What’s Sexy to a Dinosaur?

Can paleontologists identify the influence of sexual selection in the fossil record?

A sculpture of Pentaceratops outside the New Mexico Museum of Natural History and Science. Could sexual selection account for the prominent ornaments of this dinosaur? Photo by the author.

Non-avian dinosaurs were weird. That’s one of the reasons we love them so much. There’s nothing quite like a slender-necked Barosaurus, a beautifully-crested Dilophosaurus or lavishly-ornamented Pentaceratops alive today. If such dinosaurs were anything, they were bizarre, but why were they so strange? Each case demands its own explanation, and paleontologists have continuously tussled over whether particular ornaments were weapons, sexual displays or something else.

According to an in-press paper at Trends in Ecology & Evolution, at least some weird dinosaur features may best be understood in the context of mate competition, mate choice and sexual signalling. The paper, by entomologist Robert Knell and colleagues, is the latest in a long-running debate over whether sexual selection had any influence on dinosaur lives and how to detect the hallmark of such pressures.

The debate has been going on for years but only recently increased in intensity. In a 2010 study, paleontologists Kevin Padian and Jack Horner rightly noted that sexual dimorphism–or a significant anatomical difference between the sexes–has never been conclusively demonstrated among non-avian dinosaurs. The idea had been proposed for a variety of dinosaurs using a number of skeletal landmarks, but none of the hypotheses have stuck. Even if sexual dimorphism existed among dinosaurs, we lack the sample size to identify the phenomenon. More than that, Padian and Horner cited the lack of sexual dimorphism as a sign that sexual selection probably wasn’t an important facet in the origin and modification of bizarre dinosaur features. Instead, the researchers hypothesized, the various horns, crests, plates and other ornaments evolved because of species recognition–the ability for dinosaurs to quickly and easily identify members of their own species.

Other researchers disagreed. Knell and Scott Sampson had a brief exchange in the journal pages with Padian and Horner. This was followed by a paper by Dave Hone and co-authors that suggested that mutual sexual selection might explain the mystery of why dinosaurs had bizarre ornaments but don’t seem to exhibit sexual dimorphism. Under this hypothesis, both males and females may prefer mates with elaborate visual signals, and therefore the same prominent structures would be expressed in both sexes. This kind of sexual selection has been documented in modern avian dinosaurs, but, until now, hasn’t been considered as an explanation for the ornamentation of non-avian dinosaurs. Even though mutual sexual selection has not been proven as an evolutionary driver among extinct dinosaurs, it’s a possibility worth considering.

The new paper by Knell and co-authors also draws on modern examples to investigate how we might identify examples of sexual selection among prehistoric species. The paper covers a wide variety of creatures, from ammonites to birds, but, since this is the “Dinosaur Tracking” blog, I’ll focus on how the argument applies to the ever-controversial adornments of non-avian dinosaurs.

As the researchers state, there’s no simple, tell-tale way to identify sexual selection. This is partly because many strange structures are multifunctional, and structures may be co-opted for different functions during the course of their evolution. Think of sauropods. The elongated necks of these dinosaurs allowed them to feed over a wide swath of greenery, but they could have also been used as visual displays. A big fleshy neck is prime advertising space. In this case, a feeding advantage appears to have preceded any signalling function, but the mosaic nature of evolution hinders our efforts to tease apart the influence of different, interacting pressures.

All the same, there are a few clues that can help paleontologists identify possible cases where sexual selection was at play in the deep past. One possible line of investigation is sexual dimorphism, although, as I said above, this has yet to be conclusively demonstrated in dinosaurs. (And, as Knell and co-authors argue, sometimes the sexes might differ for reasons other than sexual selection.) The way prominent displays grew is another phenomenon worth looking into. We would expect that features that make a difference in mating would only appear as the dinosaur approached sexual maturity. Juvenile, and presumably sexually-immature, Lambeosaurus don’t have the full-blown crests of older individuals. Perhaps this is because the crests are sexual signals that only grow as the dinosaurs approach mating age, although it’s possible that the development of crests are related to the overall growth of the dinosaur’s skeleton.

The diversity–or disparity–of ornament shapes among closely-related species may also be important. Even closely-related species of ceratopsid dinosaurs, Knell and collaborators note, had very different horn shapes and arrangements. This could be a sign of sexual selection by way of competition and mate choice, but, as Padian and Horner pointed out, the same evolutionary pattern could be the result of selection for distinct-looking species. Finally, Knell and co-authors cite “costliness” as another potential indicator–if a trait is flashy, requires a good deal of energy to grow and comes at a cost to the organism’s survival potential, then it may be a sexually-selected trait.

Obviously, each line of evidence comes with caveats. Sexual selection can be difficult to identify even among living species, much less extinct ones. It would be strange if sexual selection played no role in dinosaur evolution, but we’re left with the question of how to detect and test the hypothesis of sexual selection. Paleontologists will have to very carefully test hypotheses about bizarre structures, paying careful attention to distinguish between competing alternatives. Ultimately, paleontologists may only be able to identify possible scenarios for the origin and evolution of bizarre features, but studies of modern species can at least provide guidelines for what researchers should look out for.

If we’re truly going to understand the visual signals of dinosaurs, though, we need better sample sizes. We need to know how individuals of the same species varied from one life stage to the next. Without this anatomical foundation, researchers will be left to argue from a typological standpoint that may misconstrue how certain features changed with age and evolved over time. Recall the “Toroceratops” debate–if Triceratops changed into a Torosaurus-form late in life, most likely beyond the onset of sexual maturity, that is certainly going to influence how paleontologists investigate and discuss dinosaur visual signals.

The influence of sexual selection, or lack thereof, will undoubtedly be debated for some time to come. But, as Knell and colleagues conclude, investigating the possible influence of sexual selection in prehistory “is neither a forlorn nor impossible task.” We may yet find out what’s sexy to a dinosaur.

For more on this study, see this post by Dave Hone, one of the paper’s authors.


Knell, R., Naish, D., Tomkins, J., Hone, D. (2012) Sexual selection in prehistoric animals: detection and implications, Trends in Ecology & Evolution DOI: 10.1016/j.tree.2012.07.015.


Get the latest Science stories in your inbox.