Like many Mesozoic mammals, it is known only from a couple of teeth. That happened to be in what was once a sea bed. Barely eroded, implying that it didn’t die far away…
Now, this was originally taken to be that it was a sea mammal, one of the first in fact. But, while they do look vaguely similar, eutriconodont molars are not functionally similar to those of piscivorous mammals: seals and cetaceans grasp, eutriconodonts, like carnivoran mammals, sheath.
Recently, studies aligned it with gliding eutriconodonts, Volaticotherium andArgentoconodon. This could suggest that, rather than a sea mammal,Ichthyoconodon was instead aerial; rather than the first sea mammal, it was probably the earliest mammalian aeronaut, predating bats by at least 93 million years.
Its worth to note that Volaticotherium’s hand is noted as “poorly preserved”, with only a few metacarpals being known…
Either way, besides the wing finger or styliforme bone, most of what you see is true to volaticotherine mammals: a large patagia, sprawing hindlimbs, a deep, almost gorgonopsid-like snout and the presence of tarsal spurs (here incorporated into the uropatagia).
In life, flying or not, Ichthyoconodon would have been a carnivore, since it was fairly large by Mesozoic mammal standards (comparable to the closely related Jugulator, weighting over 700 g), large canines and its meat-slicing molars discussed above. I asked Dylan to give it a falcon-like colour scheme, to hint at its nefarious lifestyle.
It is midday, in what will one day be Arizona. To the south, a mountain range rises, a symbol of the extensive geological activity that is shaping this world, changing the landscape in this region of the globe. These are active volcanoes, periodically spewing gases into the atmosphere, darkening the sky and scorching the slopes, into a black dust. Periodically, the world ends.
They border the horizon, so far away from this watering hole, yet their presence cannot be denied. Ashes choke the sky, but also nourish the ground, allowing a small bosque of conifers to prosper around the water. This is a godsend to the local dinosaurs: sauropods travel far and wide across the desert, stopping here for a drink. Ornithischians and tritylodontid synapsids are smaller herbivore guilds, permanent and scarcer, watching for the careless bulk of the giants. Following their herds is a Dilophosaurus, laying in the shade, observing the herbivores relaxing in the pool, while a lone Kayentasuchus rests on the bank, the sourrounding animals giving its meat-slicing jaws a wide berth.
But flying above this scene are the by far most common animals, the pterosaurs. Flocks gather here, having flown great distances, their fluttering and calls drowning the air in a strange symphony. Several land in between the sauropod giants, their groups matting the ground and risingin multicoloured clouds each time a dinosaur comes near, while others climb the trunks like ancient squirrels – and not at all unlike the haramiyidans that they share their trees with. A few swim in the lake, bathing and removing excess dirt from their pelage, launching with vapid and rapid wing beats, joining the others above or in the ground. The vast majority, however, flies above, either to move from between roosting places or to arrive or leave.
Pterosaurs have greatly diversified in the Jurassic, producing a large variety of species from the terrestrial dimorphodontids to the hawking campylognathoidids. The ashes of the Triassic cataclysms opened up a myriad of new possibilities, and these flying animals wasted no time taking advantage, reclaiming old niches and claiming brand new ones. Size has also increased, some species reaching wingspans of 2 meters or more.
One species, however, dwarfs them all.
It starts as a shadow, soaring in the horizon, just closer than the mountain range, flying in from the southwest. It glides effortlessly in the desert thermals, much like the other pterosaurs, and it quickly speeds in, its wing beats sporadic and powerful. As it approaches, however, something seems off. Smaller pterosaurs get in the way, and they seem much, much smaller as they dart beneath the newcomer.
Eventually, it approaches the watering hole, and its full bulk is evident. Its wings shadow the open canopy, their span eclipsing even the sauropods aside from the very largest. Its arrival is met by a launching of many of the smaller pterosaurs, tiny flies compared to the newcomer. Dinosaurs and mammals eye it with caution, several of the smaller herbivores darting for safety. Even theKayentasuchus is wary, bearing its jaws at the sky.
