[Taxacom] Panbiogeography

[John Grehan]

Jason,

Thank you for the comments:

"Your priors are limited to tectonics. Ergo, everything to you looks
like a nail so your interpretations are limited to the scenarios you
are willing to contemplate. It's like living in Flatland, you only see
some dimensions.

No they are not limited to tectonics. As you will see in Heads' book,
various means of providing a temporal correlation are taken into account.
Fossils are just as valid as tectonics. The problem is that the current use
of fossil calibration misrepresents divergence estimates as actual or
maximal when they are not. Tectonic correlations are also minimal estimates
with respect to time, but when there  is a biogeographic break correlated
with a tectonic break there is reason to consider that they are related.


"Panbiogeography looks to correlation with geomorphology rather than
historical hypotheses."

Biogeography in general looks at the correlation between distribution
patterns, phylogeny and then looks for mechanisms to account for this,
geology and dispersal. Thus it is not the sole domain of
pnabiogeography.

I was referring to the difference between geological patterns and
geological reconstructions. Matching a phylogenetic sequence to a
geological historical reconstruction is not the same as matching a
phylogenetic sequence to a tectonic pattern. See Heads' book for many
examples.

"The difference is that Panbiogeography does not
contemplate dispersal because, me thinks, that would make the results
fuzzier and therefore would make it impossible to justify the
tautology that in the bottom it has become: biogeography predicting
geology and viceversa"

The panbiogeographic method is about pattern analysis. In principle,
distribution data provide an empirical resource. Speculations about
imagined chance dispersal does not.


Ken has already told you that your desire to stretch maximal dates to
fit (no, shoehorn) your only contemplated mechanism (tectonics) would
often lead to ridiculous scenarios.

He may have told me that, but just saying so does not mean that such
assertions are correct.





Phylogeny calibration is as good
as the calibrating points so it is work in progress, but rather useful
when comparing relative timings within the phylogeny. Thus, it is very
likely that many lineages will, in time, be shown to be older than
initially estimated. However it is mystifying to me how it can be used
it to justify their position.

Sorry, I cannot make sense of the above.

We can only talk of the data that we
have, not the data that can possibly exist.

But one can predict data that possibly exists, e.g. tectonic formations
that have/had not been discovered.



Thus Panbiogeography asks for the impossible, that other biogeographers
prove that a fossil does
not exist. This is what I mean, in part, about Panbiogeography being
unfalsifiable.

Panbiogeography does not ask for that.

"Panbiogeographic analysis demonstrates that vicariism is
pervasive..." No, panbiogeographic analysis is a hammer and only sees
nails.But when you find a screw you insist it must be a nail.
Vicariance is common, that's all.

Sorry, the empirical data shows otherwise. Read Heads' book which is
repleat with examples of vicariant patterns.

"... and patterns of vicariism are shared by "good" and "poor" dispersers."
How do you define good an bad disperser if, according to you, they do
not exist?

Exactly my point. Organisms with purported "good" means (birds, marine
planktonic stages, aireal seeds etc) share biogeography with the purported
"bad" (flightless birds, insects, freshwater fish, frogs etc).

In any case, what is the significance of your statement? We
are talking about succesful dispersal, which is random and likely a
combination of factors.

Panbiogeographic analysis is one approach to providing information on how
dispersal works in the evolution of vicariant differentiation.

"This makes the theory of chance (extraordinary) dispersal
problematic." Are you seeking some sort of meaning in nature?

No, the patterns make the theory problematic as the patterns are
incongruous with the expectations of chance dispersal.



As to equating chance and extraordinary as equal, playing
dice is a game of pure chance but not extraordinary. It is the
combination of probabilities that would make it highly unlikely
(extraordinary). Successful dispersal becomes more unlikely with
increasing distance, the degree of tapering of which is likely
taxon-specific. So unlikely yes, but extraordinary is a choice of
words as poor as "miraculous" or "mystifying".

However you chose your words, the point is that panbiogeographic analysis
shows that distance is not a predictor of distribution range for
differentiated taxa so the 'probabilities' are contingent upon the relevant
factors of geological, climatic, or human disruption.

"If an organism is said to cross a "barrier" how does one know that it
is a barrier."
Normally, something like this: if it dies there with a certainty of a
100%, it is a barrier, if survival is equivalent to its normal
habitat, it is not and in between you have a fuzzy area. I guess that
is why Stephen thinks dispersal is more relevant in the pacific
islands (more tightly defined barrier).

But if it dies after crossing the barrier then it is not a barrier. It dies
from something else. One may think that dispesal is more relevant to the
Pacific islands, but panbiogeographic analysis offers another possibility
(again, read Heads' book).

"They have invoked dispersal as one of the mechanisms to explain biotic
distribution."

When it comes down to explaining vicariant patterns the principal choice
has been to frame the origin in terms of dispesal from centers of origin.

"...but perhaps one could use biogeographic patterns to understand how
dispersal works in the evolution of vicariant distributions."
So we assume dispersal does not exist so we can use vicariance to
understand how dispersal works... even though it doesn't exist? You
need to explain this a bit.

I have not asserted that dispersal does not exist. See Heads' book to
better understand the panbiogeographic perspective on how dispersal works.

