[Taxacom] Antarctic dispersal (was: Another blow to vicariance in ratite evolution

[Michael Heads]

Hi Ken,

You say: ' I wasn't discussing the division of Crossosomatales into two
main clades (southern and northern). That might have been by vicariance or
maybe not. What I was discussing, and which you did not address in your
comments, was the subsequent evolution and biogeography of the southern
clade'.

You've missed my point. I was suggesting that the distribution of the
southern clade was attained by vicariance with the northern clade. There is
no need for it to disperse from one side of the Indian Ocean to the other
if it already existed on both sides (and on the Seychelles etc.) at the
time of its origin.

'That is a more complicated issue which simple vicariance cannot readily
explain---unless of course, one once again uses the fossil calibration
excuse to push back the time scale tens of millions of years earlier. Funny
how that excuse comes up so often in these discussions'.

That's because it's a fundamental flaw that comes up so often in the
standard argument. The fossil record provides minimum ages. That's not an
excuse, it's simple logic.

'In any case, the maximal ages of eudicot orders are fairly well
constrained because the pollen record is so good (far better than the
macrofossil record)'.

Just because it is better than the macrofossil record doesn't mean it is
'good' in an absolute sense, i.e. that the clade ages based on it are
maximum ages. There are many problems interpreting the fossil pollen record.

'You would be hard pressed to find any botanist who believes
Crossosomatales as a whole is older than 110 million years (and 125 million
years is the accepted age of the entire eudicot clade)'.

Whether or not ideas are accepted or popular is interesting for sociology,
but is not relevant to logical, scientific argument. The centre of
origin/dispersal view is the dominant paradigm, and has been for thousands
of years, but that does not mean it's right. In any case, the eudicot age
is problematic (there is a whole section on it in my 2014 book (p. 67).
Smith et al. (PNAS 2010) wrote that:


Using the first appearance of tricolpate pollen as a fixed calibration may
underestimate the origin

of eudicots and, by extension, other age estimates that have relied on this
constraint. Tricolpate

grains first appear in separated geographical areas and the grains
themselves are not uniform in

morphology . . . , both observations implying that the tricolpate clade
originated some time before

its appearance in the fossil record . . . .


'Dispersal to Madagascar (and from there to the Seychelles and Mascarene
Islands) looks like a fairly safe bet. What's so complicated about that?'

It's not complicated at all, a child could understand it. But if you accept
that, how are you going to explain the very simple global
allopatry (northern and southern clades)?

'And I certainly wasn't arguing in favor of "everything was everywhere".'

You suggested that the elephant bird/kiwi clade was formerly also
on mainland Africa and Australia, although there is no evidence for
that and other ratites occur there. Why bring in ad hoc hypotheses?

'However, Antarctica formerly played an important role in so many taxa as
they dispersed through or from that continent'.

If you have a northern  hemisphere group with a southern
hemisphere/Antarctica sister, both groups can have evolved by simple
vicariance and there is no need for any dispersal *within the groups*,
across Antarctica or anywhere else. There may have been dispersal in the
ancestor.

'As more fossils are discovered in Antarctica, it will no doubt show that
evolution is often more complicated than we once thought, and molecular
evidence already has (ratite phylogeny certainly seems to be). Invoking too
much "simple" vicariance seems overly simplistic to me'.

Why not use a simple process for a very simple pattern? The nine ratite
families (including the fossil lithornithids in the north) are almost
entirely allopatric (rheas overlapping with tinamous in South America and
moas overlapping with kiwis in New Zealand are the only exceptions, these
can be explained by local dispersal *within *those regions). Chance
processes do not really explain this very striking global pattern, simple
vicariance does.

It is very likely that fossil ratites will eventually be found in
Antarctica, but a fossil emu in East Antarctica and a fossil rhea in West
Antarctica, for example, would not upset the global allopatry at all. What
would upset it would be, say, a fossil ostrich in Australia or New Zealand,
or a fossil kiwi in South America. That would be good evidence for more
extensive dispersal following the original allopatry.


On Tue, Jun 10, 2014 at 4:40 AM, Ken Kinman <kinman at hotmail.com> wrote:

