Re: Bipedal lizards & pterosaurs, reply

[david peters <davidrpeters@earthlink.net>]

Thank you, Darren, for making one last return visit. Your arguments are 
well-spoken, but I have a response to each one that answers your queries and 
problems with the bipedal lizard model for pterosaur locomotion. For the most 
part you have forgotten or ignored where bipedalism originated in this lineage. 
Not with pterosaurs, but with their prolacertiform ancestors. I will truncate 
many of your replies to save space. I urge interested readers to follow the 
thread backwards for more from Darren.

Also, thank you, Darren for making this large effort. Your arguments against 
this hypothesis may be those of others as well. So it's good to nip these in 
their respective buds.

Darren wrote:
-- Why do certain lizard species run bipedally?
Is it just so that they are faster? Probably not:

>>>Agreed. And that was never an issue.

Darren: What appears most likely is that bipedal running
in lizards has evolved to circumvent Carrier?s constraint: 

>>>This is a good possibility. But note that pterosaurs did not invent bipedal 
>>>locomotion within their lineage. Their lizardy ancestors, including 
>>>Cosesaurus, Sharovipteryx and Longisquama did. Rieppel even made a case for 
>>>Macrocnemus (in 1989, I think), but that may be stretching things abit.

Darren: Given that
pterosaurs would clearly not have needed to avoid Carrier?s
constraint (because, even if they were to run quadrupedally,
they would not be compressing the thorax with each stride),
why run bipedally? 

>> See above. Pterosaurs inherited bipedality.

Darren: If speed is what it?s about, it strikes me
as unlikely that animals like pterosaurs would rely on
terrestrial sprinting to escape predators or to get from A to
B. Furthermore, given that pterosaurs exhibit features
associated with leaping (see, e.g., Bennett 1997) and/or
scansoriality, it is probable that they wouldn?t need to sprint
in order to take off. 

>>> Too bad you weren't in St. Paul to see the posterior of Longisquama in my 
>>> podium presentation. Surprisingly, it has the proportions of a lemur.

Darren: Bottom line: there is no inherent ?need?
for good bipedal running abilities in pterosaurs, in contrast
to the situation in lizards.

>>> Again, you have to go back to cosesaurids to figure out the 'need' for 
>>> bipedality.

Darren: -- What allows certain lizards to run bipedally?

Dave is fond of stating that pterosaurs could run bipedally
because they exhibit an increased number of sacrals relative
to their probable outgroups, and a large preacetabular
process on the ilium. 

>>>  Inherited from cosesaurids.

Darren: As has been pointed out several times
in the literature, it?s relatively easy for a lizard to run
bipedally IF it combines these two features WITH (1)
hindlimbs that are proportionally longer than its forelimbs
(and consequently the animal has proportionally short
forelimbs), 

>>>  See Cosesaurus, Sharovipteryx and now Longisquama.

Darren: (2) a proportionally short thorax and 

>>> See Cosesaururus, Sharovipteryx and pterosaurs including basal forms. 
>>> Longisquama has a lonnnn-n-g lumbar region, as in lemurs.

Darren: (3) short
neck*, and 

>>> In Cosesaurus the neck is less than half as long as in Macrocnemus. In 
>>> Longisquama and MPUM 6009, the Milano ?Eudimorphodon ?juvenile 
>>> (cladistically the most primitive pterosaur) the neck is shorter still. 
>>> Shorter than in any later pterosaur. 

(4) a long muscular tail (see Synder 1954, 1962,
Bellairs 1969, Rieppel 1989 etc). 

>>> Long tails pterosaurs have. Muscular tails, no. But note that in 
>>> Cosesaurus, with its hypertrophied anterior ilium, the decoupling of the 
>>> caudofemoralis complex was already underway. Simplified, lizards run with 
>>> two sets of hind limb muscles. Retractors (from the tail) and rotators 
>>> (from the hips). In prolacertiforms on their way to becoming pterosaurs, 
>>> the retractors steadily diminish, while the rotators are expanded with the 
>>> invention of the prepubis forming new anchors points and both ilial 
>>> processes growing huge (not in the literature yet, and overlooked by Wild 
>>> and Dalla Vecchia, but Wellnhofer will show this soon, and I showed it in 
>>> St. Paul). So while the bipedal lizard model works well for cosesaurids, 
>>> pterosaurs have graduated many grades beyond this.

Darren:  Note that most of these
features are to do with reducing the mass of the foreparts
and thus shifting the CoG caudally. 

>>> As in Cosesaurus, Sharovipteyx and Longisquama.

Darren: On point (4), as shown
by Russell and Bauer (1992), the most important anatomical
correlate of bipedality in lizards is the presence of a large m.
caudofemoralis longus that inserts relatively distally on the
tail (thus explaining why the lizards that run bipedally are
the same ones that don?t practise caudal autotomy).

>>> Don't have the literature in front of me (but I will tonight!), as I recall 
>>> the better bipeds, like basilisks, have less robust tails and fewer 
>>> transverse processes, disappearing first caudally, as in prolacertiforms.

Darren:  *Apparently _Chlamydosaurus_ has a longer neck than
most other agamids. Its neck is still not as proportionally
long as that of a pterosaur though.

>>> The first pterosaurs (MPUM 6009) and their closest sister taxa had very 
>>> short necks.

