From left to right: Asfaltovenator vialidadi, Torvosaurus tanneri, Giganotosaurus carolinii, & Baryonyx walkeri
But then cladistic analysis in the 1980s and 1990s revealed that some of these theropods weren’t actually closely related at all. Carnosaurs weren’t a natural lineage but instead were highly polyphyletic, with the physical similarities between them seeming to be more due to convergent evolution than direct shared ancestry.
Some carnosaurs were split off and reclassified as more “primitive” types of theropod, while the tyrannosaurs were placed much closer to birds with the coelurosaurs. The remaining “carnosaurs” were just the allosaurids, carcharodontosaurs, and their closest relatives, and some paleontologists now prefer to use the name Allosauroidea for this group to distance it from the previous wastebasket mess.
…But Carnosauria might not be done just yet.
The discovery of Asfaltovenator in 2019 complicated matters once again, with a mixture of anatomical features linking it to both the allosauroids and the megalosauroids (megalosaurids, spinosaurids, and their relatives) – suggesting that these two groups might actually have been closely related to each other in a single lineage after all.
This would potentially return Carnosauria back to something surprisingly close to its original definition, with the various megalosauroids now forming an evolutionary grade leading to the allosauroids.
Most modern meat-eating placental mammals are carnivorans, a group that contains two distinct lineages: the feliforms (cats, hyenas, mongooses, viverrids, civets, linsangs, and euplerids) and the caniforms (dogs, bears, seals, raccoons, and mustelids).
The closest living relatives of these animals are pangolins, and their last common ancestor probably lived sometime between the Late Cretaceous and early Paleocene. But the actual early evolutionary history of the carnivorans themselves is rather murkier.
The earliest known carnivoran-like forms – known as carnivoramorphs – all looked vaguely-genet-like and were an ecologically diverse bunch of small predators, ranging from weasel-sized tree-climbers to fox-sized ground-based hunters, found all across North America and Eurasia during the Paleocene and Eocene. They lacked most of the anatomical specializations of true carnivorans, and didn’t quite fit into either the feliforms or caniforms, but their distinctive carnassial teeth make it obvious they were still very closely related.
From their initial discovery in the late 19th century, through to the late 20th century, these carnivoramorphs were traditionally all lumped together under the name “miacids“. As a result the group quickly turned into a big wastebasket taxon of similar-looking animals, all united more by just not being true carnivorans than by any shared characteristics between themselves.
Miacis parvivorus
But during the last couple of decades this mess has finally started to get cleared up. One distinct lineage of miacid-like animals called viverravids were split off, now thought to be the one of very earliest branches of the carnivoramorph evolutionary tree. Several other “miacids” have also been reassessed and renamed, reclassified as falling into various points in an evolutionary grade between viverravids and true carnivorans, and a coupleof species even turned out to actually be caniforms.
The true carnivorans arose from somewhere within the “miacids” during the mid-Eocene, but it’s still unclear where exactly to draw the taxonomic line between them. Forms like Quercygale and Tapocyon might be very close to the ancestral carnivoran – but they might instead be early feliforms – and some studies have also proposed that nimravids (“false sabertooth cats”) may actually be “advanced” carnivoramorphs instead of early feliforms.
There are also quite a few remaining “miacids” that still need sorting out, especially in the genus Miacis. There have to be other distinct lineages of these carnivoramorphs still hidden in the remaining wastebasket pile, and if we can eventually distinguish them from each other it might help to make early carnivoran relationships a bit clearer.
Troodontids were small bird-like theropod dinosaurs, lightly built with slender legs and sickle-shaped “raptor” claws on the second toes of their feet. They had fairly big brains proportional to their body size, rather like modern birds, and their large forward-facing eyes had good depth perception. Owl-like asymmetrical ears in some species gave them a very keen sense of hearing, suggesting they may have been nocturnal hunters using sound to pinpoint the location of small prey.
The original specimen of the namesake of the group, Troodon formosus, was a serrated tooth discovered in the 1850s, about 77 million years old and originating from the Late CretaceousJudith River Formation fossil beds in Montana, USA. It was so little to work with that it was initially mistaken for a lizard tooth, then during the 20th century it was recognized as belonging to a dinosaur and spent time classified as a megalosaurid, then a pachycephalosaur, then finally as a small theropod similar to the Mongolian Saurornithoides.