The giant circles the watering hole, looking for a good place to land. It eventually finds a fault in the tree line, just wide enough, and it descends, flapping its wings vigorously as it falls, its hindlegs absorbing the impact as the wilds fold, massive gorilla-like arms touching the ground while the wing-fingers rise as masts. As soon as it is grounded, it begins walking, surprisingly gracefully in an alien hopping motion, before crouching above the water. It wastes no time laying its toothed jaws on the surface.
This is a nomadic traveller, having flown far and wide across the globe. It has travelled over these arid lands for a while, and it sees fit to replenish its water supplies. Ideally, it would continue flying to the northeast until it reaches the Arctic coastline; this time of the year, the beaches are full of ammonite corpses, which vagrants like this tend to take advantage off. The migrant itself has flown over there many times over the years, the route well carved into its brain.
For now, however, it rests: after drinking its fill, it retreats to the shadow, laying cumbersomely on its side, wings and legs free to stretch. The Dilophosaurus is not far, and even at its massive size the pterosaur is vulnerable, though far from defenseless. The dinosaur begins to rise, but the flyer flashes its toothed jaws, prompting it to sit down again. Relaxed again, the pterosaur grooms its pelage with the long and thick teeth. The tail flaps on the ground, leaving small puffs of dust.
Above, smaller pterosaurs begin gathering. They begin threat displays, various hisses and tail swipes, some even landing on the ground to bite the giant’s tail. They begin also piling up on the Dilophosaurus, pecking at its tail and sides. Already nervous, the theeropod rises and runs away, prompting a chain reaction and leading more pterosaurs to chase after it, leaving the resting giant alone. It, however, can’t stick around for much longer, as the dinosaur outruns its aerial pursuers.
Slowly and deliberately, the massive pterosaur rises once more, balancing itself with its powerful forelimbs. Its tail swats a smaller harasser, and it walks away. If landing was difficult, launching will be even more so. It walks up to the borders of the bosque, a small elevation granting it convenient – though not necessary – high ground. Crouching, it vaults. It takes a few tries, but several pushups with its massive arms eventually propel the animal out of the ground, a high enough stroke clearing room for the massive wings to flap and immediately ascend the animal’s bulk.
All of this happens in the span of a few seconds, the titan once again airbone. It circles above the watering hole one last time, rising in the thermals, before it finally flies towards the northeast, its size quickly decreasing again as it glides towards the horizon.
Then a bomb detonates and kills everyone. Amen.
– Black (North America + Asia): Dominated by cimolodont multituberculates and metatherians; several eutherians present. Eutriconodonts until mid-Campanian.
– Red (Europe sans Hateg): Dominated by eutherians; sporadic metatherian and cimolodont remains
– Indigo (Hateg): Dominated by kogaionid multituberculates; a single eutherian specimen.
– Orange (South America): Dominated exclusively by dryolestoids and gondwanathere multituberculates
– Green (India and Madagascar): Dominated by sudamericid gondwanatheres and eutherians; a single non-gondwanathere multituberculate and the haramiyidan Avashishita bachamarensis.
– Brown (Australia): Unknown, possibly primarily australosphenidans and dryolestoids.
– Pink (Africa): Single sudamericid gondwanathere, possibly eutherians.
Schowalteria clemensi: Known only from one skull, but seems comparable to latter taeniodonts in size, ranging somewhere between 10 to 50 kg. A specialised herbivore.
2. Bubodens magnus: Represented by a single tooth. It is enormous by multituberculate standards, and probably indicative of an animal above 16 kg (it is described as “beaver sized”). Presumably a specialised herbivore.
3. Repenomamus giganticus: Only “giant” Mesozoic mammal known from fairly good material. Measuring about a meter long and weighting at least up to 14 kg. A specialised carnivore.
4. Kollikodon ritchiei: Conflicting sources on this one. It may have been up to a meter long, certainly putting it above R. giganticus (monotremes are proportionally much more robust), but some sources also list it as “platypus size”. A molluscivore or piscivore.
