Guy Hoelzer  wrote 

These incoherent fluctuations are what we call the
> "pattern" of genetic drift, which would not occur in the absence of
> sampling.

I think this is an interesting way of indicating that 
(and how) variation is the result of "sampling."  

> 
> > Secondly, the purpose of the genes copied from
> > the parent have nothing to do with estimating a
> > population. 
> 
> I agree, unless one is comfortable with quite a metaphorical stretch.
> 
> > Thirdly its bit of a stretch, at best, to
> > refer to the genes in the parent as a population.
> 
> Why?  These gene copies are spatially isolated and functionally interacting
> with each other.
> 
> > So in my mind it's a bad analogy, at best and the analogy
> > becomes worse still when you refer to differences in
> > the genes of the offspring from those of the parent as
> > "sampling error".
> 
> This is indeed an unfortunate use of this term, which derives from the use
> of sampling by statisticians bent on estimation.

I agree.

> The term "sample variance"
> is sometimes used instead, 

I don't dispute the point you're making, especially 
with respect to the fact that if there was no "sample 
variance," we'd expect all siblings to be identical.

> and does not carry the semantic baggage of
> "sampling error."  Still, I don't know any population
geneticists who get
> confused by the meaning of "sampling error" in this context.
> 
> > (And even if we overlook all of these
> > conceptual shortcomings it's still plainly fallacious
> > to refer to sampling error as causal.)
> 
> I agree in the physical sense.  The language associated with genetic drift
> as a force is used because drift happens whenever sampling happens, so there
> is the language used is really intended to indicate the association between
> sampling and fluctuation.  I think it is more accurate to think of sampling
> and drift as aspects of a single process, rather than as a sequential cause
> and effect.

I agree.  In fact this is my main argument for why I 
consider the phenomena associated with GD to actually 
be part of NS.  I just don't think it was ever 
designated/established that NS is a process devoid of 
"sample variance."

  The misconception of the logical relationship between sampling
> and drift, however, should not mislead the analytical conclusions of those
> relying on the concept of genetic drift in population studies.  Therefore, I
> don't think the misconception of drift or sampling as a force undermines
> much if any of the empirical inferences that have been made under this
> paradigm.

I agree.  IOW, the measurements that have been labelled 
GD are real.  

> On the other hand, I appreciate the heuristic value of correcting
> this misconception, and the merging of evolutionary biology with physics
> facilitated by fixing such misconceptions.

There's an interesting subject, the merging of 
evolutionary biology with physics.  One might assume 
that the obstacles are only conceptual, but IMO the 
obstacles run deeper, they are psychological or even 
moral.

Regards,

Jim
---
þ RIMEGate(tm)/RGXPost V1.14 at BBSWORLD * Info{at}bbsworld.com

---
 * RIMEGate(tm)V10.2áÿ* RelayNet(tm) NNTP Gateway * MoonDog BBS
 * RgateImp.MoonDog.BBS at 8/10/03 10:39:01 AM