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

Cearadactylus - long, but with sausages

Dear DMLers

Herewith some responses to recent comments on my Cearadactylus paper. 

Peters in a previous post noted: 
>The trouble is, even with a cladogram of 50 taxa and 160 characters, way
more than 80 percent of the characters turn out to be homoplastic. That
makes  listing a few pertinent characters almost fruitless because you
might pick the right ones and you might pick the wrong ones. Worse yet,
certain basal pterodactyloid nodes are separated from one another by at
most one or two steps - even with five distinct Pterodactylus species
for starters. One mistake or misjudgement and nodes start shifting. I
know from experience!<

I cannot comment directly on Peters' data set as it is not published and I have 
not seen the current version. Such a high incidence of homoplasy might suggest 
that character definitions and scorings need some serious attention. 

Peters also noted: 
>The firecrackers really started popping when I read your diagnosis for
the Ctenochasmatidae (Gnathosaurus, Pterodaustro, ca & ad (= common
ancestor and all descendants). abbreviated here. My mental queries
follow in parentheses.
1. rostrum ant. to NAOF more than half length of skull  (also in:
Pteranodon, Nyctosaurus, Quetzalcoatlus, and Diopecephalus)<

Yes, not surprisingly, there is some homoplasy in my analysis. But as 
Pteranodon, Nyctosaurus and Quetzalcoatlus are deeply nested within other 
clades the homoplasy is clear. If by 'Diopecephalus' Peters meant 
'Pterodactylus longicollum' then yes indeed this is so because 'P. longicollum' 
is probably part of Ctenochasmatidae. However, I'm still studying this so it 
does not feature in the Cearadactylus paper. 

Peters also noted: 
>2. anterior rostrum dorsoventrally compressed and rounded (actually your
fig. 1D looks pretty straight to me. I've seen a rounder (dorsal view)
rostrum in other anhanguerids. Plus, I wouldn't say "compressed " so
much as "excavated ventrally" to produce that maxillary "step" you

Irrespective of how we might wish to redefine the morphologies, the character 
'anterior rostrum dorsoventrally compressed and rounded' is putatively 
apomorphic for a clade that I call Ctenochasmatidae. 

Peters also noted: 
>3. Seven or more pairs of teeth in the premaxilla. (I wish you had drawn
the premax-max suture. Or noted it on a photo. You left it blank. <

I draw things in when I can see them in the fossil, or there is clear evidence 
for them in illustrations of the fossil material that has been independently 
confirmed by others who have studied the actual material. 

Peters also noted: 
>My scan [attached to personal email] shows that, like all pterosaurs,
except for the real ctenochasmatidae, only four teeth appear in the
premaxilla of basal ornithocheirds ( or anhanguerids). Here the first
premaxillary teeth are gone so three are present. They are tiny in other
forms you are aware of, so it's no big loss. The suture is visible
between 3 and 4. )<

Peters' scan shows two squiggly lines drawn across the premaxillae and 
descending between tooth position three and four. I can see no evidence for a 
suture in such a position in the illustrations available to me, and such a 
suture has not been reported by anyone else for Cearadactylus. Typically there 
are four teeth in the premaxillae of pterosaurs, so if Cearadactylus exhibited 
the standard condition I would expect a suture (were it present and visible) to 
be located one tooth position further back. Peters states that the first pair 
of premax. teeth are gone, but presents no evidence to support this. 

Peters also noted: 
>You may have mistakenly assumed that the premaxilla began at the
beginning of the jawline "step." That's not true of Huanhepterus,
either, which also has only four premaxillary teeth and the rostrum is
neither compressed nor rounded, but "sags" ventrally at the tip.<

So far as can be determined from figures of Huanhepterus (I have still to see 
the original skull) the anterior region of the rostrum was dorsoventrally 
compressed and rounded. If someone can provide photos or drawings to show that 
this is not the case I would be most interested to see them. 

Peters also noted: 
>4. Teeth project laterally...at least anteriorly. (also in: Dorygnathus,
Angustinaripterus, Haopterus and Diopecephalus).<

In most ctenochasmatids teeth stick our horizontally or sub horizontally from 
the anterior part of the jaw. The anterior teeth of Dorygnathus, 
Angustinaripterus and in fact most ornithocheirids show some lateral flare, but 
are still largely sub vertical. These two conditions are clearly distinct from 
each other and easily detectable in fossils. Interestingly 'P. longicollum' 
shows the ctenochasmatid condition - see comment above. 

Peters also noted: 
>5. Teeth in anterior dentition relatively elongate, with cylindrical
crown, tapering only at the distal tip (that's a very slight variation
on the above list, but perhaps worth noting).<

Another distinctive feature of ctenochasmatids unrelated to any of the other 

Peters also noted: 
>6. Metatarsal III more than one third the length of the tibia (what
about nearly every long-tailed pterosaur, plus P. antiquus, and at least
one Germanodactylus?)<

Ctenochasmatids are clearly not 'rhamphorhynchoids' as I think anyone can tell 
so I see no problem with this. The distribution of this character within 
pterodactyloids is discussed in the text of my paper where I note, for example, 
that P. kochi and P. antiquus exhibit the derived condition. This suggests that 
this character may be apomorphic for a more inclusive clade within 
Ctenochasmatoidea - but I have no problem with that. 