5. Oxlestes grandis: Possibly slightly smaller than R. giganticus. There is some debate about how large its skull was (10 vs 7.5 centimeters), though the former seems to be the most convincing measurements for now. A carnivore.
6. Khuduklestes bohlini: “Subequal” in size to O. grandis. Possibly carnivorous.
7. Mesungulatids: Most sources are rather vague on estimated sizes (in part due to the lack of modern analogues, in part due to how rare postcranial material is), but the larger forms likeColoniatherium seem to be around 6-13 kg. Specialised herbivores.
8. Vintana sertichi: Known from only one skull. Estimated to be around 9 kilos. Specialised herbivore.
9. Altacreodus magnus: Known from various specimens. At around 9 kilos, it is the largest of the Hell Creek mammals. Specialised carnivore.
10. Didelphodon vorax: Known from several remains. The largest of the Hell Creek metatherians at 6-9 kilos. Molluscivore or carnivore.
Above you see the phylogenetic trees for Metatheria in the papers for Tsagandelta (Rougier 2015) and Lotheridium (S. Bi 2015).
Both ultimately focus on deltatheroideans – both taxa are part of this group, after all-, but they reach a similar conclusion in regards to metatherian phylogeny, which I took care to highlight above: the closest relative of the clade leading to modern marsupials – just outside of the last common ancestor between them and sparassodonts – (highlighted in red) is the specimen known as the “Gurlin Tsav Skull” (highlighted in orange).
For those that don’t know, the Gurlin Tsav Skull is a specimen from the Late Cretaceous… of Mongolia.
Traditionally, marsupials are thought to have descended from north american species that reached South America in the Paleocene or Late Cretaceous. This makes sense: metatherian diversity was very high in North America during the Late Cretaceous, Australia was completely isolated from the northern continents during this epoch, and ultimately metatherian diversity is higher in South America than in Australia, the latter bearing exclusively one clade, Australidelphia.
Yet, as the cladograms above show, nearly all north american metatherians group within an entirely seperate clade from the south american ones, the only ones not being within this clade being a few deltatheroideans, which are even less closely related to marsupials. The graph clearly seems to indicate that south american and north american metatherian diversities were distinct radiations, with the closest relative being an asian species.
Granted, groups classically considered “stem-marsupials”, the herpetotheriids and “peradectids”, are absent. The latest study I can find on the matter seems to deem the former as part of the “northern” assemblage, while the latter are arranged as closer to the southern line, but given that most of their next closest relatives are “northern species” in these cladograms it seems likely that “peradectids” may be “northern” species as well.
Therefore, for now, we have to default to the Gurlin Tsav Skull.
This leaves us with two possibilities. The first is perhaps the most straightforward and admitely the most well supported: in the midst of the “nothern metatherians”, there was still a north american species more closely related to the Gurlin Tsav skull than to these, and it eventually reached South America and produced the “southern metatherians”. Many taxa evade the fossil reccord, after all, so there is credence to this, especially if this lineage was indeed rare; if “peradectids” – already thought to be paraphyletic – turn out to be close to marsupials after all, they immediately justify this assertion.
The other option, more radical, is that “southern metatherians” actually evolved from asian forms, that arrived to Australia during the Cretaceous or earliest Paleocene, and only later on dispersed across Antarctica to South America during the Paleocene and Eocene.
This hypothesis is not entirely uneresonable. We do know asian groups colonised eastern Gondwana in the Cretaceous: India and Madagascar in the Late Cretaceous bears a distinctively asian squamate and eutherian mammal groups (Sahni 1987 & 1994, DW et al 2006), mekosuchine crocodiles group among Paleogene eurasian clades (Alistair Glen 2014), and there is evidence of possible Cretaceous choristodere teeth in Timor (Umgrove 1949).*
Therefore, we have evidence that some tetrapod groups, even warm-blooded mammals, made the crossing from Asia to Gondwanna during this epoch, perhaps by island hopping from island systems precursor to modern Indonesia.
