re: Interpreting the autopodia of tetrapods - comment

[David Peters <davidpeters@att.net>]

A recent paper by Hone, Sullivan and Bennett (2009) concluded that “IP  
[interphalangeal] hinge lines, as conceived by Peters (2000a), can  
often be superimposed on 2-D drawings of tetrapod mani or pedes but  
have little or no value as a tool in palaeontological and biological  
research."

Thanks a lot guys! That paper has been cited fewer than a dozen times  
and you're trying to knock it down?

The observation of parallel interphalangeal lines (PILs) was first  
offered by Peters (2000a) as I reported on the virtually parallel  
lines that I found could be drawn through the interphalangeal joints  
and across the ungual tips of virtually every tetrapod manus or pes,  
including wings and flippers. That such lines were present appeared to  
indicate that the phalanges worked in sets, whether for walking   
(extension) or climbing (flexion). One can see the knuckles line up in  
the grip of the human hand, for instance, but that lineup changes when  
the hand is flat (not being used). I considered these lines obvious,  
but overlooked, understudied and potentially useful. Several dozen  
examples were presented from Devonian forms to the present. More to  
the point of the paper, I was looking for a way to understand and  
perhaps distinguish digitigrady from plantigrady in pterosaurs and I  
thought I had succeeded.

Contra the Hone, Sullivan and Bennett (HSandB) conclusions, the odd  
thing about this paper is how often HSandB report that PILs can indeed  
be drawn and may have some usefulness. They report:

"IP [interphalangeal] hinge lines can often be superimposed on 2-D  
drawings of tetrapod mani or pedes."

“Some hinge lines clearly are valid, in the sense that straight lines  
can indeed be drawn precisely through three or more joint centres or  
ungual tips for some tetrapod digits in some positions.”

“In general, there are likely to be a number of specific cases in  
which an IP hinge line does indeed operate as an axis of rotation, as  
in the metatarsus of Iguana during initial elevation of the metatarsus  
(Brinkman 1980). In such cases, the concept of a hinge line is useful  
from a functional perspective, and likely to be helpful in describing  
the observed motion.”

“At best, it might be possible to use the hinge line criterion to  
identify a broad envelope of possible configurations for a given foot  
or hand, and to exclude as implausible the inverse set of  
configurations on which hinge lines could not be drawn.”

Unfortunately, what HSandB demand is for IP hinge lines to intersect  
exactly at the middle of IP joints and to do so parallel to the joints  
themselves. While it is remarkable how often this does happen (some  
joints are even angled apparently just for this) it doesn't happen all  
the time with every joint. That’s because the IP lines are applied to  
imperfect animals that are in the process of evolving. A higher level  
of precision was never expected or encountered. In fact, as  
radiographic images of a moving pes indicate (Peters 2000a), the lines  
shift during every step cycle. Moreover the medial and transverse line  
sets seem to have evolved for extension while the transverse set seems  
to have evolved for grasping parasagittal branches and other similar  
substrates. So they don’t work together. Nothing about these lines was  
ever meant to be as precise as Hone, Sullivan and Bennett (2009)  
appear to demand. In fact, some clades are specified by hinge lines  
that intersect the middle of certain phalanges and unguals. That  
breaks all the so-called “rules,” but that’s okay, or so I have  
observed.

HS and P tested the IP hinge lines on an ornithomid pes drawing taken  
from Romer. First of all, this hypothesis was never meant to be proven  
on one-, two- and three-toed taxa. It applies only as a vestige.  
Second, the drawing they chose is one with parallel digits, as if in  
the recovery phase of the step-cycle, not spread out in use.

HS and P tested the pes of Smilodon, which I had copied from a  
plantigrade drawing. My bad. They used cylinders(?) created in a 3-d  
program to show what would have happened in a digitigrade  
configuration. Unfortunately they didn't look at a genuine Smilodon  
mount. The metatarsals don't all line up transversely. They arc. The  
medial and lateral ones fall a little behind. That's important for the  
lineup of the rest of the digits. Plus cat toes are never in a  
straight line and should never be tested that way. When it's  
configured correctly, like most museum mounts, the IP hinge lines are  
there.

HS and P note that in a few taxa, like ratites, the pads don't line up  
with the phalanges. True. Counting the number of pads per digit is the  
solution to reveal inconsistencies. If there's one for every phlalanx,  
use it. Somehow they didn't think of that.

With regard to pterosaurs HS and P reported, “there is no reason to  
think that all Pteraichnus trackways were made by ctenochasmatids, or  
even that any Pteraichnus trackways were necessarily made by  
ctenochasmatids.” Peters (2000) provided evidence with pad counts that  
matched phalanx counts and digit proportions that matched ichnite to  
pes. No counter evidence was provided. No other pterosaur pes was  
matched to the track. If a statement is false they really should have  
falsified it.

Hoping to distance pterosaurs from my digitigrade reconstruction of  
Cosesaurus (matched to the digitigrade ichnite Rotodactylus) HSandB  
report, "Moreover, the suggestion of a close relationship between  
pterosaurs and prolacertiforms such as Cosesaurus is contradicted by a  
considerable weight of opposing evidence.” Their evidence consisted of  
six cladistic analyses in which Cosesaurus was not included as a  
tested taxon and one analysis (Hone and Benton 2008) in which only a  
quarter of its characters were scored.

Red herrings don't make good arguments.

I sincerely hope you all don't take this paper at its word and forget  
about PILs. It's ideal for reconstructing scattered phalanges but  
should be used in accord with articulated sister taxa.

David Peters
St. Louis


Hone, D.W.E. Sullivan, C. and Bennett, S.C. 2009 Interpreting the  
autopodia of tetrapods: interphalangeal lines hinge on too many  
assumptions. Historical Biology iFirst article, 2009, 1–11 DOI:  
10.1080/08912960903154503

Peters D. 2000a. Description and interpretation of interphalangeal  
lines in tetrapods. Ichnos. 7:11–41.