[Taxacom] Madagascar (was: Timetree of Life)

Thu May 19 09:03:45 CDT 2011

Suppose, John and Michael, that reality is nothing like the statistical 
straw man we create?
Suppose that our stuffing of statistics is our way to trying to make sense 
of observed realities?
Suppose if you were able to talk to the organisms in your observed 
realities, would it not be
a severe shock to find out that our applied statistics really do not reflect 
reality?

In a casino we control all but one variable - the speed of turn of the 
roulette wheel.  We control
the numbers and the colors.  In theory one push of the of the roulette wheel 
gives us a number and
a particular color.  Had I pushed harder or more softly, a different number 
and a different color appear.

If I were working with cards, I have 4 sets of 13 numbers, half red, half 
black.

In biological reality, there are thousands, even often millions, of 
individuals.  And here we are trying to
cram them into our deck of cards or the roulette wheel.  We think we know 
what is going on because
in some cases we can predict.

Man are we ever conceited.

Robin

----- Original Message ----- 
From: "John Grehan" <jgrehan at sciencebuff.org>
To: <taxacom at mailman.nhm.ku.edu>
Sent: Thursday, May 19, 2011 6:23 AM
Subject: Re: [Taxacom] Madagascar (was: Timetree of Life)

Statistics can mean a lot or nothing at all - depending on what it is being 
applied to. Statistics applied to imaginary centers of origin, for example, 
just produces an imaginary statistics of centers of origin. It would be like 
trying to describe the statistical probabilities of Alice's experiences in 
Wonderland.

Henderson (1990) published a Monte Carlo simulation applied to tracks and 
nodes to showed that the number of degree vertices fell well outside the 95% 
probability limits of random distribution. Did that make any difference to 
dispersalists? No of course not.

Of course one may or may not dispute Henderson's statistics. It doesn't 
matter any more than it matters in molecular trees where various theorists 
have dismissed each other's approach as statistically invalid (or invalid 
for other reasons) while at the same time saying their approach is valid, 
and overall the molecular method is valid anyway.

John Grehan

-----Original Message-----
From: taxacom-bounces at mailman.nhm.ku.edu 
[mailto:taxacom-bounces at mailman.nhm.ku.edu] On Behalf Of Richard Zander
Sent: Wednesday, May 18, 2011 5:33 PM
To: Robin Leech; Michael Heads; taxacom at mailman.nhm.ku.edu
Subject: Re: [Taxacom] Madagascar (was: Timetree of Life)

My answer to perplexing questions about the intransigence of life is 
statistical (and therefore rocket science). For every normal distribution, 
there are (usually) two tails of stuff that just does not do what you 
expect. For every 95% credible interval, 5% is on vacation or at least 
staring into the middle distance or reading a good novel.

Apropos of this, it has been pointed out in a book I read that, in casinos 
with roulette wheels (I would not know about this in person, of course), 
they often post a list of all the numbers and colors that come up, in order. 
Apparently due to the nature of chance, the same number or color often comes 
up many, many times in a row.

This is a multiple test problem in that if you have 100 numbers, each with a 
chance it is on vacation some of the time, one of them, if you "test" each 
and every one, has a great chance of doing the very much unexpected.

So, horses and camels in the Americas went "on vacation" and never came 
back, statistically, as a sample of two of a bunch of examined species.

This is fun stuff to mess around with, but the idea that statistics must 
rein in bad or untoward behavior of models in evolutionary analysis is 
something to doubt. Five percent of 95% supported clades, and five percent 
of all statistical phylogenetic papers, could be horrendously wrong. If 
you're lucky. Hummmmmm, if 5% of morphological clades or cladograms conflict 
with molecular clades or cladograms, maybe this is to be expected, 
statistically? What do you think?

* * * * * * * * * * * *
Richard H. Zander
Missouri Botanical Garden, PO Box 299, St. Louis, MO 63166-0299 USAï¿½
Web sites: http://www.mobot.org/plantscience/resbot/ï¿½and 
http://www.mobot.org/plantscience/bfna/bfnamenu.htm
Modern Evolutionary Systematics Web site: 
http://www.mobot.org/plantscience/resbot/21EvSy.htm

-----Original Message-----
From: taxacom-bounces at mailman.nhm.ku.edu 
[mailto:taxacom-bounces at mailman.nhm.ku.edu] On Behalf Of Robin Leech
Sent: Wednesday, May 18, 2011 4:08 PM
To: Michael Heads; taxacom at mailman.nhm.ku.edu
Subject: Re: [Taxacom] Madagascar (was: Timetree of Life)

Michael,
That is rather like asking, "Why didn't something evolve this
way or that way?"
It could be:
1.  no opportunity.
2.  events (weather, floods, timing, etc.).
3.  low population when the time was propitious.
4.  any other reason or reasons you can think up.

Just as with people, "Why oh why didn't I take him up
on his offer for PhD studies?"  "Gee, I wish I had studied
more for this exam."

Horses used to be here in North America.  How come it took
the Spaniards to bring them back?  Why did North America
lose both the equids and camelids?  Why didn't they come back?

Robin

here

_______________________________________________

Taxacom Mailing List
Taxacom at mailman.nhm.ku.edu
http://mailman.nhm.ku.edu/mailman/listinfo/taxacom

The Taxacom archive going back to 1992 may be searched with either of these 
methods:

(1) by visiting http://taxacom.markmail.org

(2) a Google search specified as:  site:mailman.nhm.ku.edu/pipermail/taxacom 
your search terms here

_______________________________________________

Taxacom Mailing List
Taxacom at mailman.nhm.ku.edu
http://mailman.nhm.ku.edu/mailman/listinfo/taxacom

The Taxacom archive going back to 1992 may be searched with either of these 
methods:

(1) by visiting http://taxacom.markmail.org

(2) a Google search specified as:  site:mailman.nhm.ku.edu/pipermail/taxacom 
your search terms here