There is some evidence that metatherians dispersed multiple times across Antarctica, between Australia and South America. Microbiotheria, a group currently represented only by a south american species, were not only represented in the Palaeocene of Bolivia, but also the Eocene of Seymour Island and the contemporary Tingamarra fauna in Australia (Shiewe 2010, Nisson 2010). Though the possible australian Chulpasia species has since been found to be a possibly unrelated metatherian, the newly discovered Archaeonothos shows similarities with non-marsupial south american metatherians (Beck 2015).
Additionally the generally poor fossil Paleogene fossil reccord of Australia – only one pre-Oligocene assemblage, the Murgon Fossil Site -, and virtually none in the Cretaceous, it is a lot easier to make an argument for missing key species than in the very well documented Late Cretaceous and Paleocene north american faunas. Simply put, our window to Australia’s fauna during this epoch is very limited, and the fact that the Tingamarra Fauna contains both groups present in South America as well as a wealthy amount of incertae sedis forms makes it very likely that Australia had a non-australidelphian assemblage as diverse as South America’s.
Instead of a single metatherian lineage colonising Australia, it seems instead likely that there was a diversity comparable to South America’s, and that the latter simply saw a better fossil reccord and a higher number of post-Oligocene survivors.
Additionally, even if “peradectids” are close to “southern metatherians”, its just as equally likely that they were then recent invaders from South America to North America. There is precedent in this: the most basal of pantodonts is the south american Alcidedorbignya, implying a south american origin for this eutherian clade (Muizon 2015), and the fact that the closest known relative to dinoceratans is the south american Carodnia may similarly imply a southern origin for these “ungulates”, among the meridiungulate radiation (Burger 2015). More damning is the presence of a gondwanatherian in the Late Cretaceous of Mexico, a member of a clade that, as the name implies, is primarily gondwanan in range (SVP 2015).
In conclusion, the idea that marsupials and other southern metatherians colonised Australia first rather than South America has some merit. It should definitely be something that needs further exploration, as there is plenty of evidence of laurasian faunal groups having colonised eastern Gondwanna from Asia, and the mysterious Gurlin Tsav skull and Tingamarra Fauna both offer tantalising clues in that direction.
Multiple links in the text
G. W. Rougier, B. M. Davis, and M. J. Novacek. 2015. A deltatheroidan mammal from the Upper Cretaceous Baynshiree Formation, eastern Mongolia. Cretaceous Research 52:167-177
S. Bi, X. Jin, S. Li and T. Du. 2015. A new Cretaceous metatherian mammal from Henan, China. PeerJ 3:e896
Ashok Sahni, New evidence for palaeogeographic intercontinental Gondwana relationships based on Late Cretaceous-Earliest Palaeocene coastal faunas from peninsular India, Washington DC American Geophysical Union Geophysical Monograph Series 01/1987; 41:207-218. DOI: 10.1029/GM041p0207
Ashok Sahni, Guntupalli V R Prasad, Jaeger jean-jacques, C. K. Khajuria, Eutherian mammals from the Upper Cretaceous (Maastrichtian) Intertrappean Beds of Naskal, Andhra Pradesh, India, · June 1994
Krause, D.W., O’Connor, P.M., Rogers, K.C., Sampson, S.D., Buckley, G.A. and Rogers, R.R. 2006. Late Cretaceous terrestrial vertebrates from Madagascar: Implications for Latin American biogeography (subscription required). Annals of the Missouri Botanical Garden 93(2):178–208.
Alistair Glen, Christopher Dickman, Carnivores of Australia: Past, Present and Future, Csiro Publishing, 05/11/2014
J. H. F. Umbgrove, Structural History of the East Indies, 1949
Robin M.D. Beck (2015). “A peculiar faunivorous metatherian from the early Eocene of Australia”. Acta Palaeontologica Polonica 60 (1): 123–129. doi:10.4202/app.2013.0011.
