[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

Re: More on Birds, dinosaurs, and digits (LONG)

I hope the following makes sense, as it has now cost me my Thanksgiving
break, ticked off my wife, and caused me to spend an unconscionable sum
finally shelling out for Currie & Padian.  If anyone reading this is
familiar with ICM terminology, the thaat is the bird-therapod controversy,
the raag is the I-II-III vs. II-III-IV argument, the bandish is really
Stanley Friesen's, but any disasters in the performance are mine.  

Burke & Feduccia's argument, slightly rearranged, runs like this.  The
hypothesis to be tested is that birds are in the theropods.  One of the key
synapomorphies used to place Aves within Therapoda is the I-II-III digital
formula found in many therapods.  However, embryological studies show that
the bird manus is actually II-III-IV.  This pattern is particularly clear
from the development of the avian pes, which is analogous.  I-II-III is, in
fact, a unique adaptation of theropods which "may have evolved by strong
selection for a grasping, raking hand."  However, the "theropod" formula
"violates a consistent pattern of digit reduction in all other tetrapods."
Thus it is unlikely that birds are derived from dinosaurs because they lack
this unique and important synapomorphy.

I've left out the stuff about the "primary axis" because its meaning, if
any, is obscure.  It is a nineteenth century embryologist's guess about
develomental causation.  Frankly, given the state of molecular developmental
biology at the moment, we can do a lot better.  I hope (but don't really
expect) to be able to get to the literature on this in the next few weeks.
What I've found so far (abstracts on the net) suggests that the "primary
axis" concept is probably going to turn out to be meaningless.

The point I wanted to pursue was the rather facile assumption that the
I-II-III formula was a peculiar and irreversible synapomorphy of theropods
and could be regarded as a proxy for the clade.  I am assuming that B&F's
analysis of the embryology is entirely correct.

It turns out that the distribution of the I-II-III pattern is not at all
that simple.  According to Dodson, as I mentioned in another post,
Psittacosaurus, a highly derived ceratopsid, shows the same pattern.  This
is way over on the Ornithischian line, and about as far from the theropods
as one could find among dinosaurs.  Thumbing rapidly through C&P, one also
finds that Plateosaurus, a Prosauropod, shows a similar pattern (p. 598,
after Lucas 1994).  B&F show Herrerasuarus as an example, but it isn't at
all clear that Hererrasaurus is a therapod, or even a dinosaur.  Certainly
Lagosuchus is not, but its pes looks suspiciously like  I-II-III to me (p.
484, after Sereno 1991).  Jeff Poling mentions the possibility of an
amphibian with this digital formula, which would make the case all the
stronger.  In sum, the I-II-III is widely distributed among Dinosauria, is
not restricted to Dinosauria, and may even date back to quite basal tetrapods.

By contrast, many dinosaurs have the ordinary digital formula with no
significant reductions.   More to the point, a good number show the "bird"
II-III-IV formula.  Significantly, this includes the *Therapod* pes!  See
Tom Holtz's article in C&P on "Arctometatarsalia".  In fact, Tom's article
suggests that some therapods approached a II-IV formula by fusion of III and
IV, a very suggestive development.  Other examples include (if I interpret
the drawings correctly) Centrosaurus (p. 321) and Iguanodon (p. 321) .

If there's a pattern here, I don't see it.  What the evidence suggests is,
not that I-II-III is a proxy for Theropoda, but that *all* dinosaurs, and
pre-dinosaurians at least back to Lagosuchus, had the ability to adopt
either a I-II-III digital formula or a II-III-IV formula.  Further, the
change may have been reversible, although this is less clear.  Certainly it
possible to have both adaptations in the same lineage, e.g. Psittacosaurus
and Centrosaurus.  Thus, there is no real surprise if therapods and some
early birds had a I-II-III manus, but other birds did not (Dinogeorge has
already suggested this).

Where B&F go wrong is in buying into nineteenth century embryological
concepts along with their elderly (but still entirely valid) embryological
experimental technique.  They get tripped up by concepts like "primary axis"
and "rules" of digit reduction which were sophisticated and important
concepts for their time, but were ultimately flat guesses about
developmental causation which don't necessarily hold up outside fairly small
groups.  Turtles and crocs, for example, are fairly small groups, or at
least far less diverse than the dinosaurs and have a quite restricted
ecological range.  How much variation would one expect?  This is less true
of lizards, but I wonder: how much is really known of comparative lizard

This is my first attempt at serious biological analysis in something like 20
years, so I thank y'all for your patience if you've managed to wade through
it to this point.

  --Toby White

>At 05:27 PM 11/28/97 -0600, Toby White wrote:
>> Stanley Friesen <sarima@ix.netcom.com> said:
>>>previously.  Looking at the pictures they show of hand development in
>>>crocodilians, I saw something interesting.  While the "main axis" is fairly
>>>clearly through digit IV, this main axis is *curved* laterally through the
>>>metatarsals!  This leaves digit III as the one that is actually colinear
>>>with the humerus.  So asked myself, what would the result be if digit IV
>>>reduced in size, and perhaps digit III increased in size, to the point
>>>where distal metatarsal IV was substantially smaller than distal metatarsal
>>>III?  From the looks of things, such a process would, with *no* *further*
>>>*changes*, be sufficient to shift digit III onto the "main axis".  No need
>>>for subtle changes in proportions in the adult.
>>All of which strengthens their argument.  That is, the change needed to
>>establish III as the main axis, starting with bassal crocs, is relatively
>>slight.  That change explains what we see in dinosaurs, and is widely
>>distributed in dinosaurs.  In fact, I just ran across a reference to a
>>I-II-III manus in ceratopsians in Dawson's book.  Given its closeness to
>>crocs and wide distribution in dinosaurs, the posited III-based main axis
>>presumably arose quite early in the dinosaur lineage.  Yet, it is not found
>>in birds nor in any other group of tetrapods.  The natural conclusion is
>>that birds did not arise from dinosaurs.  
>   Actually, I think you've drawn the wrong conclusion.  If only a slight
>change is needed in croc embryos to shift the main axis to digit III, then
>one can easily see the main axis shifting to digit III in bird embryos, the
>crocs' closest living relative.  It is more likely, given the scenarios in
>the above three paragraphs, that avian digits are really I-II-III and the
>observations of Feduccia and Burke are incorrect because the main axis
>shifted to digit III long ago.
>   Now throw into the pot that the main axis in some amphibian embryos is
>apparently digit III ... still looking for that ref.
>** Dinosauria On-Line. Home of THE DINOSTORE ** "Those who trade a        **
>** (Dino stuff for sale), Jeff's Journal of  ** little freedom for a      **
>** Dinosaur Paleontology, Jeff's Dinosaur    ** little security will soon **
>** Picture Gallery, and The DOL Dinosaur     ** find they have none of    **
>** Omnipedia. http://www.dinosauria.com      ** either." -- Jeff Poling   **