Peters also noted: 
>In my own analysis of all pterosaurs, unfortunately very few skull
characters turn out to be diagnostic within the limited clades you're
working with. At the familial level, the skulls are pretty conservative.<

See my first comment above. 

Peters also noted: 
>One that you mentioned, the greatest depth of the mandible occurring at
the symphysis, is also shared by such diverse taxa as P. antiquus,
Zhejiangopterus, Tapejara and Pteranodon. In fact it's the defining
character of that large clade. The filter-feeders don't have it.<

If we are going to discuss aspects of published work then citations should be 
accurate and set in context. What I actually wrote was:

'An unusual feature of the mandibles is that they appear, in lateral view, to 
achieve their maximum height toward the caudal end of the symphysis, and are 
somewhat lower, though of about even height, from this point to the articular 
region. Moreover, as Leonardi & Borgomanero noted (1985), at this point the 
mandibles are distinctly deeper than the corresponding region of the rostrum, a 
proportion that is different from the typical condition in pterosaurs where the 
rostrum is usually the deeper of the two (e.g. Wellnhofer 1978). '

The two features cited above are relatively uncommon in pterosaurs, but I make 
no claims in my paper, or here, that they are apomorphic (they might be - I 
haven't tested this).  

Peters also noted: 
>I can't agree with you on the low angle of the quadrate versus the
jawline, nor can I agree with you that the skull was probably lower than
others have reconstructed it. Dorsal to the NAOF is where the skull
profile typically begins to curve dorsally on anhanguerids.<

Traced reconstructions of the skull lain over photos of the fossil clearly do 
not match with the rear part of the reconstructed skull which sweeps up and 
away from the outline of the actual fossil. I'm going to stick with the fossil 
evidence, not with reconstruction's that look as if they have been influenced 
by comparison with other Santana pterosaurs such as Anhanguera. 

Peters also noted: 
>In my data set, a deep jugal is homoplastic over eight separate taxa
going back to Austriadactylus, plus two more: the dsungaripteridae in
which the infilling comes in from the posterior of the orbit and

In the Cearadactylus paper I discuss this character at some length, noting that 
hypotheses of homoplasy are supported by developmental patterns preserved in 
the fossil material. 

Peters also noted: 
>Again, thank you, David, for so promptly sending me this paper. It's a
valuable report on the history of Santana pterosaurs and brought to my
attention several important characters. The status of Cearadactylus,
according to my data which remains unshaken, is still close to
Anhanguera and kin.<

I look forward to the publication of future papers that deal with the 
relationships of pterosaurs. For now I have presented evidence which supports 
the placement of Cearadactylus within Ctenochasmatidae. This evidence is 
discussed in considerable detail in the paper and, if anyone is interested in 
seeing this, I would be happy to send them a reprint. 

For those who just want the précis: There is a large set of clear-cut 
characters supporting Ornithocheiroidea and subclades within this group that do 
not appear anywhere else in Pterosauria. Consequently, it is a relatively 
simple task to determine whether a taxon is a member of this clade, or not. 
Cearadactylus lacks virtually all ornithocheiroid characters except a couple of 
features pertaining to the dentition of ornithocheirids - these I would 
interpret as homoplastic. By contrast, Cearadactylus exhibits a number of 
characters only found in non-ornithocheiroid pterodactyloids, that never show 
up in Ornithocheiroidea. If I try to force Cearadactylus into Ornithocheiridae 
forming a sister group or near sister group to Anhanguera this requires an 
additional 14 steps and some major changes to the general structure of the main 
pterosaur cladogram. With regard to the characters commented on by Peters, 
either the instances of homoplasy were already tested by the phylogenetic an!
alyses reported on in the paper, r Peters interpretation of the skull differs 
from those given by me and other authors. Probably only further material will 
be sufficient to resolve this issue. In the meantime the reader might like to 
consider the following observation: in my paper in press on pterosaur phylogeny 
the principal relationships reported (figured also in the Cearadactylus paper) 
show reasonably good congruence with the few cladistic studies published so far 
including those of Howse, Bennett, Kellner and Viscardi et al., with 
pre-cladistic works by Wellnhofer, Kuhn and Young, and with the known 
stratigraphic distribution of the main clades. Just one study (Peters 1997) 
mentions results that are radically different from any of those proposed by the 
workers mentioned above, or me, but in the absence of a published data set and 
character analysis we are not in a position to determine why this is the case. 

OK, I'm off to the Weihnacht Markt for my half meter long sausage inna bun, 
(purveyed by the inestimable Herr Dibblerschen) so Merry Xmas and a Happy New 
Year to you all on the DML. 

Bye for now.


PS High horse says he has just ejected two people who were trying to break into 
his stable... 

David M. Unwin PhD

Institut fur Palaontologie, MUSEUM FUR NATURKUNDE 
Zentralinstitut der Humboldt-Universitat zu Berlin
Invalidenstrasse 43, D-10115 Berlin, GERMANY

Email: david.unwin@rz.hu-berlin.de

Telephone numbers:
0049 30 2093 8577 (office)
0049 30 2093 8862 (department secretary)
0049 30 2093 8868 (fax)