Home-grown Shovel Tuskers?

Konobeladon

Elephants are unique and remarkable animals, looking quite unlike any other creature. They have no close living relatives among other mammals - you have to go back almost to the time of the dinosaurs to find any common ancestor with anything else. As a result, the elephant family is placed within its own order of mammals, a ranking equivalent to that given to such groups as "primates", "rodents", or "bats". With just one family, and only three living species, it isn't quite the smallest mammalian order, but it's very close.

This changes significantly if we choose to include the known extinct species. There are a great many of these, which tend to be large, heavily built creatures with elongated tusks, and, in most cases, features on their skulls that suggest they had a trunk. Indeed, this latter is the source of the official name of the order, the proboscideans. While only the elephant family survives today, at least six others are recognised to have existed in the past, and if we could see members of most of them today they'd be instantly recognisable as, if not actually elephants, at least "elephant-like".

The elephant family itself, which includes the mammoths, is not especially old, dating from around 10 million years ago, towards the end of the Miocene epoch. They are generally believed to have descended (together with a second family that died out early at the beginning of the Last Ice Age) from a group of animals called the "tetralophodont gomphotheres". 

The word "gomphothere" translates as "wedge-beast" and refers to the greatly elongated, wedge-like, lower jaw of many gomphothere species, which was typically equipped with a second set of tusks. This was probably used to cut vegetation, with the greatly elongated trunk seen in modern elephants evolving later as an alternative means of obtaining food.

The exact classification of the gomphotheres is not yet settled. The seven-family scheme for the proboscideans that I mentioned earlier lists them as just one of those families but this is likely an artificial grouping consisting of animals that are not all that closely related to one another, but just happen to look similar. How, and to what extent, we should divide them up is where the uncertainty comes in.

Among the most bizarre of the gomphotheres are the shovel-tuskers. These are not the direct ancestors of living elephants, but a side-branch that prospered during the Miocene but that died out at the end of the epoch around 5 million years ago. These days they are commonly split off from the other gomphotheres as their own family, the Amebelodontidae. (Which, if you're keeping count, brings us to eight families of proboscidean in total, and many modern schemes would recognise even more).

Shovel-tuskers were, to be honest, weird-looking. Granted, if elephants didn't exist today, we'd probably regard those as something that looked like they came out of a fantasy novel but, to modern eyes, shovel-tuskers were even stranger. Like most other gomphotheres, they had that greatly elongated lower jaw and a trunk that was likely shorter than that in living elephants. 

They also, in addition to the regular tusks in the upper jaw, had a second set on the lower one. But these lower tusks were wide and flattened, together forming what looks like a giant spatula. In the early 20th century, when the first fossils of shovel-tuskers were discovered, it was thought that this odd structure was used to scoop up vegetation from shallow rivers, hence the common name. More recent analysis has shown that the tusks were not literally used as shovels but were instead used for things such as scraping the bark off trees. Many older reconstructions of the animals also show them with flap-like flattened trunks that lay on top of the "shovel" but a 2016 analysis concluded they were probably more elephant-like than that.

The first shovel-tusker genus to be scientifically named was the one for which the family is named: Amebelodon, in 1927. This lived for around 4 million years at the end of the Miocene in North America, making it one of the last of the shovel-tuskers to die out. Various species have been assigned to the genus over the decades since, ranging from the relatively primitive A. floridanus to more advanced forms with flatter lower tusks and a more well-developed "shovel". 

In 1990, some of the more advanced forms were distinguished by being given their own subgenus. These forms were more successful than their predecessors, at least in geographic terms, with fossils having been discovered as far afield as Libya. The picture seemed reasonably clear: Amebelodon first evolved in North America in the form of A. floridanus, which evolved into the moderately advanced A. fricki, and then into the various species of the newly named subgenus some of which rapidly left America and headed to Africa via an Asian route.

In 2014, however, this simple picture became more complicated. In that year, a new species of the advanced subgenus was discovered in Greece that was as old as any in North America, calling into question in which direction the animal had migrated. This was considered significant enough to promote the subgenus status, and these species, including the late-surviving African one, now get the new name Konobelodon. To make matters worse, however, there is now evidence of what appears to be one of the intermediate forms from China that's actually earlier than the equivalent in North America.

This makes it hard to believe that these animals first appeared on the continent where they were initially discovered. The new picture is therefore that the intermediate form, A. fricki, or some as-yet-unnamed species closely related to it, evolved in China, and evolved into Konobelodon locally before the latter headed into North America - the opposite direction than originally assumed - as well as into Europe and Africa.

This naturally leaves open the question of what to do with the more primitive forms of Amebelodon, which now seem less likely to be the ancestors of A. fricki, and a recent paper suggests moving them to another new genus, Stenobelodon. Once this is considered separately, they argue that it has little obvious relationship to any of the other known shovel-tuskers from North America and, apart from the presence of the partially flattened lower tusks, resembles regular gomphotheres more than it does anything else. In which case, it could be a case of parallel evolution, having evolved the "shovel" independently by modifying the rounded and pointed lower tusks of a common ancestor.

If that's true - and it's still highly speculative at this point - the entire family of the shovel-tuskers could need revision. It may look weird, but it could be that the shovel-tusker body form was useful enough that it evolved twice.

[Picture by "DiBgd", from Wikimedia Commons.]