Miocene (Pt 34): The First Kangaroos in Australia

Ekaltadeta

During the Miocene, Australia was further south than it is today. However, it seems that the generally warmer climate of the early part of the epoch more than compensated for this, since we know that there were already coral reefs off the coasts of the main continent and also of New Zealand, which is far too cold for such things today. At the dawn of the epoch, the continent seems to have been largely covered by open woodland but as the world warmed in the Middle Miocene, and Australia edged northward, it became not only hotter, but wetter, until tropical and semi-tropical rainforests became the norm. It was only in the Late Miocene, around 10 million years or so ago, that the climate started drying again, especially in the interior, and the dense jungle began to die away, leading the way for the formation of today's Outback in the following, Pliocene epoch - although, even at the end of the Miocene, the coasts were more heavily forested than most of them are today.

The northward march of Australia also meant that the continent was approaching Asia, pushing up the chain of volcanoes that now form much of Indonesia (although most of the islands themselves are much older than this, since the tectonic plates had been colliding for some time). This, however, was not yet sufficient to allow any land-dwelling mammals to cross over from the north, leaving Australia's strange endemic fauna to live in peace. 

This is not to say that, even this far back, there were no placental mammals in Australia. This is because bats were already there and had been for some time. Our knowledge of the fauna of Miocene Australia is limited by the fact that the great majority of relevant fossils (although, to be fair, not all of them) come from the single site of Riversleigh, in northern Queensland, so that we know far more about what lived there than we do about what was happening elsewhere on the continent. Fortunately, not only is it a particularly rich site in general but, when it comes to bats, it helps that it was largely formed from limestone... and therefore had plenty of caves even that far back in time.

It's likely that these caves, like those of modern times, could hold several thousand bats each, and we have fossils of many different species, most of them from the still-living family of leaf-nosed bats. These include Riversleigha, an an unusually large example of its kind; we only have the skull, and so can't reliably estimate its wingspan, but its teeth and skull shape suggest that it chomped down on large hard-shelled beetles rather than just eating moths and the like. 

Many others are very closely related to bats from Asia and, in some cases, as far afield as France, showing the ability of bats to spread far and wide if they're given enough of a chance. However, Icarops is unusual in that its only close living relative is the bizarre "walking bat" of New Zealand; like that animal, it seems to have been at least as well adapted for walking along the ground on all fours as it was to flying, suggesting that this feature was not unique to the modern species.

However, at least when it comes to mammals, it's the marsupials that attract most of the attention and there were, as we would expect, plenty of them in Miocene Australia. The kangaroos seem to have first appeared at, or slightly before, the dawn of the Miocene; a few earlier fossils exist, of small, probably omnivorous animals, but they're sufficiently primitive, and the fossils sufficiently fragmentary, that it's hard to say where they fit in the blurred line between the first true kangaroos and the last not-yet-kangaroos.

Mind you, I am using the word "kangaroo" rather loosely here, to refer to members of the kangaroo family, the great majority of which are more accurately described as "wallabies". The earliest known kangaroos were all in this latter size range, with typical members of the group generally growing larger as the epoch wore on. Cookeroo is a typical early example; it was slightly smaller than most modern wallabies and lived in the dense forests of Early Miocene Queensland, where it seems to have fed primarily on soft leaves rather than grazing as most modern species do. 

Cookeroo lived early enough that it doesn't seem to have belonged to any of the named subfamilies of kangaroo and, while the fossils are incomplete, it probably had not yet evolved hopping as a means of locomotion, still walking on all fours with a more typical mammalian gait. Dorcopsoides, which lived much later, during the Late Miocene, had a much more modern appearance, in particular, resembling the apparently "primitive" forest wallabies that now live only in New Guinea. The fact it lived, not at Riversleigh, but in the continental interior near Alice Springs may indicate that the increasing dryness of the continent, which would have been more noticeable far from the coasts, was already driving it and its relatives towards the form that such animals now have.

Ganguroo was another small quadrupedal wallaby-like animal, sufficiently primitive that it was originally classified as a potoroo, roughly rabbit-sized relatives of the true kangaroos. These days, it more commonly seems to be placed in the kangaroo family, but it's likely not descended from the last common ancestor of the living forms, so its exact placement is arguable. Nonetheless, potoroos certainly existed at the time - probably appearing at around the same time as true kangaroos did - since one fossil species known from the Middle Miocene was so remarkably similar to the living forms that it's placed in the same genus, Bettongia, as some living species. Like them, it probably fed by digging in the forest floor for grubs, roots, and fungi.

When Ekaltadeta was first described in 1985, it too was thought to be a potoroo, or at least closer to that group than to anything else alive today. Since then it has been determined to be a relative of the living dusky rat-kangaroo, a rabbit-sized omnivore that's sufficiently odd to be given a "family" all to itself, albeit one related to the kangaroos and potoroos. Ekaltadeta, however, is notable for possessing an unusually long and sharp serrated tooth near the back of the mouth, which looks as if it would have been best suited to slicing through flesh. If so, it was presumably much more carnivorous than its living relative, although probably still omnivorous to at least some degree.

