The Mystery of the "Shovel-Tuskers"

Whenever I visit the American Museum of Natural History in New York I make sure to at least pass through the fourth-floor fossil halls before I leave, and one of my favorite displays features the shovel-mouthed proboscidean Platybelodon. In a glass case in the shadow of a mammoth skeleton is a growth series showing the development of the Platybelodon jaw, from juvenile to adult. Like many AMNH displays, however, this series was not a product of the renovation of the fossil halls in the 1990's but came out of much earlier research.

In 1932 Henry Fairfield Osborn and Walter Granger published a paper entitled "Platybelodon grangeri, Three Growth Stages and a New Serridentine From Mongolia." It would provide the basis for the current museum display, but its roots actually stretch back a few more years. In 1927 the Russian paleontologist A.A. Borissiak described the Miocene proboscidean Platybelodon from bones found in the North Caucasus region, but Granger's fossils (found in 1928) came from Mongolia and were thought to represent a new species (Platybelodon grangeri). The most distinctive characteristics of this genus were its broad, scoop-shaped lower jaw and large, flat lower tusks (incisors).

At first Osborn thought that Granger had discovered a new species of the "shovel tusked" proboscidean Amebeldon which had recently been found in Miocene-aged deposits Nebraska and described by E.H. Barbour. Clearly the strange shape of the lower jaws and tusks of these animals must have been related to some particular mode of feeding, and Osborn speculated that the extinct elephant-relative Gragner exhumed was "Adapted to uprooting bulbous plants, [and] it frequented the shallow lake waters of ancient Gobi."

By 1931 at the latest, however, Osborn changed his mind. The shovel-tusker from Nebraska and the shovel-mouthed proboscidean from Mongolia were different, with Granger's find more closely resembling Borissiak's Platybelodon. Amebelodon had a long, thin jaw with extended, scoop-shaped incisors. Platybelodon, by contrast, had a shorter but broader jaw with short, square incisors. Clearly these animals belonged to different genera.

Granger's 1928 find was complemented by the discovery of an even richer cache of Platybelodon fossils when he returned to Mongolia in 1930. (The Platybelodon fossils were so common, in fact, that Osborn thought it to be diagnostic of the Tung Gur Formation in which it was found.) Interestingly, these sites may have represented some kind of mass death assemblage. Quarry 1 yielded at least 16 individuals, most of which were adults, and Quarry 2 contained one adult, about eight juveniles, and one specimen found between the hips of an adult that Osborn considered a fetal Platybelodon. Granger had hit the Platybelodon jackpot.*

*[I assume that all the digging Granger and his crew did disturbed things quite a bit, but it would have been nice to have seen a taphonomic study done at each of the quarries. Did each bonebed represent an accumulation over a long period of time? Was there some catastrophic event? Was it a death assemblage caused by drought? I do not know.]

This large sample allowed Osborn and Granger to broadly outline the growth pattern of Platybelodon grangeri. The specimen identified as belonging to a fetus was relatively short with no real "scoop" at the front of the lower jaw (even if the sides of the lower jaw did "pinch in" a little to indicate where this scoop would eventually develop). This arrangement contrasted strongly with that of a juvenile specimen, which had a more strongly Y-shaped lower jaw, a more defined scoop, and large lower incisors.

The lower jaw was even further modified in the adult. Adult Platybelodon grangeri had mouth scoops as wide as the back of the jaw, but the jaw pinched in strongly before the part of the jaw containing the premolars and molars. Looked at from the side the lower jaw also angled down to truly form a concave scoop that differed from the tooth-shovel seen in Amebelodon. If stood up on the front incisors the lower jaws of an adult Platybelodon grangeri would have looked like a shovel with two toothy handles.

What was the function of such an arrangement? As he stated earlier (and following the work of Barbour and Borissiak) Osborn suspected that Platybelodon was digging into the dirt or mud for plants. This appeared to be confirmed by the beveled edges on the lower incisors in the specimens Granger had collected. Osborn wrote;

It seems probable that this bevel was produced by abrasion against a smooth rock or against the bottom of a shallow pond or stream in the process of scooping up vegetation, a process in which the broad short trunk was used in connection with the mandibular scoop.

It had been thought that Amebelodon was doing something similar with its tooth-shovel, too, but subsequent research carried out by David Lambert in the 1990's suggested that Amebelodon used its teeth in a variety of ways. Based upon wear patterns on its teeth it seems like Amebelodon not only used its teeth/lower jaw to shovel, but also to scrape bark from trees and otherwise gather food from different sources. Rather than being a specialist its distinctive jaws and teeth allowed it to feed in a variety of ways on different plant foods.

Neither did Platybelodon use its shovel-shaped jaws in the way Osborn envisioned. The tooth wear patterns Lambert observed were inconsistent with what would be expected if Platybelodon specialized in digging into the mud for plants with its lower jaw. How, then, was it feeding? Lambert was not sure, but he proposed an interesting alternate explanation.

Lambert hypothesized that Platybelodon used its unique lower jaw as a scythe, plucking up vegetation with its trunk and then rubbing it against its teeth to saw through it. If Platybelodon did engage in this feeding style it could account for the crescent-shaped indentations in its front, lower incisors as well as the sharpness of those teeth. Just because the lower-jaw of Platybelodon looked shovel-like does not mean it used it like a shovel.

Could Platybelodon have scooped up water plants in its jaws? They probably could have, but it does not appear that they did so with any regularity. Neither does Lambert's plant-processing hypothesis close the book on how Platybelodon used its lower jaws. Lambert's hypothesis was developed after others had failed and perhaps there is some alternate explanation that might also be consistent with the fossil evidence. Either way, it seems that Platybelodon or Amebelodon were not the sifters of swamp-muck they have traditionally been portrayed as.