Christian de Muizon, Guillaume Billet, Christine Argot, Sandrine Ladevèze & Florent Goussard (2015) Alcidedorbignya inopinata, a basal pantodont (Placentalia, Mammalia) from the early Palaeocene of Bolivia: anatomy, phylogeny and palaeobiology. Geodiversitas 37 (4): 397-634.
BURGER, Benjamin J., THE SYSTEMATIC POSITION OF THE SABER-TOOTHED AND HORNED GIANTS OF THE EOCENE: THE UINTATHERES (ORDER DINOCERATA), Utah State University Uintah Basin Campus, Vernal, UT, United States of America, 84078, 2015
Thylacosmilus atrox skull.
So, the idea that saber-toothed cats and similar mammals actually had their canines covered by lips is gaining widespread support. With enamel patterns similar to those of normal teeth rather than tusks, clear evience of sheathing and complexity not seen in tusks, it seems clear that most of these mammals followed the pattern of their closest modern analogue, the cloud leopards, and had their long canines protected, rather than visible as in classical depictions.
Or did they?
One lineage of saber-toothed predators does seem to defy these newfound conventions, the thylacosmilid sparassodonts. Already considered highly aberrant mammals by all accounts, thylacosmilids, perhaps unsurprisingly, also have aberrant saber-teeth. Their canines and other aspects of their anatomy offer highly contradictory implications in regards to the soft-tissues in their mouth, aggravating this controversy to levels never imagine.
A carnivore canine’s answer to rodent incisors
For one thing, thylacosmilids have enamel patterns seen more often in tusked mammals than other saber-teeth like machairodontines. As opposed to Smilodon, which has several enamel layers (Feranec 2004, Riviere & Wheeler 2005), thylacosmilid canines are only very shallowly “dressed”.
But, to be fair, this is actually a pattern seen in sparassodonts as whole, with non-saber-toothed species like proborhyeanids showing a progressive decrease in enamel coating. Unless these forms somehow had no lips, it stands to reason that they, like most mammals, had covered canines. Combined with the Hunter-Shreger bands seen in borhyeanids – a rare trait among mammals besides placentals -, this may possibly be simply a form of canine reinforcement.
Thylacosmilid canines do possess another bizarre trait, however: they were ever-growing. Their tooth-roots, extending all the way above the animals’ eye-sockets, essentially functioned like the rodent incisor roots, providing constant fodder for the ever-erupting teeth (Marshall 1978, Maria J. 2005, Mauricio Antón 2013).
In life, this would have made thylacosmilids easily very lucky when compared to other saber-toothed predators, as the frailty of their teeth would be compensated with regeneration. It does, however, pose a problem to animals with sheathed teeth, which would see themselves very likely to puncture their own flesh.
Once again, however, we turn to other sparassodonts. Proborhyeanids, for example, also possess perpetually-growing canines (Mashall 1978), to the point that this has been at times brought up as a possibly common trait between them and thylacosmilids (Maria J. 2005), though not all experts agree that it is a synapomorphy between both groups (Anália Forasiepi 2009).
Once again, these sparassodonts had no saberteeth.
One specimen refered to Proborhyaenidae seems to imply that the canines ceased to grow at some point in sparassodont lifespan (Bond & Pascual 1983), but a proborhyaenid identity for this specimen is controversial, and at any rate does not reflect the general trend in these sparassodonts to have perpetually growing canines (Maria J 2002). Instead, it can be inferred that regular use probably wore these teeth down.
More comparisons between thylacosmilid and proborhyaenid canines are probably needed, especially if both groups developed perpetually-growing canines independently. However, I think that proborhyaenids provide an overall good example as non-saber-toothed predators with the “tusk-like” adaptations seen in thylacosmilids, so the later’s saberteeth do not necessarily imply that they were external.
Lower jaw flanges
Thylacosmilid lower jaws, by contrast, provide the best outright evidence for sheathing among all saber-toothed predators. Most saber-tooths developed lower jaw structures vaguely akin to flanges, extensions that in life would have served as sheaths for the canines. Thylacosmilids are by far the most extreme example, with Thylacosmilus proper possessing extremely long flanges that are almost a third of the whole skull’s depth.
