Re: Hard science of evolution


"John Wilkins" <j.wilkins1@xxxxxxxxx> wrote in message news:1i2og84.12edoew5l1dv2N%j.wilkins1@xxxxxxxxxxxx
Perplexed in Peoria <jimmenegay@xxxxxxxxxxxxx> wrote:

"John Wilkins" <j.wilkins1@xxxxxxxxx> wrote...
TomS <TomS_member@xxxxxxxxxxx> wrote:

"On Fri, 10 Aug 2007 08:52:20 +1000, in article
<1i2mg54.1ktzwuz1q1fvq4N%j.wilkins1@xxxxxxxxx>, John Wilkins stated..."

Perplexed in Peoria <jimmenegay@xxxxxxxxxxxxx> wrote:

And I am a bit surprised to see you trying to conflate populations and
types.

Any part of a population that is not a single individual is itself a
population (in this case of a certain genotype).

Excuse me for being pedantic about this, but this seems to be
overly broad. Is {Torquemada, Cleopatra} a population? Why
exclude single-member populations, when a single individual
can give rise to a population?

Because you are dealing with types, rather than individuals. The set you
ostensively define here is not a typed set, unless you state the
criterion for inclusion without reference to Torquemada or Cleopatra. If
there is a type that defines a set that has one individual in it (a
singleton set) then it is a population, although a very uninteresting
one from the perspective of population genetics.

I agree (apparently with everybody) that the set {Torquemada, Cleopatra}
is neither a type nor a population. Types and populations are collections
of individuals that maintain their existence over time with a changing
cast of individual members. A population or type at any particular time
has a specific set of individuals that compose it. But the population or
type is not defined by this set, instead, the definition of the type or
population determines the set.

Agree, so far.

The difference between a population and a type is that a type is a part of
a species defined by criteria of shared genetic or phenotypic features.
But a population is usually defined geographically or by restricted gene
flow. Members of different populations rarely interbreed. That is not
usually taken to be the case for members of different types.

Here I disagree. While this is usually true, it is not, I think, the
definiens of "population". You are conflating the explicanda with the
explanans.

Maybe I don't understand "explicanda" and "explanans". I don't see that
anything is being *explained* here. I *am* probably conflating the typical
usage with the definition.

In an asexual species, this difference between the usage of 'type' and
'population' cannot be maintained. But it is the main difference for
sexual species.

Here is vociferously disgaree.

Well, you vociferously misocnstrue, anyways.

Asexuals can be defined in terms of
shared niches/adaptive peaks/hosts. On a genome-space cluster
definition, they can be defined as being not tooo far, Hamming-wise,
from a median genome (the "wild-type). I have a paper on this
forthcoming.

What I intended to say here is that when dealing with asexual species,
the distinction in usage between types and populations tends to evaporate
away.

But now I begin to understand why you raised the explicandum/explanans
issue above. You see types and populations as real things which can
be identified in nature. I am talking about their use in modeling
evolution of bigger things (species, for example). In the usages I was
thinking of, you divide the population into types or sub-populations
only as a technique for understanding the dynamics of the whole. If
you partition well, you have a good model; if you partition badly, you
don't. If you call your groupings types when you should call them
(sub)populations, or vise-versa, then you create confusion in your
audience, though your model may still be mathematically valid and in
conformance with the thing modeled.

PS. On rereading the above, I fear that I will be misinterpreted as
denying that types can be identified empirically. I am not saying that
exactly. Certainly if a property of interest is identified, it will
often be the case that the organisms will naturally partition into
types based on this property. But as a modeler, rather than an observer,
I have to point out that the choice of the property of interest is
arbitrary. There are so many properties that might be chosen that
when you take the cross product of the partitionings produced by each
property, you end up with a set of types larger than the number of
organisms.

PPS. I would ask around before using the term "wild-type" to refer to
a 'median' genome. In fact, I would recommend that you avoid the word
'median' to refer to genomes.

.

Relevant Pages

  • Re: Hard science of evolution
    ... a species defined by criteria of shared genetic or phenotypic features. ... A population is not the same thing as a "restricted gene flow" or a ... from a median genome. ... you partition well, you have a good model; ...
    (talk.origins)
  • Re: Human Endogenous Retroviruses in the Primate Lineage
    ... spread of so-called endogenous retroviruses. ... amplified in the genome and profoundly altered genome ... genomes of species in the evolutionary lineage leading to Homo sapiens. ... These elements appear to be "selfish" sequences that just exist to make ...
    (talk.origins)
  • Re: Can there be a mathematical proof of evolution?
    ... no valid hierarchy of bacteria. ... Given the FACT that eukaryotes don't replicate like bacteria do, ... and boots according to the *new genome*. ... Species A and Species B. ...
    (talk.origins)
  • Re: The incompleteness of the genome [was: No "Macro" Evolution]
    ... > people and the species bar-coders who claim that genome analysis is ... >>>Nobody is proposing, for example, that a given species can arise twice, ... > -- without evolution we cannot define species. ... then genetics is the root cause of the species ...
    (talk.origins)
Loading