Wednesday, October 6, 2010

"The Princeton Field Guide to Dinosaurs" - The Review

One of the perks to working at a bookstore is being able to borrow books, and today I checked out Gregory S. Paul's newest work.  My comments before have been based only on the snippets available on Google Books, which isn't exactly fair.  I'm only going to concentrate on the theropod section for examples, since that's what I'm most familiar with.

When people think 'Gregory S. Paul', they think excellent skeletals and restorations, and this book's overflowing with them.  My theropod favorites include Limusaurus, Acrocanthosaurus, Sinornithomimus, NGMC 2124, Sinocalliopteryx, Buitreraptor, Jinfengopteryx, Mei, the Two Medicine Troodon, Sapeornis and Protarchaeopteryx.  One thing I like is how many are rigorous in showing only the known material.  The unfortunate counterpoint is that that leaves one thinking the reconstructions which are complete indicate complete skeletons are known, but this is often not the case.  I also liked the PDW way of illustrating skulls better, with thinner lines and texturing.  The life restorations are spectacular as always.  I love the Dromiceiomimus scraping eggs and the Sinosauropteryx leaping for a Confuciusornis as two modern updates to classic themes.

The other thing people think of when they hear 'Gregory S. Paul' is taxonomic lumping. ;)  As I described in this post, he lumps a LOT of taxa together.  I think this will be confusing to most readers, who won't realize a lot of these new combinations aren't found in the literature.  While sometimes Paul notes what he changes in the taxonomy, he almost never states when something is his unique idea, or followed by other researchers.  Interestingly on page 42 he states "The phylogeny and taxonomy offered here are not a formal proposal."  This would seem to qualify for ICZN Article 8.2- A work that contains a statement to the effect that it is not issued for public and permanent scientific record, or for purposes of zoological nomenclature, is not published within the meaning of the Code.  So perhaps Paul's new combinations are all nomina nuda.

Interestingly, Paul may lump genera and many species, but he also features a wide range of unnamed species.  These include Padian's (1986) Coelophysis, the supposed Morrison Elaphrosaurus, Portuguese Ceratosaurus, problematic long-snouted Allosaurus, Dinosaur Park and Two Medicine Daspletosaurus, Dinosaur Park Struthiomimus, NGMC 2124, and Triebold and Horseshoe Canyon caenagnathids.  While some of these have been proposed in the literature, others haven't and seem to be largely stratigraphy-based.  I think the space would have been better spent including additional named taxa, such as Halticosaurus, Genyodectes, Betasuchus, Velocisaurus, Chuandongocoelurus, Erectopus, Metriacanthosaurus, Becklespinax, Archaeornithoides, Variraptor, Borogovia, etc.. 

The first 63 pages consist of standard introductory material- anatomy, behavior, trackways, history, extinction, etc..  It's largely what one finds in many dinosaur books, but is more accurate than any others I can recall.  One unique section is "Dinosaur Safari", which playfully imagines how human time travelers might deal with the Mesozoic.  I do think the book could have benefited from an additional editor besides Kirkland, as there are occasional typos (the Landian stage of the Middle Triassic, Pycnoneosaurus, Ricardoesteria, protoarchaeopterygids, etc.).  Also the accuracy of the temporal distribution of different groups on the timescale of pages 64-65 is subpar, with most maniraptoriform clades appearing too late for instance.

The main portion of the book is of course the taxonomic section.  I stand by my previous complaint that this is not effective as a field guide, since nearly all the taxa lack any useful description or indication of diagnostic features.  It's more like Sattler's 1983 "The Illustrated Dinosaur Dictionary" or Lambert's 1990 "The Dinosaur Data Book", with less information on each taxon than the former, but much more than the latter.  It's also more technical than either in including species, stages, formations and more anatomical terms.  Yet it's also less complete than either, and while obviously you can't reconstruct Ornithodesmus or Bradycneme, I don't think I'm alone among dinosaur enthusiasts in wanting my books to include every taxon.

The descriptions that do exist are often so vague as to be pointless.  For instance, coelurosaurs are described as follows- "Highly variable. Tail long to very short. Arm from longer than leg to severely reduced. Leg extremely gracile to robust, toes four to three."  Well... I guess some amphiumas and sloths are excluded at least. ;)  That doesn't tell you anything about what distinguishes the group.  The individual species descriptions suffer the same problem, when they mention any features at all.  And a disturbingly large number of taxa are merely said to be standard for their group, including rather distinctive ones such as Herrerasaurus, Segisaurus, Noasaurus, Torvosaurus, Piatnitzkysaurus, Eustreptospondylus, Gasosaurus, Siamotyannus, sinraptorids, Carcharodontosaurus, Giganotosaurus, Mapusaurus, Mononykus, Mahakala, Sinornithoides, Nanshiungosaurus and Erliansaurus.  Others are said to have insufficient information, but it's inconsistantly applied.  While Duriavenator and Ilokelesia get a description, Elaphrosaurus, Elmisaurus and Neimongosaurus don't.

