The largest land animal ever to have lived was surely a dinosaur, most likely a member of the aptly named titanosaur group. But what about the largest land mammal? It's hard to know for sure, partly because "largest" is a vague term that could be interpreted in different ways - do we mean weight or linear dimensions? But it's also just difficult to know when all you have to go on is the skeleton, and that's probably incomplete.
As a result of this uncertainty, there are at least two contenders for the crown, one of which is the Asian straight-tusked elephant Palaeoloxodon, a very close relative of living elephants that died out around 56,000 years ago, during the early part of the Last Ice Age. The other, which was taller, but perhaps not quite as heavy (although the difference may not have been great) was the giant rhino Paraceratherium.
Of course, Paraceratherium, like Palaeoloxodon, wasn't really a species, but instead a genus, or collection of very closely related species. It's partly down to this that the scientific name of the animal has a rather confused history. For much of the 20th century, prior to the reevaluation of the calculated size of the straight-tusked elephant, most popular books stated that the largest land mammal to have lived was a creature named Baluchitherium. The fragmentary fossil used to name this animal had, in fact, been identified as early as 1936 as likely belonging to the same species as the previously named Paraceratherium bugtiense (which, as it happened, had been discovered by the same palaeontologist), something that's now widely regarded as correct.
For a while, the honour of largest land mammal switched to a different giant rhino, Indricotherium based on the skull of an animal that seemed closely related to Paraceratherium, but slightly larger, and with enough distinctions in its shape to warrant a different name. It's now thought that it was, yet again, the same species and that the differences may be down to the larger skull belonging to a male, and the other to a female. Even today, the group of which they are a part is sometimes referred to as the "indricotheres", rather than the more logical (but harder to pronounce) "paraceratheriines".
But, regardless of what we call them, or even whether they're really larger than the straight-tusked elephant, what were these animals?
The giant rhinos lived during the Oligocene epoch between around 34 and 23 million years ago. This is almost half-way back to the time of the dinosaurs, and long before elephants as we know them today appeared on the scene. Most of their fossils are known from central Asia, especially in and around Pakistan, but they reached as far as China and Mongolia in the east and into Kazakhstan in the north. A number of very fragmentary remains are known from eastern Europe and Turkey, but it's hard to say much about them, and they couldn't reach western Europe because there was a sea in the way at the time (connecting, very approximately, the Baltic and the Adriatic).
These giant animals were taller than giraffes but much more heavily built, with long powerful legs and, proportionately speaking, a shorter neck. It's possible that they evolved their huge size for much the same reason that giraffes probably evolved their distinctive shape - allowing them to browse from treetops. Judging from the shape of their skulls, they probably had a short tapir-like snout. It's been proposed that they might also have had elephant-like ears, but obviously those wouldn't fossilise, and it's fair to say that this isn't widely accepted.
Technically, they weren't rhinoceroses, in the sense of belonging to the modern rhino family, although this did already exist at the time in the form of some early hornless rhinos in North America. They were, however, more closely related to rhinoceroses than to anything else alive today and are merely part of the larger "indricothere" group, of which the oldest-known definitive member was the horse-sized Juxia living around 40 million years ago.
A number of species have been described, even ignoring those that have ended up being synonymised with those previously described. A couple of weeks ago, a new one was added to the mix, and it sheds some new light on the evolution and diversification of the group. This has been named P. linxiaense, named for the Chinese city of Linxia just north of where the fossil was discovered. This is a rugged, almost mountainous part of the country and, notably, is on the northeastern edge of the Tibetan Plateau.
The fossil consists of a complete skull and jaw, as well as a few bits of the backbone. This obviously isn't enough to tell us, for example, how tall the animal might have been, although the fact that the skull alone was over a metre (3 feet) in length suggests that it was in the same general size range as P. bugtiense, the best-known species, and for which we have a relatively complete skeleton.
Nonetheless, the skull is often the most diagnostic part of any mammal skeleton, and this one is in good condition, considering its 26.5 million-year-old age, sufficient to allow us not only to be confident that it is, in fact, a new species, but to see how it fits in the wider history of giant rhinos.
It seems to be one of the larger members of the genus (although, obviously, they're all pretty big) and to be relatively specialised, which makes sense given that it's also one of the more recent. In particular, the discoverers note that the shape of the atlas - the vertebra immediately underneath the skull - implies a more flexible, and presumably longer, neck than seen in other species. A detailed analysis of this and other features indicates that the new species was more closely related to P. bugtiense, from Pakistan, than it was to species known from closer, but older, sites in China.
The analysis concludes that giant rhinos first evolved somewhere around Mongolia, where the oldest known fossils have been found, and then migrated south and westward, reaching China and Kazakhstan before eventually crossing all the way over to what is now Pakistan. It is there that it reached its greatest size, with some of these exceptionally large individuals later heading back north to return to China and give rise to the newly named species (among others).
Today, this would not be an easy route for a herbivore that presumably required a very large amount of foliage to keep itself going. The eastward route, through Indochina, would be lengthy, and we don't know of any relevant fossils from that area - although that doesn't necessarily rule it out, so far as I can see. On the other hand, the westward route, cutting through Kazakhstan and perhaps giving rise to the European and Middle Eastern species on the way, won't work at all. Once again, this is because there was a sea in the way during the Oligocene, a vast eastward expansion of today's Black Sea that terminated in a narrow gulf reaching all the way across to somewhere around Nepal.
Yes, Pakistan had a north coast at the time.
It was there because the Indian subcontinent hadn't quite finished colliding with Asia and raising the Himalayas and the Tibetan Plateau more generally. Those mountains did already exist, but they were lower than today, perhaps with wide passes between them. For that matter, fossil imprints of the leaves of subtropical palm trees have been discovered from central Tibet (to be precise: here) that date from 25.5 million years ago - right in the relevant time period. Which rather implies that the elevation was considerably lower then than it is today.
This means that they didn't need to go through Myanmar and Vietnam and that they could take the much more direct route by crossing Tibet. It wasn't so difficult when the Plateau hadn't yet fully formed...
[Photo by Ryan Somma, from Wikimedia Commons.]