knuckle-walking (was: hominid classification)

[John Grehan]

> Behalf Of Ken Kinman
The African ape clade (Gorilla, Pan, Australopithecus, Homo,
> among other fossil genera) seems to have support beyond that of the
> molecular data:

"seem to"? I have contacted about 10 prominent primate paleosystematists
 None have been able to provide me with synapomorphies in support of a
Gorilla-Pan-Australopithecus-Homo clade (let along a
Pan-Australopithecus-Homo clade). What characters are being invoked
above?

>      Perhaps the most interesting morphological synapomorphy for the
> African clade is the so-called "stiff wrist"----formed by a shelf of
bone
> preventing the wrist from bending backward much more than 30 degrees.
> This "stiff wrist" aids in knuckle-walking in Gorilla, Pan, and is
also
> present in the earlier Australopithecus species (anamensis and
afarensis).

In my reading of the papers by Richmond and Strait 2000 making such a
claim is about the most tenuous of arguments I have seen and one that
was counter argued by Dainton (2001). The knuckle-walking argument
appeared to me to be more of a reading into the data than out. My view
is that the case has yet to be made, and the current evidence does not
really point in the direction of knuckle walking any more than pointing
any where else. 

> The abandonment of knuckle-walking by the hominid line (as bipedality
> became more "obligate") was a gradual process, 
 
Or it did not occur at all.

By the time
> of Australopithecus africanus, knuckle-walking was pretty much
abandoned
> in favor of a flexible wrist that was free and flexible for a variety
of
> tool-usage (which would culminate in weapons for hunting).  

Basically there is no compelling evidence that hominids preceding
Australopithecines ever knuckle walked. If the orangutan is the nearest
relative of humans then one would not expect any such evidence.

This link
> between morphology and behavior (and the gradual, concerted evolution
of
> both) 

Or the evolution may not be gradual.

is something I find very intriguing and less vulnerable to the
> weaknesses of arguments based on molars (which can more easily
fluctuate
> with diet).

Assertions about fluctuation of characters because of their function
roles seems to rest primarily on the argument that structures with an
functional role are unreliable because they are the result of selection
and therefore prone to be independent of phylogeny. This is seems to be
a fantasy.

>       THEREFORE, my efforts thenceforth will concentrate on two
African
> ape topologies:  (1) Hominidae as splitting from a Gorilla-Pan clade;
and
> the somewhat more popular (2) Gorilla as splitting from a
Pan-Hominidae
> clade.  Frankly, the third alternative (Pan splitting from a Gorilla-
> Hominidae clade) doesn't seem to have any serious support.  If the
> orangutan-Hominidae adherents can't (or won't) find or even look for
> molecular support for their hypothesis, I give up on them.

Fine, but you still have the hominid problem that cannot be addressed by
DNA base sequences. Further, there is no necessary reason why the
similarity of base sequence matches of humans and chimpanzees is a
necessary predictor of phylogeny (not to say that it cannot predict
phylogeny, just that it may not always do so even if the same pattern is
found in different parts of the genome). 

I do wonder if the pre-judgment of DNA similarity over morphology is
really a kind of creationist argument since it removes fossil evidence
from phylogeny and evolution altogether.

John Grehan