In most restorations, these flanges are depicted as lip-less, giving way to the canines. I personally always found this extremely weird, even before this sheathed-saber-tooth era; virtually all mammals possess lips, and in metatherians suckling is extremely important in early stages of life, as joeys are attached to their mother’s nipples. If marsupials are suspected of less facial diversity than placentals because of this, what chance did the closely-related-but-still-pouched sparassodonts have?
If thylacosmilids did have exposed sabers, we would expect some seriously radical facio-muscular changes in their development, a kind of change not reccorded even among placentals. As it stands, the most logical conclusion is that my suspicions are correct and that their flanges were covered by lips in life.
Which would, by default, imply sheathed sabers after all.
A problem I posited occasionally is how sheathed saber-tooths would clean their flanges. Indeed, an animal like Thylacosmilus would have very long and deep areas where food residue and other waste would gather to extremely detrimental effect. A long tongue should probably be enough, and studies on the thylacosmilid hyoid could definitely be the turning point in this debate.
Conslusion, and Lotheridium
Based on the above, I think that the thylacosmilid “tusk-like” canine anatomy does not imply a different scenario from other saber-tooths. “Tusk-like” canines are found in sparassodonts with proportionally shorter teeth, and the very existence of lower jaw flanges implicates that these fangs were protected.
What does show, however, is that thylacosmilids were basically OP by saber-tooth standards. They had rather robust, perpetually growing canines that were protected by the best mouth covering nature could provide, making these far-less vulnerable than the sabers of Smilodon or any other machairodontine, nimravid or barboroufelid.
To think that these animals were outcompeted by the comparatively less efficient placental predators seems rather laughable, and indeed, the first saber-toothed cats only appeared in South America several million years after the last thylacosmilids became extinct (Francisco et al 2013). I know concepts like “superior” and “inferior” are utterly meaningless in any discussions about evolution, but I wager that thylacosmilids were functionally better saber-teeth than their placental counterparts.
Lastly, no discussion about metatherian saber-teeth is complete without the most recent addition to the roost, the deltatheroidean Lotheridium. Deltatheroideans have long been assumed to be carnivorous mammals specialised to feed on vertebrate prey, so a saber-toothed carnivore, even if weasel-sized, is not terribly surprising.
Comparisons between Lotheridium and thylacosmilids are, as far as I know, nonexistent. We do know that Lotheridium appears to lack lower jaw flanges, but its canines are noted as being laterally compressed (Li et al 2015), a characteristic usually associated with saber-toothed predators. This to me seems to implicate a functional analogy very strongly, even if it was nowhere near as specialised.
In life, therefore, Lotheridium probably also did cover its canines. Though probably lacking the remarkable attributes of thylacosmilids, it served as a nice prototype, being among the first true mammals to develop saber-teeth.
Robert S. Feranec, Isotopic evidence of saber-tooth development, growth rate, and diet from the adult canine of Smilodon fatalis from Rancho La Brea, 2004
Antón, Mauricio (2013). Sabertooth.
Forasiepi, Analía M. (2009). “Osteology of Arctodictis sinclairi (Mammalia, Metatheria, Sparassodonta) and phylogeny of Cenozoic metatherian carnivores from South America”.Monografías del Museo Argentino de Ciencias Naturales 6: 1–174.
Marshall, L. Evolution of the Borhyaenidae, extinct South American predaceous marsupials. Berkeley: University of California Press, 1978.
Prevosti, Francisco J.; Analía Forasiepi; Natalia Zimicz (2013). “The Evolution of the Cenozoic Terrestrial Mammalian Predator Guild in South America: Competition or Replacement?”.Journal of Mammalian Evolution 20: 3–21. doi:10.1007/s10914-011-9175-9.
S. Bi, X. Jin, S. Li and T. Du. 2015. A new Cretaceous metatherian mammal from Henan, China. PeerJ 3:e896