Darren: Pterosaurs obviously don?t have proportionally short
forelimbs, but more importantly they don?t have the short
neck seen in bipedal lizards, nor do they have a tail that
would have supported a large m. caudofemoralis longus:
even in basal long-tailed forms, transverse processes (and
hence a reliance on m. caudofemoralis longus) are
extremely reduced (and, incidentally, there is no indication
that pterosaurs switched to the knee-based retraction system
seen in birds). 

>>>  The knee-based retraction system is a Padian bird-convergence hypothesis 
>>> that is here conisdered erroneous. Very little if any knee retraction is 
>>> used or was necessary (based on moving models through pterosaur footprints, 
>>> and the femoral rotation model for pterosaur leg movement) in the 
>>> prolacertiform hypothesis.

On the relevance of this reduction in
caudofemoral musculature to bipedal locomotion, Synder
(1954) writes ?while a long, heavy tail does not necessarily
indicate bipedal habits, a short, lighter tail precludes the
possibility of this type of locomotion? (p. 9). 

>>> Was this statement given after Snyder cut the tails off of bipedal lizards 
>>> to see if they could still run bipedally?  Or was that someone else?  I'll 
>>> check tonight.  It is important that the given biped have a long tail, so 
>>> the best bipeds (think basilisk and frillneck) appear have a longer, 
>>> lighter tail. Also, remember that the transition occurred in cosesaurids, 
>>> not pterosaurs.

Darren: 
-- So what of the alleged correlates of bipedality present in
pterosaurs? Dave suggests that an increased number of sacral verts and
a hypertrophied preacetabular process on the ilium are
indicative of ?improved? bipedality in pterosaurs. The
problem is that, firstly, the features discussed above are
needed as well (viz, proportionally short neck, big m.
caudofemoralis longus etc), 

>> previously covered.

and, secondly, when the sacral
and iliac features are present without these others, they may
not be indicative of bipedality but of quadrupedality. Look
at (e.g.) ceratopsians. 

>>> Descended from bipeds. And armored forms, as Greg Paul notes in DA, also 
>>> have extended ilia and sacral counts, by convergence. The same dynamics are 
>>> not at work there.

Relative to basal ceratopsians,
ceratopsids have a longer preacetabular process and an
increased number of sacrals (10-11 compared to 6), so
according to your criteria ceratopsids might be better suited
to bipedality than psittacosaurs. 

>>> Possible, according to Paul, but again, different dynamics are at work 
>>> here. They're not lizardy.

Darren: Parareptiles come to mind
too: in nycteroleterids, nyctiphruretids, procolophonoids
and sclerosaurs there are (usually) 3 sacrals and a short or
absent preacetabular process, 

>>> Don't forget nothosaurs. Cosesaurids start at four sacrals.

Darren: but in pareiasaurs - most
notably in big derived forms like _Scutosaurus_ - there are
4-6 sacrals and the preacetabular process may be so
hypertrophied that the pelvis looks much like that of a
pterosaur (see Fig. 14E in Lee 1997). 

>>> Armored.

Darren: As in ceratopsians,
these sacral and iliac trends are to do with improved
quadrupedal abilities.

Similarly then, the features you cite in pterosaurs might
therefore indicate improved _quadrupedal_ abilities relative
to their (?) more bipedal ancestors.

>>> True. They are better quadrupeds. Especially on tree trunks. And later on 
>>> land (see Unwin's latest abstract).

Darren: A better way of testing for bipedality in Pterosauria might
be to look at intermembral indices (viz, forelimb:hindlimb
ratios), 

>>> Comments to come when the paper is seen.

Darren: at the CoG (as I mentioned, Sangster has been
working on this), 

>>> Wish I could see her work. Can't comment otherwise.

Darren: at unambiguous soft tissue evidence (e.g.,
the Crato azhdarchoid with its preserved brachiopatagium),

>>> That tissue preserves aktinofibrils and impressions of wing bones. Two 
>>> reason to consider them outer wing membranes, not inner ones. Plus the 
>>> tissues cross under the tibia and can be seen on both sides as a continuous 
>>> ribbon. 


Darren: or at trackway evidence? 

>>> we have cosesaurid tracks. Their bipedal, narrow gauge and digitigrade. 
>>> Then the big Jurassic blank slate and then Cretaceous waders. 

Darren: and right now the evidence from
all of these areas shows that quadrupedality is better
supported, or in other words that pterosaurs were more
likely quadrupedal.

>>> Agreed. Pterosaurs were quadrupedal.

>>> Conclusive reiterations skipped.

Darren: - The presence of a well-developed iliopubic ligament
(=ligamentum inguinale) might indicate that pterosaurs
were good at elevating the thorax, but given that everyone
agrees that pterosaurs must have been bipedal when
opening or closing their wings (and they thus would have
needed to elevate the thorax at least occasionally), this
doesn?t necessarily indicate bipedal running. 

>>> Thank you for those concessions.  But remind the pterosaurs not to wander 
>>> from that spot or they might be accused of walking.

Darren: Incidentally,
see Hutchinson (2001, pp. 156-8) for a discussion of
iliopubic ligament distribution in Reptilia. Because reptiles
including ceratopsids, pareiasaurs and turtles appear to have
had a hypertrophied iliopubic ligament as well, the
correlation between this structure and an enhanced bipedal
ability is not immediately clear.

>>> Thanks for the reference. Darren made some very good points. But it is 
>>> important to remember the heritage of the Pterosauria. Remember, as in 
>>> birds, the wings came last.

David Peters
St. Louis