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


--Original Message-- From: Allan Edels <Edels@email.msn.com>Date: 06
November 1998 08:13

First of all Allan, thanks for replying.  OK - handshake over - let battle
commence! . . .

>    Sorry, John -  I don't buy it.  Your statistics are dealing with (or
>more precisely NOT dealing with) a difficult-to-quantify factor, namely the
>likelihood of fossilization of small, rare fossils in often barren
>of sediment.

The entity under consideration is: "Potentially Findable Fossils" -  a
compound of existing fossils of that group, availability, and chance
investigatory bias.

>Jurassic exposures are more scarce and less examined than
>Cretaceous.  (I am NOT including the Morrison formation here - anyway,
>exposures are for a relatively limited timeframe, in comparison with the
>rest of the Jurassic).  Certainly, the Jurassic has only recently been
>examined for the smaller creatures (AND here I exclude Solnhofen -  so sue
>me :) ).  We need to look at 180 -150 mya, and we need really, really good
>fossils!  We need Solnhofen or Liaoning class preservation!

If it's a combination of preservational and investigatory bias that has led
to all that group's remains being found in the K, how come there is an
enormous bias the other way for small pterosaurs?  Is it 90% of
(post-Triassic) small non-marine deposited pterosaurs found in the J, or
100%?!  Also, why do all the J manis/arctos have to be small?  Actually, I
don't mind that - there *were* some things close to that group - apprentice
Archaeopterids - small, arboreal olshevskyans not quite yet having converted
fur to modern feathers - but I'm happy with that.

>    The problem that you do deal with, i.e. the time problem, is a
>one.  You fault Cladistics because it doesn't deal with time relationships,
>and I agree it is an annoying lack.

I don't really fault it for not being able to incorporate time into its
processing, rather for producing temporally paradoxical results (amongst
other faults!).

>    The fact that we find bunches of Maniraptorians in the Cretaceous, and
>one earlier bird (at the end - almost - of the Jurassic) does not mean that
>they did not share a common earlier ancestor,

Probably not in terms of certainty, but certainly in terms of probability.
(We can never be certain of vert. palaeo family trees).  But just having the
notion mentioned occasionally - that would be nice to be getting on with.

>especially since we have not examined the preceding
>10 -30 mya before _Archaeopteryx_ in sufficient detail.

Oh, I think we have!

>(I expect arguments
>on this - tell me I'm wrong!).

(See my accompanying posting.)

>We find birds in the Cretaceous - not a hell
>of a lot, but quite a few - are these descendants of a known Maniraptor
>merely because they occur after
>it?  Or perhaps all modern birds are late Cretaceous Maniraptorian
>descendants, since they occur after the K-T boundary (excluding the
>mentioned parrot), and are different than _Archaeopteryx_ and other early
>birds of the Mesozoic - i.e. the enantornithine birds?  These are patently
>ridiculous statements, but, if one looks at _just_ the time relationships,
>then these statements are likely outcomes.

Granted, "later" than another member in a group by no means implies "from".
However, what about "later than the *first* known member of the group"?
Still not certain of course but a lot more possible.  (We can never be
certain - but why oh why won't the supporters of the contrary notion stop
implying it *is* certain!!).  For us to suspect some creature were close to
the ancestral member of the group it would need something special about it.
Might not the first feathered flier fit in well with the ancestor of the
most successful group of K theropods?

>    As I said the time problem is a difficult one, and I don't have a very
>good answer for it, except to reiterate the need for more fossils in the
>sweet spot (ie. 10-30 mya prior to _Archaeopteryx_).  You are essentially
>supporting George O.'s BCF theory, which, as I've said before, is a nicely
>constructed logical theory - missing only sufficient evidence to back it up

...just like the opposing theory!  No-one has evidence for *any* J precursor
to Archae apart from a few unusually small teeth which could be anything.
Apart from cladistics and historical accident, there is no evidence that the
direction of relationship is _Velociraptor_ -> Archae; and time would seem
to point the arrow the other way.

>{By the way, the only fossil that I think comes close to beginning to build
>evidence for BCF is _Caudipteryx_ - I think that it was secondarily

It's very nice to hear you say that Allan, but I have to say I don't think
it's safe for you to stand there!  Where do you now draw the line?  Between
_Caudipteryx_ & _Protarchaeopteryx_?  Between them and _Oviraptor_?

>Even George has said that BCF versus BADD (or BAMM) would most
>likely be very subtle - even with the best fossils.

Oh yes, in some ways they're very close.  It's so hard to tell for sure,
isn't it?  No more than an vague feeling one way or another.  You could
easily cross the line without hardly realising you had done it - and you
need never look back.  And if you did, all you would see would be a long
line of people slowly getting up and following you! :-)

>    A modest suggestion - perhaps difficult to enact, but I think it would
>be interesting to do:  Can we find a way to add time relationships to
>Cladistic analyses - essentially, weighing the results of a Cladistic
>analysis with the time relationships of the samples?

Not a question aimed at me perhaps, but by the time you've added all the
complications and amalgamations, you've got a massive stew of everything and
anything.  How could you reasonably justify it theoretically?  Who would
have the cheek to attempt to calibrate such a monstrosity?  Non-cladists try
to combine all that in their heads of course, but at least we have the
honesty to call it guessing!

>    Everyone knows that you can do nearly anything you want with
>depending on how you load them.

I'm sorry, I thought I heard someone snorting as you said that - did you say
"statistics" or "cladistics"?

>(I remember calculating the statistical
>death of the universe - based on entropy - of course, I don't believe it
>will hold true - but it does work).

It's sometimes hard to trust statistics, but I believe they can be made to
do useful work.

> - beautiful, elegant theories are not necessarily true.

True.  Though I don't claim much of those qualities in my latest offering.
I would claim this though - it makes better use of the principle of
parsimony than advanced cladistics!