Restoring Dinosaurs (was Hadrosaur necks, etc.)

["Ralph Miller III" <gbabcock@best.com>]

Andy Farke <andyfarke@hotmail.com> wrote:
> Also, regarding paleo reconstructions--I don't agree with taking 
> evidence for one animal (e.g., sauropod spikes) and applying it to all. 
> Likewise with feathered dinosaurs--just because one had them doesn't 
> mean all did. I won't hold that against anyone though--I still love G. 
> S. Paul's dinosaurs, spiked sauropods and all!

Unfortunately, (or fortunately, for artists who enjoy putting a personal
stamp on their work), dinosaur restoration requires some degree of
speculation.  And in the nitty gritty details of this "fleshing out" --
where there tends to be a paucity of solid evidence -- paleolife artists
can only do their best to guess correctly.  But the research that they do
is essential to making an educated guess.

Given that _non-ossified_ spiky features are known for a single undescribed
sauropod specimen (said to resemble _Barosaurus_ and _Diplodocus_), and
ossified scutes are known for several titanosaurs, what can we say about
other sauropods?  The rarity of described fossils of soft tissues (or more
to the point, anything but the bones) among the dinosaurs raises serious
problems for the paleolife artist.  It may be safe to suggest that more of
the bony elements (such as the titanosaur scutes) should have been found
and described by now on a wider variety of sauropods if these features had
been ubiquitous among the sauropods.

But even here, one must exercise caution, for -- and this was especially
true in the early days of dinosaur science -- excavations and initial
descriptions are not always as proper and painstaking as one would hope,
and misidentifications have occurred on many occasions.  For example, in
_Dinosaur Lives_, Jack Horner describes a dinosaur nesting horizon chock
full of eggs that had been mistaken for fossil clams!  It has been remarked
that in the past paleontologists accustomed to finding and excavating only
bones may have ignored and inadvertently destroyed fossilized skin
impressions.  In his _Skin_ article in _The Encyclopedia of Dinosaurs_,
Stephen Czerkas points out that the first discovery of dinosaur skin
impressions (by S. H. Beckles in 1852) was not heralded as such, because
the cetiosaurid sauropod skin was presumably of crocodilian origin.

Even so, it seems fair to assume that there was more than one sauropod
genus with spiky elements, and hence, that there is some basis for applying
these features to other sauropod genera.  The point is that at this point
we cannot prove the case one way or another because such details are so
rarely preserved as fossils.

To the critics, I ask this question:  Given that there are perhaps two
dozen skin impressions attributed to specific dinosaur types, how would YOU
go about restoring the unknown hides of the hundreds of known dinosaur
genera remaining?  If you were Stephen Czerkas you would work from known
impressions of the most closely related animals, scaling them up or down as
needed, in order to produce a hypothetical approximation of a correct
integument.  You would put _Centrosaurus_ and _Chasmosaurus_ skin on your
_Styracosaurus_, for example, because _Centrosaurus_ and _Chasmosaurus_ are
the closest relatives for which you have good impressions.  Obviously, the
more detailed and complete the fossil evidence, and the closer the
relative, the better.  This is essentially no different than the use of
size-adjusted _Gorgosaurus_ skeletal elements to extrapolate missing bones
for a _Tyrannosaurus_ mount, and it is the best you can do under the
circumstances.  You could be proven wrong, but as a paleolife restoration
artist, you're not going to illustrate or sculpt only half of an animal
because paleontologists have yet to recover a 100% complete dinosaur!  Yes,
you could illustrate a partial _skeleton_, (as Tracy Ford does), but that
is a different subject.  Admittedly, artists are sometimes enlisted to
restore an animal that is known from very fragmentary evidence, and this
leads to serious errors.  But that is a far cry from the legitimate
incorporation of the known elements of closely related genera.      

If you are restoring the life appearance of a sauropod which is more
closely related to _Barosaurus_ and _Diplodocus_ than _Saltasaurus_, then
it seems to me that you would be very justified in applying spikes to your
sauropod.  Looking at a particular case, _Brachiosaurus_, we read that
Wilson and Sereno (1998) place the Brachiosauridae clade closer to the
Titanosauria than to the Diplodocoidea, so it is unclear whether the spikes
are appropriate.  On the other hand, there is no evidence for
titanosaur-style bony scutes among the brachiosaurs -- which one might
think SHOULD have been preserved -- so you must guess for yourself what you
think is appropriate, and be open to the possibility that your restoration
may prove wrong as new fossils are unearthed.

