine animals best described as “problematic”, known from the Early Cambrian (~518-507 mya) of China, Greenland, Canada, and Australia. They had bulbous but streamlined bodies with a mouth opening at the front, no eyes, a thick exoskeleton-like cuticle, and a segmented swimming tail. Most also had five pairs of openings which may have been gill slits.
Their evolutionary affinities have been uncertain for a long time, and at different points they’ve been classified as arthropods, chordates, kinorhynchs, basal deuterostomes, or even given their own unique phylum. A genus named in 2014, Nesonektris, has been interpreted as having a possible notochord – making vetulicolians chordates, and potentially placing them close to the tunicates – but their exact relationships are still unresolved.
(Skeemella also complicates matters, having some features considered more arthropod-like than other vetulicolians. But since it’s only known from a single specimen, more fossil material is needed to figure out what’s going on with it.)
Trilobites are common and recognizable fossils, found around the world from the Early Cambrian to the Late Permian (521-250 mya), and ranging in size from 1mm to 72cm (0.03″ – 2′4″). They were some of the first organisms on Earth with complex eyes, and some groups also developed ornamentation like spines, horns, and tridents. The image above depicts a particularly elaborate genus known as Dicranurus.
Occasionally fossils have been found showing fine details of trilobite anatomy like antennae, legs, gills, and digestive organs, and we’ve even recently discovered their eggs.
And yet we don’t really know where they came from. Much like the pterosaurs we started the month off with, trilobites appear suddenly in the fossil record with no intermediate or ancestral forms to definitively link them to other groups. We know they were definitely arthropods, but which arthropods they were most closely related to is still uncertain.
They might be related to the chelicerates (arachnids, horseshoe crabs, and eurypterids), or they might be part of the mandibulates (crustaceans, insects, and myriapods). But the exact relationships of these major arthropod groups are still in dispute, too, and phylogenetic results can vary wildly depending on whether trilobites are included in the analysis or not.
It’s probably going to be some time before any sort of consensus is reached.
Chitinozoans are tiny microfossils (50-2000µm in size) commonly found in marine deposits all around the world between the end of the Cambrian and the start of the Carboniferous (~489-358 mya). Often described as “flask-shaped”, they have a variety of external ornamentation, are sometimes found in linked chains, and are important as Paleozoic index fossils.
But we don’t know what sort of organism actually made them.
They’ve been proposed to originate from a wide range of creatures, but currently the main hypothesis seems to be that they were the egg cases of certain marine animals – such as annelid worms, polychaete worms, molluscs, or even conodonts. Or possibly they might be immature graptolites. Or relatives of living ciliates. So far, though, no single identification seems to have gained any widespread acceptance.
In the early 1970s an opalized dinosaur leg bone in a South Australian gem shop came to the attention of paleontologist Neville Pledge. The specimen’s owner allowed it to be borrowed and studied, and it was eventually named as Kakuru kujani – Kakuru after the Rainbow Serpent of Australian Aboriginal mythology, and kujani after a variant spelling of the Guyani, the local indigenous people. Later the fossil was auctioned off to another private owner and lost to science for nearly 30 years, until finally being acquired by the South Australian Museum in 2004.
But all we really know about Kakuru is that it was some sort of theropod dinosaur. The 33cm (1′) tibia probably belonged to an animal up to about 2m long (6′6″), living during the Early Cretaceous (~125-112 mya), but any placement in a specific group is almost impossible. Based on particular features of the bone – such as a tall and narrow astragalar process – it’s been proposed to be either an oviraptorosaur or an abelisaur. But more recent examinations have concluded the bone’s preservation is too poor for those features to be confidently identified, and consider Kakuru to be a basal coelurosaur or even just a dubious name for an indeterminate theropod.
It’s all a bit of a mess, really, and more and better material is needed to clear up this mysterious dinosaur’s identity.
I’ve restored Kakuru here in three different ways, to illustrate just how varied the interpretations are – on the left, an early oviraptorosaur; in the middle, a generic coelurosaur; and on the right, an abelisaur.
(Yes, the abelisaur is fluffy. South Australia was within the Antarctic Circle during the Early Cretaceous, and while the climate there wasn’t as cold as it is today it was still chilly enough for some floofy insulation to be useful.)
Ammonites (or “ammonoids” in technical terms) are one of the most recognizable types of fossil, found in such high abundance that they’re frequently used to precisely date rock layers. They’re absolutely everywhere in fossil collections, and are even made into jewelry.
So we must already know everything we possibly could about them, right?
Except… we really don’t know what their soft parts looked like.
The fossil record for ammonite soft tissue is surprisingly empty for a group that existed for over 300 million years. A possible ink sac and a few organs have been found, but nothing else.
Based on their other cephalopod relatives, they probably had at least ten arms (the two longer tentacles shown on this Collignoniceras are a little speculative), along with a siphon for propulsion – but until we find that elusive exceptional preservation we just don’t know for sure.