In the late 1980s it was merged together with multiple other troodontids (including Stenonychosaurus of speculative “dinosauroid” fame), and since Troodon had been the first of all of them to be named it took priority as the genus name.
And then for a while every single Late Cretaceous troodontid specimen from North America was also lumped into Troodon, turning it into a wastebasket taxon.
The problem was that all these troodontids came from locations separated by thousands of kilometers and millions of years of time, and it’s unlikely that they all actually represented just one single species. But they were only known from rare fragmentary remains, making distinguishing them from each other difficult, and the original Troodon tooth didn’t really have any distinctive features either – it turns out most troodontid teeth all look exactly the same!
It was becmoning increasingly dubious whether Troodon was even a valid name at all, and during the 2010s several paleontologists began trying to sort the mess out. The old names Pectinodon and Stenonychosaurus were revived, and some ‘Troodon’ fossils were also split off and given completely new names, becoming Albertavenator and Latenivenatrix*.
As of 2022, Troodon itself is now in a sort of taxonomic limbo, with some paleontologists abandoning it as a dubious name while others are still arguing in favor of continuing to use it. The name could potentially be properly rescued if the original tooth can be clearly linked to better fossil material, letting Troodon take over priority again from one of the other better-established troodontids, or by defining a new type species similar to what happened with Iguanodon.
…But with how incredibly generic that tooth is, both of those options would be very difficult.
Bellerophonts were small snail-like marine molluscs that were either early gastropods or very close relatives of them. They had symmetrically-coiled shells superficially shaped like those of nautiluses, with about half of the shell covered by their mantle similarly to some modern seasnails, and some fossil shells also preserve hints of banded color patterns.
First appearing in the late Cambrian (~500 million years ago), these molluscs existed all the way until the early Triassic, surviving the Great Dying mass extinction (~252 million years ago) only to go extinct just a short time later (~249 million years ago) – a phenomenon known as “dead clade walking”, when a group just barely scrapes through a mass extinction event but doesn’t manage to actually recover afterwards.
The whole group is something of a wastebasket of similar-looking shells, and might actually be more of an “evolutionary grade” made up of various early gastropods and gastropod-relatives than a single defined lineage.
But there’s also another wastebasket nestled inside this wastebasket: the namesake of them all, the genus Bellerophon.
Bellerophon tenuifascia
Originally named in 1808, this genus has had a huge number of species assigned to it over the last couple of centuries. This gives a false impression that Bellerophon-like molluscs didn’t change for hundreds of millions of years, and it makes figuring out their actual long-term patterns of evolution and extinction much more difficult.
In the last few decades some mollusc paleontologists have been gradually chipping away at Bellerophon, and multiple new genera have been broken off from it. But even today it remains a very bloated mess – there are still well over a hundred named species spanning about 230 million years of geologic time.
Studies do indicate the whole genus is highly polyphyletic, made up of a tangle of multiple different lineages that all really need to be revised and renamed – but there’s a lot of work still needing to be done to clean up this particular wastebasket.
The three living species of elephants are the last surviving members of the proboscidean lineage – but up until the end of the last ice age about 11,000 years ago their relatives were much more numerous and widespread, found on every continent except Australia and Antarctica. Mammoths are probably the most famous of these recently-extinct proboscideans, closely related to modern Asian elephants, but there were also the more distantly-related stegodonts and mastodons…
Traditionally any proboscideans that fell into the evolutionary grade between mastodons and elephants-and-stegodonts were all labelled as gomphotheres. As a result by the late 20th century this group ended up as a wastebasket full of elephant-like forms that didn’t easily fit anywhere else, defined more by what they weren’t rather than by any features they all had in common.
This big collection of gomphotheres was highly diverse. Some species independently evolved similar convergent features, and there was also considerable individual physical variation within species, making the actual taxonomy of these animals very difficult to figure out. But over the last few decades there’s been a lot of revision of proboscidean evolutionary relationships, and gradually the gomphothere wastebasket has been clearing up. Groups like the choerolophodontids, amebelodontids, and anancids have been split off, leaving a more defined lineage of gomphotheres that do have shared anatomical characteristics – distinctive three-lobed trefoil-shaped wear patterns on their molar teeth.
