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

Fwd: Polytomy question (and sickle claws)

David asked me to forward this...

Jordan Mallon

BScH, Carleton University
Vertebrate Palaeontology & Palaeoecology

Paleoart website: http://www.geocities.com/paleoportfolio/
MSN Messenger: j_mallon@hotmail.com

---------- Forwarded message ----------
From: David Marjanovic <david.marjanovic@gmx.at>
Date: Oct 26, 2005 4:48 PM
Subject: Re: Polytomy question (and sickle claws)
To: jordan.mallon@gmail.com

Could you please forward this to the list? For, presumably, some reason
Listproc seems to annihilate my messages without telling anyone. Mickey
Rowe is investigating.


> > What I find interesting is that every evolutionary
> > change is always modeled as a split.  How about
> > species that simply change over time without spliting
> > into two different lineages?  This is rarely captured
> > in any phylogenetic tree.  Is it because of lack of
> > evidence?  Would we know them if we see them (not me
> > of course but you get the point).

We would not know them if we saw them. We could only suspect, more or less

If a cladogram shows two taxa as sister-groups, if one of them completely
lacks autapomorphies ( = derived features of its own), _and_ if its age
fits, then it _enters consideration_ for being an ancestor of its "sister-

However, here we must stop. All we have here is negative evidence. We can
find out that something is _not_ the ancestor of something else by finding
its autapomorphies; we cannot do the opposite. (If we don't find
autapomorphies, this doesn't mean they weren't there. We might overlook
some; more importantly, we cannot find autapomorphies that aren't

This is not a failure of cladistics, it's a failure of science as a whole.
In other words... if you're talking about late Pleistocene diatoms found
in deep-sea sediments, go ahead assuming you have found an ancestor
because it fits the above criteria; but if you're talking about Mesozoic
vertebrates, I recommend to forget about it. The (known) fossil record is
bad enough that the chances that you've really found an ancestor are
extremely small.

> The most important thing to remember here is: CLADOGRAMS ARE NOT
> PHYLOGENIES.  They're two different things.

Well, not quite. Cladograms _are_ hypotheses about phylogeny, with the
additional assumption that none of the OTUs is an ancestor of any of the
others. In many (though clearly not all) cases, this assumption is
reasonable; cladistics was invented by neontologists for neontologists
(Hennig even stated that it should only be used for taxa of the same time
slice, without ever explaining how thick such a slice would be allowed to
be), and most of the fossil record is bad enough that we can't expect to
find an ancestor.

> Unfortunately, many people seem to miss this point and treat the two
> as the same.  This is a problem that is of special interest to me
> lately, as some of my buddies and myself are looking into what we
> think might be an anagenetic series of mosasaurs in the WIS.

An anagenetic series doesn't need to be an unbranched line of ancestors
and descendants (though it obviously _can_ be). It's enough if each branch
has its autapomorphies in some characters other than those which show the
anagenetic change. In other words, you can't tell a phylogeny from a
pseudophylogeny this way.

On the topic of sickle claws... I still don't understand how they could be
useful in climbing. They are called "sickle" rather than "hook" because
they have a cutting edge. For climbing I'd expect a round, if not outright
flat, ventral surface that can carry weight, not just a pointed tip alone.

Telefonieren Sie schon oder sparen Sie noch?
NEU: GMX Phone_Flat http://www.gmx.net/de/go/telefonie