Re: Late night thoughts: Pathetica and Interspersal

[David Marjanovic <david.marjanovic@gmx.at>]

Currently there's so much traffic that I sometimes receive replies before 
the originals. So I'll only answer this one post now. I don't have oodles of 
time anyway...

----- Original Message ----- 
From: "Tim Williams" <twilliams_alpha@hotmail.com>
Sent: Wednesday, June 20, 2007 7:32 AM

> No, David didn't say that at all.  What he said was that after a dramatic
> increase in size that occurred early in their evolution (from
> _Anchisaurus_-sized to _Isanosaurus_-sized and beyond), sauropods had
> essentially hit their maximum size by the Middle Jurassic.

No, not that either. I said that after a drastic increase in the size 
_range_, into both directions, with -- in sum -- _no trend_, the smallest 
ones (like *Anchisaurus*) died out, and the bigger ones kept diversifying, 
though neither the minimum nor the maximum size changed much from the MJ 
onwards.

Again: there is no trend in sauropod size over the Mesozoic, and there is no 
trend in sauropodomorph size over the Mesozoic, according to my unpublished 
results.

> Within each of the major sauropod lineages (diplodocoids,
> brachiosaurids, titanosaurs) average size
> appears to have peaked, and then gone downhill.

No, not even. There is no trend. :-)

> For example, although the
> Titanosauria produced some extremely large taxa in the later Cretaceous
> (e.g., _Argentinosaurus_, _Puertasaurus_) the average is dragged down by
> the appearance of a slew of much smaller titanosaurs (e.g., saltasaurids).
> This is what Carrano found.

Yes. I only found a statistically not significant decrease, perhaps because 
I used fewer immature sauropods as data points, so I found no statistically 
significant trend, only diversification and extinction.

> You may argue that certain sauropods got bigger as a
> defense mechanism, which might well be true.  But this argument of
> "directional selection" cannot be upheld as a general overarching
> theme for sauropod evolution.

That's because directional selection would result in a trend, and there is 
no trend, unless my analysis is flawed.

> If your hypothesis is that sauropods were driven to large body size by 
> theropods (and/or vice versa), you must demonstrate a correlation.  Such a 
> hypothesis makes a prediction that there is a correlation between 
> increased body size in theropods and sauropods.  Thus, it can be tested. 
> If such a correlation is demonstrated (quantitatively) to exist, then we 
> have room to speculate on causation.

The method I used takes the phylogeny into account and compares descendants 
to their reconstructed ancestors. If repeated size increases and extinctions 
had happened, I'd have found that as a trend of size increase. I didn't.

Hone et al. did not reconstruct the sizes of ancestors and just compared 
younger relatives to older relatives. They, too, didn't find a trend.