These gomphotheriids were widespread, found across Africa, Europe, Asia, and the Americas – and they were especially successful in the latter. They arrived in North America during the Miocene (~16 million years ago) via the Beringia land bridge, and rapidly spread across the continent and down into Central America. They went on to become the only proboscideans to disperse into South America during the Great American Biotic Interchange, with two different lineages arriving at separate times – Notiomastodon around 2.5 million years ago, and Cuvieronius around 750,000 years ago.
Cuvieronius hyodon, a 2.3m tall (7’7″) South American gomphotheriid with distinctive spiraled tusks.
The exact relationships of the gomphotheriids to other elephant-like proboscideans are still a little uncertain. Both protein sequences and mitochondrial DNA have recently been recovered from 35,000-13,000-year old Notiomastodon specimens, but these studies have given different taxonomic conclusions – with the protein results suggesting gomphotheriids were most closely related to mastodons, and the DNA results suggesting they were much closer to true elephants.
At first much of this dinosaur’s anatomy was poorly understood, and at first it was misidentified as a hadrosaur. The skull remains were fragmentary and ceratopsians hadn’t yet been identified as a group, so Monoclonius‘ horns weren’t even recognized as being horns and a piece of the frill was initially misinterpreted as part of a breastbone.
Once the much better-preserved Triceratops was discovered in 1889, and the existence of ceratopsians was recognized, Monoclonius was re-examined and identified as a similar dinosaur – and three more species were quickly described within the genus, also based on very fragmentary fossils.
Monoclonius crassus
Then for a while afterwards every ceratopsid fossil that wasn’t clearly a Triceratops was then just dumped into Monoclonius, quickly turning the genus into a wastebasket full of dubious indistinct remains.
But then…
Centrosaurus apertus was named in 1904, from the similarly-aged Dinosaur Park Formation in southern Alberta, Canada. It had originally been one of the various species of Monoclonius, but was now claimed to be different enough to deserve its own separate genus name – and this started a decades-long controversy between several paleontologists.
Over the new few decades arguments went back and forth over whether Centrosaurus was actually valid or if it was just a junior synonym of Monoclonius. As more and better ceratopsid fossil material was discovered several other Monoclonius species were eventually split off into their own separate genera, too, creating Styracosaurus, Chasmosaurus, and the somewhat dubious Brachyceratops. But other new species also continued to be lumped into Monocloniusup until 1990, meaning that over its century of existence this wastebasket taxon had at one point or another contained at least 16 different species.
During the 1990s opinion began to turn against Monoclonius, increasingly regarding it as a dubious name. Its original type specimen was a chimera of multiple different individuals (and possibly multiple different species), and it just didn’t have any distinct enough anatomical features to distinguish it from other ceratopsids.
Centrosaurus, meanwhile, was further validated by the discovery of huge bonebeds containing thousands of individuals, making it into one of the best-known of all ceratopsians.
Today Monoclonius‘ name remains attached to a few fossil specimens, but only the ones that are too indistinct to classify as anything else. Some “Monoclonius” have also turned out to actually be juveniles and subadults of other ceratopsians – it seems many young centrosaurines had a Monoclonius-like stage in their growth, before they went on to develop their own species’ distinctive horn and frill shapes.
So Monoclonius may never have been a distinct genus at all – it was just a bunch of different ceratopsian teenagers!
Most of the wastebasket taxa featured this month are completely extinct and known only from fossils, but to start things off let’s take a look at a major example of how even groups with living members could have their classification muddled up for centuries.
The name Insectivora first came into use in the early 1820s, and was used to refer to various “primitive-looking” small insect-eating mammals, with modern shrews, moles, hedgehogs, tenrecs, and golden moles as the original core members.
By the early 20th century insectivorans were considered to represent the “primitive” ancestral stock that all other placental mammals had ultimately descended from, and any vaguely similar fossil species also got dumped under the label. Extinct groups like leptictids, cimolestans, adapisoriculids, and apatemyids all went into the increasingly bloated Insectivora, too, making the situation even more of a wastebasket as time went on.
