Natural Science Forum / Biology / Evolution / February 2007

The content seems OK.  A more conventional format - with abstract,
conclusion, references and links might help.

Here's a relevant link, for example:

http://www.blackwellpublishing.com/ridley/classictexts/haldane2.pdf

Tim Tyler <seemy...@cyberspace.org> wrote:

One need go no farther than Haldane's own
writing to debunk "Haldane's dilemma", and it
doesn't take some fancy new mathematical analysis to
do the trick, either.

Speaking of the peppered moth, Haldane says:

       Now if the change of environment had been so
       radical that ten other independently
       inherited characters had been subject to
       selection of the same intensity as that for
       colour, only (1/2)^10, or one in 1024, of
       the original genotype would have survived.
       The species would presumably have become
       extinct.

Yet from Wikipedia

 http://en.wikipedia.org/wiki/Peppered_moth

we learn:

       The female lays about 2,000 pale-green ovoid
       eggs about 1 mm in length into crevices in
       bark with her ovipositor.

So with the usual two surviving offspring to replace
her and her mate, the peppered moth is _already_
undergoing 1000::1 selection pressure, and the
species has _not_ become extinct.

Thus Haldane, right from the start, was working from
entirely false premises, that intense selection over
a few generations would extinguish a species.

All that the ten selection-susceptible
characteristics he proposes would have accomplished
is to modify (with some factor for the actions of
blind luck) just _which_ two (on average) of the
eggs survived to be a mating pair of moths.

No "excess deaths", no threat to the species, just
nature acting as nature in fact does act, ruthlessly.

FWIW

xanthian.

Tim Tyler <seemy...@cyberspace.org> wrote:

That doesn't particularly matter; Haldane's claim
was that 1024::1 selection pressure would extinguish
the species, yet that is in fact the species' normal
mortality, so all that would happen is that what you
claim next to be "random" mortality would instead be
mortality as directed by natural selection, but it
wouldn't be _changed_ mortality.

Why?

Because as fast as the absolute count of peppered
moths dropped due to selection pressure, it would
rise again due to excess abundance of resources from
that diminished population level, as better fed,
healthier moths produced
excess-to-normal-expectations offspring.

No "excess deaths" would occur, demolishing the
basis for Haldane's entire argument.

But that's nonsense.

It might apply in the case of the eggs of a single
moth, but it would not apply in the case of the
breeding cohort of the moth species.

Haldane wasn't describing some generic species, he
was describing a specific species with a specific
known selection event.

These weren't roughly cloned individuals, like
cheetahs are, there was in fact variation against
which natural selection could and did work.

To claim that the mortality of the entire species
was "random" is to deny evolution entirely.

You might see being eaten as eggs as "random", but a
species some of whose members laid bright yellow
eggs self-flagged "poisonous as hell" would see it
as the punishment natural selection imposed on the
rest of the species for laying moss green eggs
clearly self-flagged "delicious and nutritious".

Again you've missed the point here, which was that
Haldane presumed a level of mortality supporting
simultaneous "50% mortality each" natural selection
for ten characteristics would be one that
exterminated the species, when it was in fact, that
species' usual mortality within a couple of percent,
for that particular exemplary species he had chosen
as his starting point for his argument.

Beginning with that false premise, the rest of his
argument fails in turn.

The calculations we're seeing from talkorigins here
ignore another characteristic of natural selection,
which is that it starts with what it's got, it
doesn't have a "pull needed mutation" functionality.

Thus, the usual starting point is a variation which
up until the new selection pressure is neutral _and
probably widely spread in the breeding population
already_.

Same for ten variations. Natural selection _isn't_
starting with ten instances of one extant variation
each, that's a complete misconception.

It is starting with a population where the neutral
variants have had _the entire lifetime of the
species up to the point of the new insults_ to have
been created by mutation and spread by genetic
drift, and in the most likely case, being truly
neutral, are some substantial fraction of all the
alleles of the particular gene locus, probably a
fraction quite close to 1/"number of long existent
neutral in relationship to each other alleles at
that locus".

Haldane also assumes incorrectly that the _normal_
situation is rapid environmental change. Asteroid
impacts and volcanos losing their tops excluded,
most environmental changes are fairly slow. It is
one of today's tragedies that we've pushed climate
rate of change far past its historic values, and
_now_ species are having trouble keeping up with
that change, which managed nicely to keep up with
normal change rates; but that just shows that the
rapidity of current change is _not_ the norm, the
opposite to what Haldane posited.

FWIW

xanthian.

On Feb 22, 2:47 pm, "Kent Paul Dolan" <xanth...@well.com> wrote:

Obviously, bacteria and viruses can take that kind of selection
pressure and survive, but we're also discussing mammalian evolution
here.

No, his calculations still apply in that case.  The "excess deaths" in
this case are the deaths of the moths who died for lack of the
beneficial allele.  Under extreme selection pressure (s=1, and low p-
values for the beneficial allele), you end up killing about 1x the
entire population (1.65 in diploid organisms).  The fact that a
species has high birthrates allows it to bounce back very quickly, but
that doesn't mean "no excess deaths" would occur.

Haldane does talk a great deal about moths in his paper, but it's not
true that his ideas and calculations apply only to moths.  His
calculations apply to fast-breeding species (like moths), but that
they have a large breeding excess, so it's not as interesting for
those species.  He says: "The number of loci in a vertebrate species
has been estimated at about 40,000.  'Good' species, even when closely
related, may differ at several thousand loci, even if the difference
at most of them are very slight.  But it takes as many deaths, or
their equivalents, to replace a gene by one producing a barely
distinguishable phenotype as by one producing a very different one.
If two species differ at 1000 loci and the mean rate of gene
substitution, as has been suggested, is one per 300 generations, it
will take at least 300,000 generations to generate an interspecific
difference.... Zeuner after a very full discussion of the Pleistocene
fossil record, concluded that in mammals about 500,000 years were
required for the evolution of a new species ... The agreement with the
theory here developed is satisfactory" (page 521-522)  Additionally,
there is no reason to conclude his model would apply only to fast-
breeding species like moths, but not to mammals.

No, he's claiming that deaths that are random do not go towards paying
the substitution cost, and that there is probably some proportion of
deaths are essentially random.

But, if the selection pressure is for dark-winged moths, and the eggs
of light-winged moths versus dark-winged moths are indistinguishable,
then those deaths are random.

No, because his model could still apply to slower-reproducing
species.  Further, while he talks a great deal about moths, he makes
it very clear that it's not just about moths.  He mentions mammals and
vertebrate evolution.

His calculations still work out under weaker selection pressures.
Using very strong selection pressures, you can drive down the
substitution cost, but a weak selection pressure (say s=0.01 or
s=0.001 instead of s=0.1) results in slightly higher death tolls
(about +5%) over the long term.  The fact of the matter is that using
a weaker selection pressure causes fewer deaths per generation, but it
takes many more generations before the beneficial allele reaches
fixation, resulting in a total substitution cost that is nearly the
same in both cases.

http://www.blackwellpublishing.com/ridley/classictexts/haldane2.pdf