[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]
RE: up!up!and away!
Responses below in UPPER CASE:
> From: Bill Adlam[SMTP:firstname.lastname@example.org]
> Sent: Wednesday, March 24, 1999 10:26 AM
> To: Derkits, Gustav E, JR (Gus)
> Cc: email@example.com
> Subject: RE: up!up!and away!
> Gustav Derkits wrote:
> > Steven J. Gould presented this argument in "Drunkard's Walk"
> > and also at his talk at Dinofest. This is the kind of thinking
> > that led Luis Alvarez to call paleontologists "stamp collectors".
> I'm no expert on philately, but I understand that it does not
> typically involve the derivation and testing of quantitative
> mathematical models. I think you are misrepresenting Alvarez.
OH NO I'M NOT! HE SAID IT! GOULD RESPONDED TOO!
(ALVAREZ MAY ACTUALLY HAVE BEEN QUOTING LORD KELVIN.
I DON'T HAVE THE UTLIMATE SOURCE OF THE QUOTE.)
> > A one-dimensional random walk model does not work.
> It correctly predicts the numbers of species of different complexity,
> does it not? And it does so with a minimum of assumptions.
> NO. IT DOES NOT. IT PROVIDES A VERY QUALITATIVE
"PICTURE" THAT RESEMBLES THE SPECIES COMPLEXITY
BUT BEAR IN MIND THAT THERE IS NO UNIVERSALLY-AGREED-UPON
INDEX OF SPECIES COMPLEXITY.
> > The random walk that constitutes evolution is a kind of diffusion
> > in two related spaces: sequence space (Manfred Eigen) and
> > phenotype space (Dawkins, Maynard-Smith)...
> Obviously evolution isn't a random walk at all, because of selection.
> But Gould's model incorporates the null hypothesis that there is no
> consistent selection for greater or lesser complexity. As the
> observed facts fit the model, there is no evidence to reject the null
I AM WRITING SHORT EMAILS. OBVIOUSLY EVOLUTION
HAS FEEDBACK LOOPS THAT STRONGLY FILTER THE OUTCOME.
HOWEVER, NOTE THAT "JUNK DNA" OR NON-CODING DNA
DOES EVOLVE BY RANDOM WALK.
> [some evolutionary genetics snipped]
> > Small bacteria have about 10-100 kB of genetic information.
> > Humans have about 600 MB and we do not have the
> > largest number of genes(I think a mudfish does). There
> > has been a gradual increase in the number of genes
> > in the most complex living beings and in the information
> > stored in those genes as time has gone by. The "force"
> > driving the enlargement of the genome is entropy
> > and polyploidy makes it possible to access the sequence
> > space into which life expands.
> What do you mean why you say entropy is a "force" driving the
> enlargement of the genome? Do you mean that genomes expand and shrink
> at random, just as Gould claims phenotypic complexity does? Or do you
> mean that entropy forces genomes to expand (in which case, why have
> the majority not expanded)?
I KNEW THE WORD "FORCE" WOULD CAUSE PROBLEMS, WHICH IS
WHY I PUT IT IN QUOTES. THE DRIVING FORCE BEHIND COMPLEXITY
INCREASE IS THE SAME THAT CAUSES RAINDROPS TO EVAPORATE
- THERE IS AVAILABLE SPACE. THE THINGS THAT REDUCE THE EFFECTIVE
SPACE ARE FITNESS AND ACCESSIBILITY. UNTIL ARTIFICIAL LIFE REALLY
COMES INTO EXISTENCE, LIFE CAN ONLY ACCESS PARTS OF SEQUENCE
SPACE NEAR EXISTING SPECIES, SO THE SEARCH PATHS THROUGH SEQUENCE
SPACE ARE LIMITED. THIS IS CHANGING AS WE SPEAK. THE QUESTION
OF "WHY HAVEN'T THE MAJORITY EXPANDED?" IS ILL-POSED. THEY HAVE.
WE ARE THEM. THE SUBDOMAIN OF SEQUENCE SPACE IN WHICH
DIFFERENT FORMS ARE SUFFICIENTLY FIT TO SURVIVE IS JUST
SMALLER. RECENT(THIS WEEK) WORK INDICATES THAT ABOUT 1/3 OF HUMAN
RANDOM MUTATIONS ARE FILTERED OUT BY FITNESS IN VERY FEW
> Bill Adlam
> Do You Yahoo!?
> Get your free @yahoo.com address at http://mail.yahoo.com