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

Re: Dinosauria---Rejected Name?



First, thanks for responding.

My discussion of a single paragraph of your observations, quoted so the
context of each section would be available, wasn't questioning the (to me
self-evident) assertion that 'any two organisms share a most recent common
ancestor.'  I'd also assume you wouldn't argue that every possible clade
must be named; given the propensity for cladistic definitions to include
specific species, you'd end up with unwieldy appendices.
Naming conventions are human constructs with criteria including usefulness
in communication.  They are not in themselves analyses, but names may be
formulated as the result of analysis.

You gave an example:
<The absolute composition of a clade does not change with analysis, just
*our hypothesis* of the clade's content.
Holtz named Arctometatarsalia, and hypothesized that it included o-mimids,
tyrannosaurs and trodontids. He now suggests the latter two groups are not
members of the clade he named. The TRUE content of the historical entity
Arctometatarsalia has not changed, but our hypothesis of its content has
changed, based on exclusive groupings recovered from phylogenetic analysis.>

As I read this, there are animals which HP Holtz decided were linked in a
way important enough to be worth naming, and which was given the name
Arctometatarsalia.
(The 3 sets of animals he included must share a common ancestor, but not all
sets of animals are worth formulating a name for.  These were.)
Now two of these groups are not considered to be well enough linked for the
concept implied by the name to have 'real' importance, so they are removed.
You're right, the 3 groups do have a common ancestor; I'm just noting the
existence of a criterion which determines whether a set of animals is
included or not.
Suppose that for some reason he had removed all 3 groups.  You would then
have a clade with a name, but no group of animals meets the criteria to be
included.  The clade name Arctometatarsalia would, however, continue to
exist.  In that sense, there is stability in the continued existence of the
name, but I'd suggest that the utility of the construct has been
substantially diminished.  Why shouldn't the name disappear, at least until
members can be found?

You also note:
 <The ancestor is absolutely, 100% NOT hypothetical, in the absolute sense.
...   Now, that ancestor's relationships to other species IS a matter of
hypothesis and test.>
We cannot point to that ancestor; a clade's existence does not rest on the
ability to find a specific fossil to be the ancestor.  To the extent that
primitive/derived characters are identified, an implicit description of the
ancestor is developed.  Because that description is the result of inference,
I used the term hypothetical.
Because the group name, or entity name if you like, may be derived from some
attribute of the hypothetical (descriptively) ancestor, this becomes
important.  The attribute which identifies the clade would be considered
_essential_ to the clade, no?

Your argument has to be followed carefully because it's the opposite of
expectation.  I would have expected that any group is named because of
resemblances among the membership, and that finding the resemblance of any
single member was coincidental would call the relationship of the other
supposed members into question until some way to confirm that the other
members were not also related by coincidence could be found.  The existence
of the clade would in this case be controlled by the hypothesized
relationship among the membership.  You certainly set a high value on the
unimpeachability of a construct.
Complicated stuff.