Re: Cretaceous taeniodont -- long, combined answer

[John Bois <jbois@umd5.umd.edu>]

On Tue, 13 Apr 2004, David Marjanovic wrote:

> > Is there a fossil bias against finding large mammals?  I'm assuming not.
>
> There is
> 1. a fossil bias against finding _anything_. That's a BIG bias.
> 2. a fossil bias against finding anything terrestrial. I know of 2 (two)
> places in the _world_ that record the start of the Cretaceous on land, for
> example -- Purbeck and Cherves-de-Cognac.
> 3. a fossil bias against finding anything terrestrial smaller than a
> sauropod femur. Isn't all that big, but...
> 4. ... there's a perception bias against finding anything terrestrial
> smaller than an *Allosaurus* skull.
> 5. There's also disarticulation. Very, very few Mesozoic mammal sites have
> yielded mammal skeletons. The normal thing to find is "the tooth, the whole
> tooth, and nothing but the tooth". You know the size of your molars. There
> are sites where a tooth of half that size would shine dark brown-black (like
> Cherves) or white (like the Gobi), but in many places such a thing is quite
> inconspicuous.

Still, all other things being equal, one would expect to find bigger
things (i.e., teeth of larger mammals) more easily than small.

> In other words: We are _not_ going to be able to make statistically
> meaningful statements about the abundance of big mammals over time.

I wonder why we keep looking, then?

>... fossils are not equally but
> _randomly_ distributed between the strata.

That's the same thing--if you sprinkle a thousand salt crystals over a
meter square pan at random, you can sample a couple of areas, multiply
by area, and come up with a thousand.  I mean, in statistics, we make
solid assumptions based upon randomly distributed data all the time.

> We do not know if we have a parallel size increase, a parallel size
> _decrease_, or _any_ trend _at all_ here. The sample is _too small_.

So, back to my original qustion: at what point can we say anything at all?
Another way of saying it, what statistical test did you use to conclude
the data cannot reject the null hypothesis of no trend?

> Predatory birds? Yes, there is evidence for a clade of predatory birds in
> the Mesozoic. That clade is called Avisauridae, and it starts (at least)
> with *Cuspirostrisornis* in the middle Early Cretaceous -- you would need it
> _far_ later.

As I understand it, birds have by far the worst fossil record--so, back at
ya with your not enough data argument.  I frankly like my argument better,
viz., old birds replaced by new birds for reasons having nothing to do
with asteroid, but, rather, adaptations.  And, adaptations that worked for
this regime change, would also influence the entire biota.  But I think
we've already agreed there is not enough data to resolve this.

> Pterosaur decline? First show me that there _was_ a pterosaur decline. The
> fossil record of LK pterosaurs is simply too bad to judge _this_ at the
> moment. In other words: There may not be anything for you to explain!

Pterosaurs were plentiful and gradually declined over the Cretaceous with
only a few species appearing in the record of the latest K.  Are you
suggesting they enjoyed a last minute rebound at the latest K only to be
KOd by asteroid?

> > If size gave immunity from predation, why didn't this fuel an arms
> > race as it seems to have done in dinosaurs.
>
> Because there was a big heavy lid on this arms race. A too large mammal
> would have had to evolve into an occupied niche. There may or may not have
> been such an arms race, but it was finite if it was at all.

Good point.

> > you can sustain more predation because
> > your reproductive success is better.

> Your reproductive rate is _worse_ when you're bigger.

Size alone is not the determining factor, e.g., sauropods had bigger
clutch than elephants.

> At the K-T boundary, it seems... Enantiornithes and Ichthyornithidae, at
> least, survived to the end.

And became extinct due to as yet unknown forces.