Fossil Icons of Male-Female Difference May Not Be So Different After All

For more than a century, the giant horned mammals known as uintatheres have stood as one of the most celebrated textbook examples of sexual dimorphism in the fossil record — the phenomenon in which males and females of the same species differ strikingly in body size and ornamentation. Generations of paleontologists pointed to these Eocene giants, with their imposing skulls, bony protuberances and formidable canine teeth, as evidence that males of ancient species were dramatically larger and more heavily armed than their female counterparts. Now, a new study from the University of Kansas, published in the journal PLOS One, argues that this long-standing assumption may rest on far shakier statistical ground than anyone realized. When the skulls of one extinct species, Uintatherium anceps, were measured and analyzed with modern methods, the supposed two-sex split all but dissolved into a continuum of variation.

Uintatheres emerged roughly 58 million years ago, fewer than 10 million years after the asteroid impact that ended the age of dinosaurs, and they persisted in the fossil record for around 20 million years. They were among the largest land mammals of their time, notable for their bulk, their prominent horns and their large canine teeth. Because living relatives of such animals are long extinct, paleontologists have traditionally inferred differences between the sexes from clusters of size and ornament in fossil samples. If a collection of skulls seemed to sort into a big-horned group and a small-horned group, the natural interpretation was males and females. For uintatheres, that interpretation hardened into orthodoxy: males were depicted as much larger animals bearing far more dramatic cranial features than females, a picture repeated in museum halls, popular books and scientific literature alike.

The new research began, fittingly, with a puzzle far smaller than a meter-long skull. Kevin Mulcahy, a doctoral student at the University of Kansas Biodiversity Institute and Natural History Museum, and his doctoral adviser K. Christopher Beard, senior curator at the institute, were trying to identify a single broken upper molar collected during fieldwork in Montana, a tooth they suspected might belong to the uintathere group. To solve that problem, the two traveled to the American Museum of Natural History in New York, whose storied vertebrate paleontology collections hold entire shelves of uintathere skulls, each nearly a meter long. These are famous specimens, described and redescribed in the literature for decades, and the original goal was simply to compare teeth and pin down the identity of the Montana fossil.

What Mulcahy found instead was a pattern that did not match the textbooks. As he examined the celebrated skulls to compare their teeth, he noticed an enormous amount of variation — but it did not appear to fall into two neat categories. Working quickly, he took a series of measurements of the skulls with a ruler and began exploring them statistically. The logic of the test is straightforward: if a sample of fossils truly reflects a strongly dimorphic species, measurements of traits such as skull size, horn development and canine dimensions should fit a bimodal distribution, with two peaks corresponding to average males and average females. If the species is not strongly dimorphic, the same measurements should instead follow a single, unimodal distribution. When he ran the analysis, the expected two-peak pattern was nowhere to be found.

“It wasn’t like there’s one group of specimens that look bigger than another, and then there’s another group that are smaller with smaller horns and canines,” Mulcahy said. That observation raised an uncomfortable question: were uintatheres really such good examples of sexual dimorphism after all? Was there actual evidence in the morphology of the skulls — in the distribution of the ranges of morphology — that two sexes were being sampled? To find out, Mulcahy broadened the project, gathering additional measurements from specimens held at other institutions and applying newer statistical procedures. His final dataset included about 30 skulls of Uintatherium anceps, the species at the heart of the dimorphism story.

The results were underwhelming for the traditional view. “Maybe there were some little signals of dimorphism here and there, but nothing really concrete,” Mulcahy said. Yet a skeptic could still object that the methods themselves were too weak to detect dimorphism if it existed. Fossil samples carry a fundamental handicap: no one actually knows which specimens are male and which are female. Researchers can guess, assigning the most extreme individuals to one sex and the least extreme to the other, and then test whether the groups differ — but as Mulcahy points out, that procedure is deeply circular. Of course the groups will differ if you built them by sorting extremes in the first place.

To break that circularity, Mulcahy turned the same statistical techniques on a living animal whose sex could be verified with certainty: the modern bison. Bison offered an ideal comparison. They are another very large herbivore, they occupy an ecology broadly similar to what has been proposed for uintatheres, and they carry substantial horns, making them a plausible modern analogue for testing how dimorphism signals appear in skull measurements. The contrast with the fossils was striking. In the bison sample, the dimorphism signal was clear and strong: females clustered with females and males clustered with males in an almost perfect separation. The methods, in other words, were fully capable of detecting sexual dimorphism when it genuinely exists — which made its near absence in the uintathere skulls all the more meaningful.

If the statistics were sound and the sample was adequate, then the flagship example of sexual dimorphism in the mammalian fossil record may never have deserved the title. Mulcahy suggests that the mistaken assumption has deep historical roots, reaching back to the late 19th and early 20th centuries, when so many fossil species were first described. This was the era in which evolutionary theory, sexual selection and natural selection were entering scientific discourse, and the field was dominated by male paleontologists working within a distinctly Victorian-era mindset. Within that intellectual climate, the idea that males would be bigger and more fearsome than females was less a hypothesis to be tested than an assumption to be illustrated. “When you step back and actually test that hypothesis, it doesn’t really hold up,” Mulcahy said.

The implications extend well beyond uintatheres. If a species long regarded as the classic, near-definitive case of dimorphism yields only weak and inconclusive signals under modern analysis, then other fossil species whose dimorphism was asserted on similar grounds may deserve re-examination as well. The finding also speaks to a broader debate about the evolution of size differences between the sexes across the mammalian family tree. Strong, big-male-style dimorphism was probably not ancestral at the root of the mammalian tree, Mulcahy noted; among living mammals where the pattern does appear, it has almost certainly evolved independently several times. The assumption that males are typically larger than females, he argued, reflects a long-standing bias in which animals scientists chose to study and which questions they found interesting.

That bias has been quantified in recent work on living mammals. As Mulcahy points out, a paper published just a couple of years ago showed that among modern mammal species, most either show no sexual dimorphism at all or actually feature larger females. The popular image of the animal kingdom as a place where males invariably outweigh and out-arm females, he said, is really an artifact of which creatures biologists in the past chose to look at. For paleontology, the lesson of the uintatheres is a methodological one: claims of dimorphism in fossil species need to be tested against explicit statistical expectations, ideally benchmarked against living species of known sex, rather than assumed from Victorian intuitions. A broken molar from Montana set that reassessment in motion, and the iconic horned giants of the Eocene may never look quite the same again.

Subject of Research: Sexual dimorphism in the extinct Eocene mammal Uintatherium anceps

Article Title: Study: ‘Classic example’ of sexual dimorphism in fossil record actually shows nuance between males and females

Article References: Study: ‘Classic example’ of sexual dimorphism in fossil record actually shows nuance between males and females. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: uintatheres, sexual dimorphism, paleontology, Eocene mammals, Uintatherium anceps, fossil skulls, statistical analysis, bison comparison, sexual selection, University of Kansas, PLOS One, mammalian evolution