The theropod groups (Theropoda, Tetanurae, Coelurosauria, Maniraptora, etc.) are never implied to include pygostylians in the introductory sections, so reading that e.g. maniraptorans only lasted to "the end of the dinosaur era" is rather odd.  All major theropod taxa have the following in their notes- "absence from Antarctica probably reflects lack of sufficient sampling."  Not only is it repetitive, it's unecessary since we have both Molnar et al.'s (1996) basal tetanurine tibia and Case et al.'s (2007) supposed dromaeosaur from the continent.  The habits section gets particularily speculative.  For instance, herrerasaurs are said to be pursuit predators.  Really?  Of the included taxa, Alwalkeria and Chindesaurus are far too poorly known, Eoraptor seems omnivorous and Herrerasaurus is massively built with shorter tibiae than femora.  I suppose Staurikosaurus may have been, but the topic has never been studied to my knowledge.  One excellent practice is taxa known only from young specimens aren't given a size and are instead marked "Adult size not certain."

"The Princeton Field Guide to Dinosaurs" is thus extremely accurate in general and full of amazing artwork, but isn't very useful for learning about individual dinosaur species or groups.  It also contains a lot of information which is not based on the literature and often contradicts the consensus, which most readers won't know.  I'd recommend it for young readers who aren't ready for Glut's encyclopedias or The Dinosauria yet.

Next up, theropod criticism and commentary.

Paul, 2010. The Princeton Field Guide to Dinosaurs. Princeton University Press. 320 pp.


  1. So perhaps Paul's new combinations are all nomina nuda.

    Erk. Mickey. You know better than this. They're not nomina nuda because that term is meaningless in this context. As far as the ICZN is concerned, availability only applies to species and genera names separately, not to combinations. Combinations are only regulated in terms of spelling and avoiding homonymy, but, once a species name is published, you can move it into any damn genus you like without producing a "new name".

    In practice, it doesn't make any real difference, of course, though it does bug me a little that zoologists have a history of being a bit sloppy when it comes to producing new combinations without necessarily specifying what the animal was called earlier.

  2. "I'd recommend it for young readers who aren't ready for Glut's encyclopedias or The Dinosauria yet."

    For the same readers, don't forget Holtz and Rey's fabulous (and ridiculously cheap) dinosaur encyclopaedia.

  3. Hmm, that's an interesting point. I always assumed the ICZN's availability constraints counted for combinations too, so that if I stupidly sink Velociraptor into Tyrannosaurus on my website as T. mongoliensis, it wouldn't stop someone else from naming a new species of Tyrannosaurus T. mongoliensis. "Nomen nudum" is probably the wrong term for such a thing, but surely there are such things as unofficial combinations.

    What about when Paul lists something like "Appalachiosaurus (or Albertosaurus) montgomeriensis" for a taxon that has only been called Appalachiosaurus montgomeriensis in the literature? Is Albertosaurus montogomeriensis a valid combination which Paul should be cited for? Or does the ambiguity count against it?

  4. Overall, I agree. As a fairly-well educated layman, I was really impressed by the pictures (the only real flaw I saw was the "bunny hands" on the one painting), but really let down by the text - I thought it would be more along the lines of PDW. I got bored quite quickly reading the text, and just started leafing through to look at the pictures.

    I think the best format would have taken something from Holtz's recent book (each of his "groupings" (can't say clade for obvious reasons) could have had its own sub-chapter, with a table listing basic information like size, fossil formation, particularly distinctive characteristics, etc. And then include a page or two of information about the "group," along with reconstructions. This way repetition would be cut way down, and there would be more room for exposition about the various taxa - which is what a dino-enthusiast probably wants anyway.

  5. I always assumed the ICZN's availability constraints counted for combinations too, so that if I stupidly sink Velociraptor into Tyrannosaurus on my website as T. mongoliensis, it wouldn't stop someone else from naming a new species of Tyrannosaurus T. mongoliensis.

    Gets a bit complicated here :-). In that case, the names would be secondary homonyms (homonyms due to a combination change rather than conflict of original combinations), which only count as such if the two species are "currently" regarded as belonging to the same genus. So if the other person doesn't regard 'Velociraptor' mongoliensis as belonging to Tyrannosaurus, they could potentially argue that there wasn't a conflict (but this is one of the reasons why there's a recommendation that authors avoid using species names that have been used in closely related genera for new species if they can).

    The ICBN, offhand, doesn't distinguish between primary and secondary homonyms (as far as I know). But the ICBN does have requirements for regarding a name as 'validly published'.

  6. I got this and The Dinosauria for Christmas as I'm looking at going back to school to get into the field of paleontology. Skimming through I noticed some vagueness to the descriptions myself but I have a feeling that possessing and reading this will be beneficial in cross referencing sources despite its vagueness. One thing I immediately flipped to was the Marginocephalian section and I noticed he definitely takes Horner et al.'s work on Pachycephalosaurus and Triceratops to heart in his descriptions and models. That's my main interest in the field so of course that had to be what I skimmed first.