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

Re: New Cretaceous bird and other papers

> > But without a big mass extinction in between, unlike the K-T. Fossils
> > regularly show enormous radiations after mass extinctions, molecular
> > _never_. Sort of suspect, no?
> Maybe I'm misunderstanding your point, but morphological change (as seen
> in radiations) should not necessarily be reflected in genes used for
> molecular clocks.  Usually (I think) clock genes are relatively
> stable--not subjected to selection--and indfependent of morphological
> change (e.g., mitochondrial DNA).  Otherwise they would be of little
> value.  In this respect how would a mass extinction be recorded in the
> genome at all?

My point is that such divergence times never indicate closely spaced
divergence of lots of clades. From fossils one would expect that most extant
big clades of placentals and neornithines diverged in the Paleocene and
early Eocene, but molecular divergence times don't cluster in this time,
instead they are smeared out over pretty much the entire Cretaceous.
Assuming that the genes used for molecular clocks are reliable (not
subjected to selection, as you write, respectively subjected to a constant
amount of stabilizing selection over hundreds of Ma), then explosive
radiations should be detectable by clustered divergence times.
        AFAIK molecular clocks yield relative times from which absolute ones
are calculated by some calibration points from the fossil record, such as
the basal divergence of amniotes 310 Ma ago. Who knows what's wrong with the
calculation methods*. Often molecular clocks are applied to questionable
phylogenies, thus artificially extending all divergence times. Add to this
the impressive error margins.
        When a mass extinction has created an empty world without
competition, it is conceivable that every somehow viable mutation can found
a new species and a new big clade. This may mean that species with wildly
mutated molecular clock genes can survive and acquire further mutations that
restore _full_ functionality millions of years later, which in turn will
pretend grossly inflated divergence times. But I'm not sure whether this
scenario is even testable.

* This may just be a difficult way of saying that _I_ have practically no
idea about how it's done. :-)

> > But assuming almost the entire diversity of Neornithes in
> > the Cretaceous is IMHO a stretch.
> A bigger stretch is imagining how a limited neornithine diversity in LK
> could survive while the _total_ diversity of enantiornithes is knocked out
> by bolide magic bullets.

It is IMHO no stretch to assume that...

...there were enants everywhere in the world except in Antarctica in the
Maastrichtian. Evidence: All known birds from the Maastrichtian site on
Seymour Island (next to the Antarctic Peninsula) are Neornithes.
...the effects of an impact depend on latitudinal distance. Then Antarctica
is as far as you can get.
...100 % of non-neornithines died out, along with (guesses) 99 % of
neornithine species and 99.999 % of neornithine individuals.
...100 % of Neornithes died out everywhere north of (guess) 20° south.
Allows for survival of ratites in New Zealand, Madagascar and India as
predicted (but maybe not necessary, assuming that the ratites of that time
were not flightless) by the newest mtDNA phylogeny of ratites.