Having said this, there are clearly different opinions on how the available
fossils are to be interpreted.  For example, (and I hope I am interpreting
the articles correctly this time!), Stephen Czerkas favors the application
of _Kritosaurus_-type dermal spines more generally on the duckbills,
eschewing the ornamental "skin ribbon" frill interpretation which Gregory
S. Paul supports for edmontosaurs and corythosaurs.  However, Gregory S.
Paul does concur with Czerkas on the nonbony hornlets (the aforementioned
dermal spines) on _Kritosaurus_, and on the possibility of a "deep,
withers-like neck" on the ornithopods with strongly downcurved backs.  But
there is no across-the-board consensus among contemporary paleontologists
and paleolife artists on some of these issues. 

Regarding feathered theropods, the problem again is the improbability of
finding preserved fibers or impressions thereof.  It seems to me that
protofeathers or feathers of some sort can reasonably be applied to
restorations of any small theropod which is a member of the clade comprised
of the last common ancestor of _Sinosauropteryx_ and _Archaeopteryx_ and
all of their descendants.  This may also apply to hatchlings and juveniles
of the larger species, and this could even apply to all hatchling dinosaurs
for all we know, for there are no good scale impressions known for small
dinosaurs.  Indeed, some degree of endothermy should likely apply to the
aforementioned "feathered" clade as well.  It is also possible that some of
these features were secondarily lost in the descendants.  We don't know. 
We do know that _Carnotaurus_ was covered in small tubercular scales
surrounding low, conical (nonbony) studs.  _Ceratosaurus_ is known to have
had a single row of bony ossicles atop the neck and tail.  And small
patches of _Tyrannosaurus_ skin impressions have been found, revealing
small tubercular scales.

Restoration can be most confidently accomplished where the fossils are most
detailed, most numerous, and most complete.  The most conservative approach
would be to leave the bones locked in the matrix and leave them alone.  If
you choose to restore a prehistoric animal, you can leave off the spikes,
the protofeathers, the withers, the frills, the wattles, and the dewlaps,
but that doesn't mean that the resulting image will be any more lifelike or
authentic than a more elaborate interpretation.  Any restoration will raise
some hackles among purists because to many viewers it will present the
defining image which says: "Here it is.  This is what the animal looked
like."  This merely points to the need for explanations which reach out to
the public, informing them of what is "real" and what is "speculation" or,
as some among the lay public would say, "faked."  

As to those professional illustrators who slavishly ape the restorations of
other artists with no knowledge of the subject (especially going back to
the works of Knight and Burian), they do so at their peril, and they are
not doing their jobs.  Shame on them.  Paul, Franczak, and Ford (among
others) are in a particularly critical position when they reconstruct a
skeleton or restore an animal, for their work inevitably will be reflected
in the images produced by many of their contemporaries, but I feel secure
that the aforementioned are doing their best under the circumstances to
present the animals as they really were.  These artists are well aware that
they will probably make mistakes when they engage in a work that
incorporates speculation, or they wouldn't be in this profession.  Would
the public be better served by images of dinosaurs that are drab and
featureless?  That is the problem with the "conservative" approach to life
restoration.  Rest assured that if new evidence surfaces to disprove the
spiny appearance of certain sauropods or the withers of hadrosaurs, Gregory
S. Paul and other artists of integrity will cease producing works with the
offending features, and will touch up some of their outdated works. 

I think that the following passage from Gregory S. Paul's (recently
extinct) _Predatory Dinosaurs of the World_ sums it up nicely (Turn your
hymnal to page 123):

"What it all comes down to is that, at this time, and despite what I know
are the traditionalist's strong sentiments, it is every bit as speculative
and no more legitimate to portray small dinosaurs in naked skins as it is
to show them insulated.  Eventually, enough impressions of small dinosaur
integument will be found to settle things.  Until then, I always draw my
small theropods feathered.  For those who feel I should do some naked for a
more "balanced" view, plenty of others have and are doing that."

The same general sentiment could be applied equally to the spiny sauropods
and withers-necked hadrosaurs.  To each his (or her) own.  These images
were produced by artists who did their best to work with the fossil
material available at the time, and they committed their personal
interpretations -- speculative though they may be -- to canvas, bristol
board, or sculpture.  Go ahead, leave the darn spines off your sauropod,
but be prepared to put spines on as new discoveries come to light.     

-- Ralph Miller III     gbabcock@best.com