The taxonomic issues with Pentaceratops aquilonius

[Manatee]

Several days ago, I read a Blogger post about the new named species of Pentaceratops, Pentaceratops aquilonius named by Nick Longrich. However, its position on the phylogenetic tree of the Chasmosaurinae is quite odd:
http://waxing-paleontological.blogspot.com/2014/10/taxonomic-messes.html

[triceratops83]

I've always liked what Gregory S Paul did with lumping species. If todays' animals were fossils, palaeontologists would name our different species as different genus to get the status of having named something. Australia's two crocodiles, both are Crocodylus, but if you found fossils of them, you would not think they are the same genus. One's a monitor lizard sized piscivore, while the other's a massive apex predator. Their skulls are as different from one another as a Tyrannosaurus is to an Alioramus.

[Manatee]

I agree that either extant genera are very lumped, or dinosaurs need to be more so.

[Sim]

The naming of "Pentaceratops aquilonius" appears to have no logic behind it.  I hope this gets revised so something meaningful comes from the classification of this animal.  I also think Rhinorex is actually a species of Gryposaurus.

[Newt]

I'm not qualified to comment on this particular group, but there are good reasons for the prevalence of monotypic genera in dinosaurs.

Determining the relationships of animals known only from fossils, often incomplete and with very low sample sizes, is even more difficult than determining the relationships of extant animals. Placing a new species in an existing genus is stating a hypothesis of relationships, and if your hypothesis of relationships changes,so must the taxonomy. On the other hand, if all the species in the group are treated as monotypic genera, then the taxonomy remains stable; it's basically the same as saying, "We don't know how these animals are related to one another." You could make an argument here that the whole difficulty here is caused, not by aquilonius being placed in Pentaceratops, but by anything but the type species being placed in Pentaceratops.

These seeming minutiae don't just affect taxonomy. Workers in other areas assume taxonomy is meaningful, and use it in their own research. So, a worker might discuss the changing distribution of Pentaceratops through time in the context of palaeoclimate reconstructions or correlates with geologic vicariance events - all of which are invalid if "Pentaceratops" is non-monophyletic. Such studies can still be done with groups of monophletic genera, but they must explicitly choose a hypothesis of relationships of those taxa rather than just assuming that Pentaceratops represents a monophyletic group.

This is not to say that there is never a reason to assign species to the same genus in dinosaurs, just that it is problematic. It's also a bit nasty to assume that any paleontologist who erects a new genus is doing it for status or glory. Erecting a new genus is actually a conservative act that allows the new taxon to be viewed objectively.

I hope that's clear. It's early and I haven't had my coffee yet.

[CAWCarcharo]

Well, currently there is also a new issue regarding the validity of Pentaceratops aquilonius as it has now been possibly regarded as actually belonging to Spiclypeus shipporum.

[Sim]

I read what the paper that names Spiclypeus says about the validity of "Pentaceratops aquilonius".  I'm pleased to see it find "Pentaceratops aquilonius" a nomen dubium.  I think it proves that "P. aquilonius" wasn't even based on good enough fossil remains to warrant naming a new species.

[Soopairik]

Wow, this was an interesting read.

[CAWCarcharo]

In my opinion, I think that Pentaceratops aquilonius is either a Juvenile Female or an Adult Female Spiclypeus shipporum.

[Tyto_Theropod]

I'm not qualified to comment on this particular group, but there are good reasons for the prevalence of monotypic genera in dinosaurs.

Determining the relationships of animals known only from fossils, often incomplete and with very low sample sizes, is even more difficult than determining the relationships of extant animals. Placing a new species in an existing genus is stating a hypothesis of relationships, and if your hypothesis of relationships changes,so must the taxonomy. On the other hand, if all the species in the group are treated as monotypic genera, then the taxonomy remains stable; it's basically the same as saying, "We don't know how these animals are related to one another." You could make an argument here that the whole difficulty here is caused, not by aquilonius being placed in Pentaceratops, but by anything but the type species being placed in Pentaceratops.

These seeming minutiae don't just affect taxonomy. Workers in other areas assume taxonomy is meaningful, and use it in their own research. So, a worker might discuss the changing distribution of Pentaceratops through time in the context of palaeoclimate reconstructions or correlates with geologic vicariance events - all of which are invalid if "Pentaceratops" is non-monophyletic. Such studies can still be done with groups of monophletic genera, but they must explicitly choose a hypothesis of relationships of those taxa rather than just assuming that Pentaceratops represents a monophyletic group.

This is not to say that there is never a reason to assign species to the same genus in dinosaurs, just that it is problematic. It's also a bit nasty to assume that any paleontologist who erects a new genus is doing it for status or glory. Erecting a new genus is actually a conservative act that allows the new taxon to be viewed objectively.

I hope that's clear. It's early and I haven't had my coffee yet.

You make some very good points. I have always thought that the way taxonomy is set up makes people (including serious palaeontologists) far too inclined to assume two fossil species are close relations simply because they have the same name, when in reality they may not be. If all the experts blithely assumed that, for example, Velociraptor mongoliensis and V. osmolskae were absolutely, definitely sister taxa and then evidence turned up that proved V. osmolskae to be closer to Linheraptor than it was to V. mongoliensis, it could potentially invalidate quite a lot of research. While I honestly don't mind that V. osmolskae hasn't been given its own generic name (I'm not a professional palaeontologist, after all, so I assume that those who are know what they're doing more than me), but I personally think that it's important to remember that taxonomy is a human-created system and is therefore subjective, especially with fossil organisms.

Writing this out recalled to me that I'd read a blog post on it somewhere. It could have been from Mark Witton or Darren Naish, but I'm not sure now...

[Newt]

Thanks, Tyto_Theropod. I believe there have been some discussions about this on SVPoW as well.

It's easy to lose sight of just how meager the data for most dinosaur species is, compared to most living organisms or even many fossil taxa. When I was in college I worked in the vertebrate collections; we probably had more specimens of, say, Fowler's toads or stonerollers on one shelf than there are known specimens of any non-avian dinosaur species. A single lump of rock may have more foraminiferan fossils in it than there are specimens of all the non-avian dinosaur species put together.

It's very difficult to understand the range of variation in a taxon with only a handful of complete specimens and assorted scraps (if we're lucky) to compare, and that has serious consequences for how certain we can be about taxonomic assessments - not to mention that even a "complete" dinosaur specimen is missing a lot of data that we take for granted when looking at the taxonomy of modern animals. Just to mention one example - if you look at North American hylid frogs, some of the major distinctions between genera and species include breeding season, call structure, testis pigmentation, skin texture and pattern, and chromosome count, as well as biochemical (enzyme and DNA) data - all of which are virtually or actually unknowable for Mesozoic animals.