>  Hi Michael,
>        I wasn't discussing the division of Crossosomatales into two main
> clades (southern and northern).  That might have been by vicariance or
> maybe not.  What I was discussing, and which you did not address in your
> comments, was the subsequent evolution and biogeography of the southern
> clade.  That is a more complicated issue which simple vicariance cannot
> readily explain---unless of course, one once again uses the fossil
> calibration excuse to push back the time scale tens of millions of years
> earlier.  Funny how that excuse comes up so often in these discussions.
>
>        In any case, the maximal ages of eudicot orders are fairly well
> constrained because the pollen record is so good (far better than the
> macrofossil record).  You would be hard pressed to find any botanist who
> believes Crossosomatales as a whole is older than 110 million years (and
> 125 million years is the accepted age of the entire eudicot clade).
>  Dispersal to Madagascar (and from there to the Seychelles and Mascarene
> Islands) looks like a fairly safe bet. What's so complicated about that?
>
>      And I certainly wasn't arguing in favor of "everything was
> everywhere".  However, Antarctica formerly played an important role in so
> many taxa as they dispersed through or from that continent  As more fossils
> are discovered in Antarctica, it will no doubt show that evolution is often
> more complicated than we once thought, and molecular evidence already has
> (ratite phylogeny certainly seems to be).  Invoking too much "simple"
> vicariance seems overly simplistic to me.
>
>
> ------------------------------
> Date: Mon, 9 Jun 2014 16:45:32 +1200
> Subject: Re: Antarctic dispersal (was: Another blow to vicariance in
> ratite evolution
> From: m.j.heads at gmail.com
> To: kinman at hotmail.com; Taxacom at mailman.nhm.ku.edu
>
>
> Hi Ken,
>
> The Indian Ocean clade of Crossosomatales (Madagascar - New Zealand and
> vicinity) has its sister group neatly allopatric to its north (New Guinea,
> Sri Lanka etc. northwards), so the simplest explanation is that they both
> originated by vicariance. The break between the two (located between
> Madagascar and Sri Lanka, between New Caledonia and New Guinea, etc.)
> coincides with belts of major rifting as well as breaks in countless other
> groups.
>
> The 80 m.y. date for the clades is fossil calibrated and so it is a
> mimimum age. This is useful as it rules out earlier events (not older ones)
> as possible causes.
>
> For the ratite clade, you seem to be supporting the idea that everything
> was everywhere and then gradually the distributions formed by extinction.
> So how and why were they everywhere to start with? It just gets more and
> more complicated for a very simple case of allopatry.
>
>
> On Mon, Jun 9, 2014 at 2:55 PM, Ken Kinman <kinman at hotmail.com> wrote:
>
> Hi Mike,
>         The Crossosomatales southern clade is probably only about 80
> million years old, so dispersal to Madagascar seems to be a good bet (and I
> believe it also dispersed to the Seychelles and Mascarene islands as well).
>  This clade probably existed in Antarctica and Australia (fossil pollen
> records), so the current distribution is no doubt a combination of early
> vicariance, later dispersals, and then extinction in intervening areas
> (including Antarctica and Australia).  I don't know much about the beetle
> family Chaetosomatidae, but perhaps Stephen Thorpe might have an opinion on
> that.
>
>          In any case, since only that beetle family has the same
> distribution as the elephant bird-kiwi clade, I would certainly not call
> that much of a pattern!!!  And the elephant bird-kiwi clade is probably
> even younger than 80 million years old, so dispersal of a flighted ancestor
> (of kiwi and elephant birds) from Antarctica in the early Cenozoic seems a
> very probable hypothesis, and again followed by extinction of the clade in
> Australia (due to competition with other ratites) as well as extinction in
> mainland Africa (due to competition with early ostriches).
>
>         So if there is a pattern to be seen in such cases, I would suggest
> that Antarctica is going to be a major factor, and that the massive
> extinction of the biota in Antarctica during the early-to-mid Cenozoic (and
> paucity of a known fossil record there) makes vicariance more appealing
> than it actually is for many taxa.  In particular, the multiple losses of
> flight among ratite taxa makes perfect sense to me.  Why John calls such
> suggestions ignorant (much less stupid) baffles me.
>                          ----------------------------Ken Kinman
>
>
> ------------------------------
> Date: Mon, 9 Jun 2014 11:12:39 +1200
> Subject: Re: [Taxacom] Another blow to vicariance in ratite evolution
> From: m.j.heads at gmail.com
> To: kinman at hotmail.com; taxacom at mailman.nhm.ku.edu
>
> Hi Ken,
>
> On p. 74 of my 2014 book I cited the beetle family Chaetosomatidae, known
> only from Madagascar (Malgassochaetus) and New Zealand (Chaetosoma,
> Chaetosomodes). A similar pattern in plants occurs in the southern
> hemisphere clade of Crossosomatales; it's in Madagascar (also the adjacent
> coast of Africa and Cape Town) (Aphloia, Geissoloma) , New Zealand (Ixerba)
> and New Caledonia (Strasburgeria). (Both examples are cited, with
> references, in my 2014 book, p. 74).
>
>
>
>


-- 
Dunedin, New Zealand.

My recent books:

*Molecular panbiogeography of the tropics.* 2012. University of California
Press, Berkeley. www.ucpress.edu/book.php?isbn=9780520271968

*Biogeography of Australasia:  A molecular analysis*. 2014. Cambridge
University Press, Cambridge. www.cambridge.org/9781107041028