John Grehan


On Wed, Mar 26, 2014 at 8:09 AM, JF Mate <aphodiinaemate at gmail.com> wrote:

> John, sorry for the delay in replying.
>
> "Historical reconstructions are hypotheses. I am not aware of
> geological hypothesis made from geological research necessarily
> falsify of an incongruent prediction (hypotheses) made from
> biogeography. But it is not a simple case of geology/genes/phylogeny
> being wrong. In fact, as shown in Heads' book, it is not a simple
> matter of geology/genes/phylogeny being 'wrong' but in how they may be
> interpreted."
>
> Your priors are limited to tectonics. Ergo, everything to you looks
> like a nail so your interpretations are limited to the scenarios you
> are willing to contemplate. It's like living in Flatland, you only see
> some dimensions.
>
>
> "Panbiogeography looks to correlation with geomorphology rather than
> historical hypotheses."
>
> Biogeography in general looks at the correlation between distribution
> patterns, phylogeny and then looks for mechanisms to account for this,
> geology and dispersal. Thus it is not the sole domain of
> pnabiogeography. The difference is that Panbiogeography does not
> contemplate dispersal because, me thinks, that would make the results
> fuzzier and therefore would make it impossible to justify the
> tautology that in the bottom it has become: biogeography predicting
> geology and viceversa.
>
> "Panbiogeography looks at the spatiality of genes (whether in
> molecular or morphogentic form) rather than their interpretation as
> absolute molecular clocks based on misrepresenting minimal divergence
> dates as absolute or maximal, and looks to the spatiality of phylogeny
> to examine patterns of common connections and breaks rather than as
> imagined centers of origin. Molecular divergence estimates have been
> widely hailed as the falsification of panbiogeography and yet it is
> the divergence estimate that rested on the false representation as
> absolute or maximal dating rather than minimal."
>
> Ken has already told you that your desire to stretch maximal dates to
> fit (no, shoehorn) your only contemplated mechanism (tectonics) would
> often lead to ridiculous scenarios. Phylogeny calibration is as good
> as the calibrating points so it is work in progress, but rather useful
> when comparing relative timings within the phylogeny. Thus, it is very
> likely that many lineages will, in time, be shown to be older than
> initially estimated. However it is mystifying to me how it can be used
> it to justify their position. We can only talk of the data that we
> have, not the data that can possibly exist. Thus Panbiogeography asks
> for the impossible, that other biogeographers prove that a fossil does
> not exist. This is what I mean, in part, about Panbiogeography being
> unfalsifiable.
>
>
> "Panbiogeographic analysis demonstrates that vicariism is
> pervasive..." No, panbiogeographic analysis is a hammer and only sees
> nails.But when you find a screw you insist it must be a nail.
> Vicariance is common, that's all.
>
> "... and patterns of vicariism are shared by "good" and "poor" dispersers."
> How do you define good an bad disperser if, according to you, they do
> not exist? In any case, what is the significance of your statement? We
> are talking about succesful dispersal, which is random and likely a
> combination of factors.
>
> "This makes the theory of chance (extraordinary) dispersal
> problematic." Are you seeking some sort of meaning in nature? As to
> equating chance and extraordinary as equal, playing
> dice is a game of pure chance but not extraordinary. It is the
> combination of probabilities that would make it highly unlikely
> (extraordinary). Successful dispersal becomes more unlikely with
> increasing distance, the degree of tapering of which is likely
> taxon-specific. So unlikely yes, but extraordinary is a choice of
> words as poor as "miraculous" or "mystifying".
>
> "If an organism is said to cross a "barrier" how does one know that it
> is a barrier."
> Normally, something like this: if it dies there with a certainty of a
> 100%, it is a barrier, if survival is equivalent to its normal
> habitat, it is not and in between you have a fuzzy area. I guess that
> is why Stephen thinks dispersal is more relevant in the pacific
> islands (more tightly defined barrier).
>
> "..In panbiogeography the question is less about dispersal and
> vicariance being competing mechanisms as they are complimentary.
> Biogeographers have traditionally invoked a concept of dispersal
> (chance, long distance etc) by which to interpret biogeographic
> patterns,..."
> They have invoked dispersal as one of the mechanisms to explain biotic
> distribution.
>
> "...but perhaps one could use biogeographic patterns to understand how
> dispersal works in the evolution of vicariant distributions."
> So we assume dispersal does not exist so we can use vicariance to
> understand how dispersal works... even though it doesn´t exist? You
> need to explain this a bit.
>
>
> Michael
>
> "The difference between normal, observed dispersal discussed by
> ecologists (e.g. weeds dispersing into a garden), and chance, 'jump'
> or 'long distance' dispersal as invoked by evolutionists, is that the
> former does not involve differentiation, whereas the latter is
> proposed as a mode of speciation."
>
> Not true and no different (except in scale that is). We may argue
> about the actual probability the range of dispersal, but dispersal is
> a probability curve with a long tapering tail with increasing
> distance. Also dispersal is but one mechanism leading to allopatry
> (Mayr or Dobzhansky mention "barriers" emerging as well).
>
> "Dispersal theory explains range overlap by dispersal, but also
> explains allopatry by dispersal. Vicariance theory explains range
> overlap by dispersal, but explains allopatry by vicariance."
> Dispersal is one mechanism to explain allopatry. Vicariance in
> Panbiogeography is the only mechanism contemplated for reasons that
> have been explained many times before by people far abler than me.
>
>
> "Note that the dispersal invoked in vicariance theory is caused by
> geological change, whereas dispersal as invoked by dispersal theory to
> explain allopatry, is caused by chance."
> So you think chance is somehow wrong? You need to elaborate.
>
> _______________________________________________
> Taxacom Mailing List
> Taxacom at mailman.nhm.ku.edu
> http://mailman.nhm.ku.edu/mailman/listinfo/taxacom
> The Taxacom Archive back to 1992 may be searched at:
> http://taxacom.markmail.org
>
> Celebrating 27 years of Taxacom in 2014.
>