On Friday, September 25, 2015 at 12:21:59 PM UTC-4, John Harshman wrote:
> On 9/25/15, 8:37 AM, Peter Nyikos wrote:
> > On Friday, September 25, 2015 at 1:27:02 AM UTC-4, John Harshman wrote:
> >> On 9/24/15, 7:13 PM, Peter Nyikos wrote:
Picking up a bit before I left off, for continuity:
> >>> With the ur-vogel now being pinned down to a taxon of no more disparity
> >>> than the average Linnean archosaur family, it is like tying our hands
> >>> behind our backs to avoid giving a name to that taxon and to have the
> >>> nearest "real group" include pterosaurs as well as ornithischians, etc.
<snip bit I dealt with in my first reply>
> > > The narrowest real group that we
> >> are sure contains the "ur-vogel" (itself a poorly defined term) is
> >> probably Paraves.
> >
> > The "ur-vogel" is the LCA of what used to be called Aves.
>
> Could you define what used to be called Aves?
Romer seemed to regard possession of feathers as the defining
characteristic, Colbert [1956] gave feathers and "warm-blooded"
[his quotes]. Both can be investigated using fossil evidence.
Carroll [1988] wrote: "most paleontologists consider *Archaeopteryx*
to be a bird because of a single, albeit complex character --
feathers -- that are uniquely shared by modern birds."
> I suspect that too is a
> poorly defined term.
See above. You of course disagree with the definition, but
it certainly was unambiguously defined.
> An additional problem is that relationships in that
> neighborhood are not clear, so just what node in what tree would be
> referred to is also unclear.
Use true feathers, not proto-feathers, which you yourself
admit are too rare to be used by cladists for classification.
But if they WERE used, the lack of clarity might disappear.
> > And if Archaeopteryx has no known apomorphies that would rule
> > it out in that role, it is still a candidate for "ur-vogel".
>
> > In fact, that is how I *define* "ancestor-candidate": an organism
> > known from a well preserved and highly detailed fossil that lacks such
> > apomorphies.
>
> That's nearly as good a definition as we could come up with, though
> there's enough vagueness to cause problems. Are there apomorphies that
> wouldn't rule it out as well as apomorphies that would? Or does each
> apomorphy just make its candidacy a little less likely without ruling it
> out completely? There are after all such things as reversals.
Finally, you actually engage what I write! and with unusual helpfulness.
These are questions I would love to leave to old-fashioned naturalists
well versed in these kinds of judgments, but they may all be dead by now.
So, all I can say is, there are apomorphies of both kinds, as well
as a sizable gray area. If you ask me about specific apomorphies,
there's a ghost of a chance I might be able to answer well.
But alas, no more than that.
By the way, some anatomical judgments are so obvious, it's inexplicable
why anyone could entertain an opposite judgment. For instance, there was
once a wild hypothesis that Megachiropterans and Microchiropterans developed
wings independently. But any look at the structure of the wings,
especially the closeness of the "index finger" to the "middle finger"
and their relative lengths, should convince anyone of the silliness of
this hypothesis.
> More importantly, what good does it do us to identify some species as an
> ancestor candidate? Is science thereby advanced in any way?
Now THAT is a no-brainer. Scientific pedagogy is advanced immeasurably
when one can point to a fossil, and assure students that it is an
ancestor-candidate by the above criteria. "A picture is worth 1000 words,"
and here the case is even better. Only students that ace a course
in comparative anatomy could master the technical terms for the
various features that the LCA ought to have, and even they might
never make the effort to try and figure out a complete description.
And even if one does, less nerdy students might meantime learn about
twenty ancestor candidates, and have a much better start on being
paleontologists for the love of it, rather than just for the sake
of making a living or having others to hobnob with.
> > But your cladophilia is so dogmatic, it refuses to accept even such
> > concepts as being helpful.
>
> You might have a point if such concepts were actually helpful. Can you
> explain the helpfulness?
See above.
Will you post some snarky comment like "I looked, and didn't see
anything."?
> >> There is no such thing as an "ancestral group".
> >
> > Oh, but "stem X" includes what traditional systematists would call
> > ancestral groups to X. The closer they are to "species" in the
> > Linnean hierarchy, the more valuable they are to someone wanting
> > to learn about the tree of life.
>
> "Stem X" isn't a group.
"Total X" is a group to the extent that Christine Janis claimed
that she was correct in making *Gnathostomata* refer to the total
group without explanation, even though the term officially designates
a crown group.
And since Total group = crown group + stem group, you seem to
be claiming that "real + unreal = real". Or do you want to
take issue with what Christine claimed?
> Why are they valuable?
Extension of what I wrote about ancestor candidate. Students
and even researchers could greatly benefit by being told that the
LCA is as surely in the subfamily ____________ as X is "sister group"
to Y. And I'm sure you've seen lots of generally accepted choices
for X and Y.
The extension becomes complete if students/researchers are referred to some
good reproductions of representative fossils in subfamily __________
so they can get a really good feel for what the LCA should be like.
Concluded in next reply to this post, but not until Monday, if then.
It's time to hit the sack and begin my usual weekend break from posting.
Peter Nyikos
Professor, Dept. of Mathematics -- standard disclaimer--
University of South Carolina
http://www.math.sc.edu/~nyikos/
nyikos @
math.sc.edu