Re: Gould criticism and reviews
- From: rem642b@xxxxxxxxx (Robert Maas, see http://tinyurl.com/uh3t)
- Date: Sat, 01 Mar 2008 14:46:43 -0800
From: John Harshman <jharshman.diespam...@xxxxxxxxxxx>
... what PE is about. It's about events on the species level.
Species (supposedly) remain unchanged throughout their lifetimes.
I find it hard to believe that Gould in his PE writings said any
such thing. Let me offer an alternative statement: The genotype
evolves over time, with different loci constrained to different
degrees.
Most loci are totally unconstrained, and consequently evolve at
nearly the same rate at all times, with variation of rate caused by
general stresses such as mutations caused by radiation or chemical
toxins and times when the cell doesn't have enough spare energy to
invest in DNA repair. But those variations over time are nothing at
all like the variation in rate claimed by PE.
A small minoroty of loci are constrained by adaption. Any
significant deviation from the already-obtained fitness maximum is
quashed. So long as the environment remains the same, these loci
remain in near stasis, which can mean these loci hardly change at
all over millions of years.
Some loci, typically third base of a triple, are only
semi-constrained, where two possibilities are OK and the other two
are harmful. So we need to factor that four-way choice into two
two-way choices, one of which is strongly selected hence in stasis
most of the time, and one of which is free to drift.
Some other loci if changed will result in a different amino acid
which only slightly affects the resultant protein, so these are
only weakly selected. There's a spectrum of selection, all the way
from totally free to drift, to fully constrained. Somewhere along
that spectrum is the boundary between selection pressure dominating
over noise and vice versa. For loci near that boundary, we can
dismiss the case as borderline between the drift and selected
models and not worth discussing in the larger context. So we ignore
some range of borderline cases, and the rest are divided into the
two cases of freely drifting versus constrained by selection.
So, for the set of loci which are constrained by selection, we have
stasis for long periods of time when the environment is stable and
the organisms remain within the same niche, then we have a rapid
re-adaption whenever the environment drastically changes (loss/gain
of preditor or prey or disease, asteroid strike, volcanic eruption,
etc.) or when the population discovers a new niche to occupy. Thus
PE occurs for *those* small fraction of loci, which include the
traits we observe in fossils, while PE doesn't occur for the vast
majority of freely drifting loci. But when looking at the fossil
record, we can't see the effects of the freely-drifting loci, only
those very few of the constrained loci which cause visible changes
in the fossils (the structure of the fossils themselves, or their
locations which show something about the behaviour of the
organisms). So the fossil record shows PE and virtually nothing
else, while any DNA we can scrape up from semi-fossils shows the
other side of the picture, the vast majority of freely drifing
loci.
So would Gould have agreed for the most part with what I said
above? (Do you know of anything he wrote which explained it in a
way similar to how I did just above?) If so, would you agree with
our viewpoint on that topic? If not, if he and I have serious
disagreement about this topic, can somebody cite where he said
something that clearly contradicts what I said above?
Or was everything he wrote sufficiently unclear that nobody can
tell whether he would have agreed or disagreed with my summary
above?
No modern species is particularly old.
How would anybody determine that?? It's my understanding that there
is no way from looking at a fossil to know whether the animal that
made it could theoretically mate with another animal that made a
different fossil (assuming a time-machine to bring the two animals
together for the experiment of course). Thus the biological species
concept is meaningless when comparing ancient fossils for which no
DNA is available. At best we may have a one-sided decision
possible, where we might be able to say that two fossils are so
grossly different in form that it's totally unreasonable that the
two animals that made those fossils could ever have mated even if
they lived at the same time and happened to meet each other. But in
most cases, the variation in form within a living species (today)
is so great that all sorts of "grossly different" fossil forms
can't be ruled out as able to mate with each other on that basis of
being too different. The only case where I am pretty sure we can
say that two populations were definitely not of the same species
(per BSC) is when they in fact lived in the same locale at the same
time, intermixing their micro-locales in a way that implies lots of
encounters between them, yet their fossil forms drifted
independently from each other, showing essentially zero gene flow
between them despite plenty of theoretical oppertunities to
interbreed. But that does nothing to resolve the question of
whether a distant ancestor is the "same" species as the more recent
descendents from that population. I really don't think we can say
with any certainty whether (per the BSC) they were the same species
or not. Thus your statement I quoted must be totally unsupported.
But here we get into the central problem of PE, which is that
species are identified in the fossil record by their characters.
If the characters change much, we call it a different species; if
they don't change much, we call it the same species. Neither fits
well with the biological species concept as applies to extant
species.
It sounds like you agree with me on that point.
We really can't recognize species in the fossil record,
And we agree on that major point too.
so PE boils down to a statement that morphological change occurs
only when there is morphological change.
No, that's not what PE says. Both PE and gradualism use the same
evidence, of a bunch of fossils from one time that look a
particular way, then another bunch of fossils from a more recent
time that look a different way. Gradualism explains it as looking
at two frames from a continuous action movie. PE explains it as
looking at two frames from a nearly-still-shot animation, where
there really is hardly any change at all for a period of time
(stasis), during which we get the first bunch of fossils, then
after a near extinction and subsequent radiation, after the
population had adapted in a new way, and reached a new stasis
*different* from the previous stasis, then we get that second bunch
of fossils.
If we had really good time resolution on fossils we might be able
to tell the difference between the two models, and then PE or
gradualism would be a fact (and the other dismissed) rather than
both being possible theories. Performing regression between
accurate time and morphology within each of the two bunches of
fossils, we might discover either near-zero correlation between
time and morphology, totally inconsistent with gradualism, or we
might discover a clean signal of correlation between time and
morphology, with slopes consistent with each other and with the
longterm trend *between* the two bunches, hence showing clear
consistency with gradualism and apparent refutation of PE. But with
such noisy time data, we can't get any decent regression within a
single bunch of fossils which all were deposited at nearly the same
time to define a single paleo-species. Um... am I correct here?
(I'm not an expert paleontologist, and I'd rather not play one on the net!
But I'm a really bright guy who *might* be correct on such topics.
But I'm very insecure and would like some feedback real quick now.)
.
- Follow-Ups:
- Re: Gould criticism and reviews
- From: John Harshman
- Re: Gould criticism and reviews
- Prev by Date: Re: Letter to the editor: Evolution theory full of holes: student
- Next by Date: Re: Atheists, read this.
- Previous by thread: Re: Gould criticism and reviews
- Next by thread: Re: Gould criticism and reviews
- Index(es):
Relevant Pages
|