In a similar vein, Miocene Queensland was also home to a now-extinct family of kangaroo-like animals known as the balbarids. Thought to be ancestors of living kangaroos when first identified in the 1980s, it later became clear that they were actually a side-group that died out in the Middle Miocene, leaving no descendants. In 2000, however, they achieved some degree of fame when a relatively complete skull was discovered for the first time... revealing that they had large fang-like canine teeth that may well have been visible even when the mouth was closed.

Christened Balbaroo fangaroo, there was immediate speculation that this, and other known balbarids, were active predators, sometimes described as "killer kangaroos". Leaving aside the fact that they were actually about the size of a typical modern wallaby, it's notable that their other teeth seem better suited to a browsing, herbivorous, diet. So it's perhaps more likely that they used their fangs for other purposes, such as fighting for access to mates, in a case of parallel evolution with musk deer. Evidence from other parts of their skeleton suggests that, like most other "kangaroos" of the time, they still walked on all fours, and that they may have been nocturnal, or at least preferred dusk to full daylight.

Wombats also existed at the time, although we have relatively few fossils, perhaps because they were not common at the few good fossil-forming sites of the right age. The oldest known example dates from very early in the Miocene, at around the same time as the earliest known kangaroo-like fossils. Rhizophascalonus, however, lived in South Australia; analysis of its teeth suggests that it may have fed primarily on plant roots.

Rather better known are the so-called "giant wombats" or diprotodontids. These were not literally wombats, although they belong to the same general branch of the marsupial family tree. Nonetheless, they looked quite like wombats, being heavy, quadrupedal animals with a herbivorous diet. They were not so large in the Miocene as they would later become and many were not that much larger than a modern wombat. Neohelos is a typical example, and one that is known to have been especially widespread, with fossils identified from at least Queensland to South Australia. With a skull a little over 20 cm (8 inches) long, it probably wasn't much larger than a living wombat although its bodily proportions may have been different. Many different species existed, with a general trend to become larger over time; later relatives, such as Kolopsis from the Late Miocene, were at least 50% longer and presumably much heavier.

Nimbadon was a rather more unusual diprotodontid, even though it seems to be relatively close to Neohelos and its kin in the larger family tree. Known from a sufficiently large number of fossils that it's possible to trace the development of its skull as it changed from suckling in its mother's pouch to feeding on its adult diet of leaves, analysis of the rest of the skeleton did not come until 2012. When it did, it became clear that Nimbadon had powerful forelegs with highly flexible shoulders and elbows, large curved claws and a semi-opposable thumb and big toe. All of this suggests an animal that spent much of its life climbing in the trees; not something we'd associated with a real wombat. Given that the animal is estimated to have weighed around 70 kg (150 lbs) and may well have lived in large social groups, it must have been an impressive sight.

The obvious comparison, of course, is with a koala. The koala family dates back slightly further than the kangaroo family although this may well be a matter of definition; Miocene "kangaroos" were diverse enough to be placed into several families but koalas never have been, so even the earliest fossils get placed in the same group for lack of anywhere else to put them. 

Having said that, it's also true that koalas have changed relatively little in the 25 million years or so since they first appeared. The Early Miocene Priscakoala, for instance, would likely have been readily identifiable as such, being an arboreal leaf-eating animal of about the same size as the living koala. The primary feature that marks it out as one of the most primitive koalas known is the shape of its teeth, which would hardly have been obvious if one saw it up a tree; they lack the specialised grinding surfaces found on other koala teeth and suggest that it ate comparatively soft leaves and likely wouldn't have gone near eucalypts. 

Priscakoala probably represented an early branch in the koala family tree since even the few koalas that predate it (from the preceding, Oligocene, epoch) look more like the modern species than it does. However, it lived alongside at least two other genera of koala that are much closer to its modern relative. These are the "dwarf koalas" Nimiokoala and Litokoala, both of which were about half the size of the modern species. Litokoala is thought to be the closer of the two to the modern form, due in part to the fact that it had developed more of the short face that we recognise in such animals, while Nimiokoala had a longer snout. 

Both species, however, had large eyes, and may have been nocturnal. The bones around their ears are also much larger than is typical for mammals, suggesting acute hearing, a trait shared with the living species. Differences in their jaw structure suggest that they were not yet specialists on eucalypt leaves, although they would have likely found these easier to eat than Priscakoala did; the shift towards such a diet probably began in the Middle Miocene, but was not completed until much later. Being smaller, both dwarf koalas were probably more agile and active than their modern kin, although the acute hearing suggests that they may have been equally chatty.

But even kangaroos, wombats, and koalas do not represent the totality of Australian marsupials today, and this was equally true in the Miocene. Next time, I will be looking at some of the other native animals of the epoch...

---

Synapsida will be taking a break next week and will return 14th August.

[Picture by Nobu Tamura, from Wikimedia Commons.]