From:    Chris Colby
To:      All                                    Msg #86, Jul-24-93 12:52PM
Subject: punk eek as artifact and my own macro theory

Organization: animal -- coelomate -- deuterostome
From: colby@bu.bu.edu (Chris Colby)
Message-ID: <22s7e2$rlo@news.bu.edu>
Newsgroups: talk.origins

There's been a little talk about punk eek recently on t.o. Since I
hadn't read anything about it for a couple of years, I drug out my
'tempo and mode' folder and browsed through it. I came across one
argument against punk eek that I had forgot all about. I also have
come up with my own theory of macroevolution.

Species in the fossil record are identified by some diagnostic
characteristic. In order to be useful, this trait should be
discontinuous with other morphological traits in related species.
So when Gould and Eldredge state that a lot of morphological change
goes on at speciation and most species level change happens rapidly,
this may be because species are _defined_ in the fossil record as
organisms with a distinct trait (or traits).

Let me give an example to illustrate. Imagine an praying mantis
lineage. We'll follow two traits through time in this (imaginary)
example -- arm length and number of "spikes" (or whatever those
things on a mantises arms are called). Let's say that over time,
arm length increases gradually in several closely related species.
As time goes on, the arm could get long enough that there is room
for one or more spikes to be placed on it. Now, a paleonologist
studying this group of insects doesn't know when speciation
occurred in any of the species; but, following conventions
would call insects with different numbers of spikes each a
new species. He would focus on the discontinuous trait for ease
of classification and ignore the continuously varying trait. Then,
when he was gathering evidence for punk eek, he would look at the
"species" and say -- yep, these characters (spike number) stayed
the same throughout the life of this species, then changed abruptly
at speciation.

I've used a discrete trait here (can't have half spikes), but this
needn't be the case. Imagine that (as before) arm length was
gradually increasing, but body weight stayed the same for long
periods then shot up, then stayed the same for awhile and so
on. The insects would then be divided up into "species" based
on size (which is discontinuous in rate of evolution). But,
there is no way of knowing if the increase in body size was
accompanied by a speciation event or not -- that is assumed.

Any paleo types out there want to comment on this? Nedin?

Here's the reference:

See Levington and Simon, 1980, A critique of the punctuated
equilibria model and implications for the detection of speciation
in the fossil record, Syst. Zoo. 29 (2): 130-142

-----------------------------------------------------------------

And now (trumpets blare) MY OWN THEORY OF MACROEVOLUTION!!!!

I will show that my theory is just as consistent with current
biology as any rival theory, but has one important benefit;
since I'm "publishing" on USENET and not a real scientific
journal, I can easily disavow any responsibility for it by
saying it was just idle speculation 8-)

Many populations exist as a single gene pool over a large
geographical range. But, subdivision occurs due to
uneven distribution of habitats. Small subpopulations can
(and do) adapt to their local surroundings, but gene flow
reintroduces alleles lost to drift or selection. Constant
gene flow (even at low levels) means alleles from all
subpopulations can migrate to all other subpopulations. Alleles
that confer a benefit in some areas may be slightly deleterious
in others, so continual gene flow acts similar to mutation in
depressing mean fitness. Recessive alleles that are not expressed
in the individual that migrates (and his/her children if the
allele is absent in the new subpopulation migrated to) could
especially contribute to this fitness depression.

Now, if a subpopulation suddenly acquired reproductive isolation,
mean fitness would quickly rise as the population flushed out
the alleles that were adaptive in other subpopulations, but
present only due to gene flow in the now isolated population.

How could a population suddenly become reproductively isolated?
Recently, some biologists have shown that many "good" species
in the wild can breed with a closely related species if one of
the two are fed antibiotics. These species (most of the current
examples are in insects) harbor cytoplasmic bacteria that destroys
the chromosomes of non-infected sibling species. When the bacteria
is killed, the barrier to reproduction is gone. It has been shown
(in _D simulans_) that a cytoplasmic incompatibility (CI) factor
can quickly spread through a population.

The quick release from the fitness depression (due to gene flow)
could also trigger more evolution. The population would
be resistant to invading alleles because invading individuals
would not carry the cytoplasmic factor and their matings would
not produce offspring. But, any signal for individuals to pick
local (infected) individuals over migrants would be favored as
it would prevent the carrier from wasting energy on matings
(with uninfected "outsiders") that would not produce offspring.
The signal must be present in a much greater frequency in the
isolated population than in neighboring subpopulations. The
signal could initially be present only in low frequency in the
isolated population if absent altogether in neighboring populations.

If at low frequency initially, this signal will increase in
frequency because individuals mating with others assured to
be members of the isolated population will have a higher reproductive
success than those who mate with non-signal carriers and run the
risk of mating with an outsider.

Thus the initial invasion of the CI factor causes at least one
burst of evolution, the flushing of "outsider" alleles and
possibly a second burst to evolve a mate recognition system
beneficial to individuals in the population. If the signal is
morphological (it wouldn't have to be), this theory would
also predict more morphological change around the speciation
event (although for different reasons than punk eek).

My theory doesn't require the subpopulation to be either small
or peripheral, just isolated enough so that the CI factor
could quickly invade and confer reproductive isolation. (I've
actually glossed over some details about how CI factors work,
but I can elaborate on how it is possible if asked.) The
release from deleterious gene flow and subsequent mating
system evolution are separable -- both need not occur. The
big initial population might not have local adaptations so
the first part wouldn't occur.

Of course, the success of my model won't depend on how well
it meshes with the data. Really successful theories need
splashy titles, like punctuated equilibria. Has a nice ring
to it (and can be nicely shortened into cool slang -- "punk
eek".) Therefore I've decided to call my theory "prophylactic
flush". The CI factor shields the isolated population from
"infecting" alleles and releases it to a "flushing" of deleterious
alleles.

Comments?

Chris Colby
email: colby@bu-bio.bu.edu