…Insectivora?
The problem was that the only characteristics that really united these various animals were very generic “early placental mammal” traits – small body size, five clawed digits on the hands and feet, relatively unspecialized teeth, and mostly-insectivorous diets – and attempts at making sense of their evolutionary relationships were increasingly convoluted.
The rise of cladistic methods from the 1970s onwards resulted in a lot of “insectivores” finally being recognized as unrelated to each other, removing them from the group and paring things back down closer to the name’s original definition. The idea that insectivorans were ancestral to all other placentals was abandoned, instead reclassifying them as being related to carnivorans, and the remaining members were recognized as just retaining a superficially “primitive” mammalian body plan.
Just shrews, moles, hedgehogs, solenodons, tenrecs, and golden moles were left, and to disassociate from the massive mess that had been Insectivora this version of the group was instead now called Lipotyphla.
Lipotyphla
But there were still no unique anatomical links between the remaining lipotyphlans. And then once genetic methods became available in the late 1990s, something unexpected happened.
Lipotyphla was suddenly split in half. For a while it was unclear if even the remaining shrew-mole-hedgehog-solenodon group was still valid – hedgehogs’ relationships were especially unstable in some studies – but by the mid-2000s things began to settle down into their current state.
Finally, after almost 200 years of confusion, the insectivore wastebasket has (hopefully) now been cleaned up. The remaining “true lipotyphlans” do seem to all be part of a single lineage, united by their genetics rather than by anatomical features, and are now known as Eulipotyphla.
A few fossil groups like nyctitheriids and amphilemurids are generally also still included, but since this classification is based just on their anatomy it isn’t entirely certain. The only exception to this are the nesophontids, which went extinct recently enough that we’ve actually recovered ancient DNA from them and confirmed they were eulipotyphlans closely related to solenodons.
Ornithoprion hertwigi here was one of the first eugeneodonts found with fossilized skull material, and helped with the early understanding of just how their weird jaw anatomy actually worked.
It lived during the Late Carboniferous, about 315-307 million years ago, in a shallow tropical sea that covered what is now southwestern Indiana, USA.
At only around 50cm long (~1’8″) it was one of the smaller eugeneodonts, and along with a small Helicoprion-like tooth whorl it also had a distinctive highly elongated chin. Similar to modern halfbeak fish this structure may have served a sensory function, helping Ornithoprion to detect prey in dark or murky waters.
But now the recent discovery of Jakapil kaniukura is suggesting a lineage of early thyreophorans actually survived for much much longer than previously thought – all the way into the Late Cretaceous, about 97-94 million years ago.
Just 1.5m long (5′), Jakapil lived in what is now southern Argentina, in an ancient desert with a braided river system. It was bipedal, with a short beak, small arms, and a body bristling with spiky armor, and its unusually deep lower jaw and heavily worn teeth indicate it fed on rather tough vegetation that required a lot of chewing to process.
It’s currently only known from somewhat fragmentary remains, so reconstructions of its full appearance are rather speculative and there’s already been some dispute about whether Jakapil actually was a thyreophoran. One proposal is that it shared a lot of anatomical features with early ceratopsians instead, which if true would make it an incredibly weird armored ceratopsian, and also the first definitive member of that group from South America. But the ceratopsian-like features could also just be due to convergent evolution – and a Jakapil-like dinosaur might actually help explain the only other known dubious South American “ceratopsian” Notoceratops, and the similarly-disputed Australian Serendipaceratops.
But whatever it was – late-surviving basal thyreophoran, southern armored ceratopsian, or even a previously unknown lineage of ornithishcians entirely new to science – it’s an exciting and unexpected discovery.
Around 12cm long ~(4.75″), it had a compact skull, small eyes, a short neck, shovel-like hands, an elongated body and slightly reduced hind limbs – all features that indicate it was a burrowing animal, digging tunnels and feeding on underground invertebrates.
Its exact relationships are uncertain, but recent studies have suggested it was an early amphisbaenian, representing a point in the group’s evolution before the full loss of their legs and the development of their extremely long worm-like shape.
