Re: Genealogical Momentum
From: Joe Felsenstein (joe_at_removethispart.gs.washington.edu)
Date: 02/25/05
- Next message: Jim McGinn: "Edser: semantic insanity."
- Previous message: Robert Karl Stonjek: "Article: 'Raptor' dinosaurs roamed far and wide"
- In reply to: Claudius: "Re: Genealogical Momentum"
- Next in thread: Claudius: "Re: Genealogical Momentum"
- Messages sorted by: [ date ] [ thread ]
Date: Fri, 25 Feb 2005 02:24:39 -0500 (EST)
In article <cvm8ev$1f9b$1@darwin.ediacara.org>,
Claudius <claudius_denk@yahoo.com> wrote:
>
>Joe Felsenstein wrote:
>> As you may realize from the past discussions here, the
>> basic question between McGinn and everyone else is whether
>> Hamilton's formula for kin selection is wrong. McGinn
>> implies that if there is no different behavior of pairs of
>> genes, whether they are the same due to IBD or the same
>> by (say) parallel mutations creating identical alleles,
>> then Hamilton's formula must be wrong.
>
>Yes, because R (as defined, r = genes IBD), which
>is an essential part of Hamilton's formula, would
>be a false distinction. IOW, it's a phantom
>quantity. We are left with no choice but to
>exclude it from the formula.
If you have two individuals who are relatives, and you want to
know what is the probability that genes sampled from both carry an allele,
the correct calculation is to use Sewall Wright's 1921 formula which
is p^2 (1-f) + p f, as I mentioned. So there is a real and
legitimate use of Wright's quantity f. And that is how it is
used in Hamilton's work. So there is no basis for excluding it
from the formula.
I would answer McGinn's first question "no", his second "does not
apply" and his third by making the above point. Hamilton's model
specifies that the individuals benefited are partially relatives
of the one doing the behavior. To work out the net effect on the
frequency of the allele, we then need to know the joint probability
that both altruist and beneficiary carry the allele, which brings
in Wright's quantity f very naturally, without any need to think
that the amount of the benefit depends on the f.
>You seem to be making a lot of general comments
>here. I was specifically addressing the issues
>that emerge in the context of McGinn's three
>questions.
Why did he ask those questions? You yourself just indicated
that they persuaded you that the "r" in Hamilton's formula
should not be there. So the issue was, as always, Hamilton's
formulas.
>> If you feel that McGinn's argument invalidates
>> Hamilton's, could you perhaps explain why?
>
>I don't think McGinn's argument invalidates
>Hamilton's Rule. Keep in mind that I answered
>'yes' to the first question.
It is unclear why you would *not* think McGinn's argument
invalidates Hamilton's, if you think Hamilton has no basis
for using the quantity r. So I think you have stated two
contradictory positions here.
>> In the arguments in this newsgroup, spin is often
>> visible. But it's the spin of arguments, not molecules.
>
>If Hamilton is any example, the "spin of arguments,"
>are not limited to this newsgroup. For me it wasn't
>until you posted his words (in that other thread the
>title of which I will not mention) that it became
>apparent that Hamilton's arguments were nothing but
>spin. It seems that in science argumentive spin is
>only a problem when it remains invisible.
Well, all I can say is that a couple of generations of
population geneticists and animal behaviorists have
found Hamilton's argument to be clear, and they have
led to much good work. I disagree with you, obviously.
My two initial postings in "Logic of kin selection" show the
Hamilton formula successfully at work in a particular
case. Do you think my arguments there are wrong?
----
Joe Felsenstein joe@gs.washington.edu
Department of Genome Sciences and Department of Biology,
University of Washington, Box 357730, Seattle, WA 98195-7730 USA
- Next message: Jim McGinn: "Edser: semantic insanity."
- Previous message: Robert Karl Stonjek: "Article: 'Raptor' dinosaurs roamed far and wide"
- In reply to: Claudius: "Re: Genealogical Momentum"
- Next in thread: Claudius: "Re: Genealogical Momentum"
- Messages sorted by: [ date ] [ thread ]
Relevant Pages
|
