Musings on extinct Phyla

[Paul Ciszek]

There are about 13 phyla with no fossil record. There are phyla with
very few surviving members--Priapulida are down to only 18 known species.
Onchyphora lost all their non-terrestrial members at some point. I am
thinking that there are probably many exitinct taxa that would deserve
the rank of "phylum" if we knew about them. And that humans may just
manage to extinguish a phylum or two.

--
Please reply to: | "Evolution is a theory that accounts
pciszek at panix dot com | for variety, not superiority."
Autoreply has been disabled | -- Joan Pontius

[John S. Wilkins]

Paul Ciszek <nos...@nospam.com> wrote:

> There are about 13 phyla with no fossil record. There are phyla with
> very few surviving members--Priapulida are down to only 18 known species.
> Onchyphora lost all their non-terrestrial members at some point. I am
> thinking that there are probably many exitinct taxa that would deserve
> the rank of "phylum" if we knew about them. And that humans may just
> manage to extinguish a phylum or two.

If you think "phylum" is an objective rank, go look at what are phyla
amongst bacteria and single celled organisms...
--
John S. Wilkins, Associate, Philosophy, University of Sydney
http://evolvingthoughts.net
But al be that he was a philosophre,
Yet hadde he but litel gold in cofre

[Darwin123]

On Nov 4, 12:45 am, j...@wilkins.id.au (John S. Wilkins) wrote:
> Paul Ciszek <nos...@nospam.com> wrote:
> > There are about 13 phyla with no fossil record.  There are phyla with
> > very few surviving members--Priapulida are down to only 18 known species.
> > Onchyphora lost all their non-terrestrial members at some point.  I am
> > thinking that there are probably many exitinct taxa that would deserve
> > the rank of "phylum" if we knew about them.  And that humans may just
> > manage to extinguish a phylum or two.
>
> If you think "phylum" is an objective rank, go look at what are phyla
> amongst bacteria and single celled organisms...

I agree that the rankings among bacteria and viruses are far more
subjective than the rankings in eukaryotes.
Among bacteria and viruses, there is a large amount of lateral
gene transfer. The "hybridization barriers" in lateral gene transfer
is much weaker for lateral gene transfer in meiotic gene exchange
(i.e., true sex).
I don't know anything about the rankings in single celled
eukaryotes. Are phyla well defined among protozoa and single celled
fungi (e.g., yeast)?
I am conjecturing the situation is a bit clearer among eukaryotes,
even the single celled ones. However, I am not sure.
It does appear to me to be true that phyla in animals and
divisions in plants are close to objective among extant animals. In
fact, I think that the major phyla of animals seem well defined all
the way back to the Cambrian explosion.
There is little doubt there were "transitional" phyla previous to
the Cambrian. There are a few fossil animals (?) in the Edicaran
period, just before the Cambrian. I think that many of those animals
have either ambiguous or extinct phyla.

[Darwin123]

On Nov 3, 7:12 pm, nos...@nospam.com (Paul Ciszek) wrote:
> There are about 13 phyla with no fossil record.  There are phyla with
> very few surviving members--Priapulida are down to only 18 known species.
> Onchyphora lost all their non-terrestrial members at some point.  I am
> thinking that there are probably many exitinct taxa that would deserve
> the rank of "phylum" if we knew about them.

Do you want to count the "transitional phyla?" We may in the
future find fossils from peCambrian animals that have features from
two or more extant phyla. Although their phyla may be ambiguous, I
wouldn't consider them as coming from "extinct phyla." They would be
animals on the boundary between two phyla.
There is one phylum of animals (?) which I consider extinct. Dead
as door nails. The phylum of animals that some call rangomorphs seem
to have gone extinct, left no survivors and don't seem transitional to
any other phyla. The rangomorphs are dead as doornails.
There are fossils from the Edicaran period, which is the period
of time just before the Cambrian period. Many of those creatures have
a fern like profile (although their three dimensional structure was
far more complex). The Edicaran period extended over 40 MY before the
Cambrian explosion (circa 550 MYA).
Some paleontologists classify these organisms as rangomorphs. The
rangomorphs are a phylum (or maybe kingdom) that has gone extinct. No
one is certain precisely how rangomorphs fit in a cladogram. It does
not seem to have given rise to any extant phylum.
I tentatively present the rangomorphs as a truly extinct phylum.
So far as I can tell, there are not other known phyla that have gone
extinct.
There are fossil eggs with embryos from preCambrian times. There
are eggs from about 750 MYA. E can tell that they were triblastic
animals. One can distinguish between deutrerosomes and protosomes in
these eggs. However, there is no way to tell what phylum these guys
were in. That doesn't mean that they were or weren't from extant
phyla. However, scientists don't yet know.

[John Harshman]

Darwin123 wrote:
> On Nov 4, 12:45 am, j...@wilkins.id.au (John S. Wilkins) wrote:
>> Paul Ciszek <nos...@nospam.com> wrote:
>>> There are about 13 phyla with no fossil record. There are phyla with
>>> very few surviving members--Priapulida are down to only 18 known species.
>>> Onchyphora lost all their non-terrestrial members at some point. I am
>>> thinking that there are probably many exitinct taxa that would deserve
>>> the rank of "phylum" if we knew about them. And that humans may just
>>> manage to extinguish a phylum or two.
>> If you think "phylum" is an objective rank, go look at what are phyla
>> amongst bacteria and single celled organisms...
> I agree that the rankings among bacteria and viruses are far more
> subjective than the rankings in eukaryotes.

You're agreeing with something Wilkins didn't say.

> Among bacteria and viruses, there is a large amount of lateral
> gene transfer. The "hybridization barriers" in lateral gene transfer
> is much weaker for lateral gene transfer in meiotic gene exchange
> (i.e., true sex).

Here you are confusing the objectivity of ranking (which Wilkins denies)
with the objectivity of the existence of clades (which he may or may not
deny in some taxa). But the point is that he's talking about ranks.

> I don't know anything about the rankings in single celled
> eukaryotes. Are phyla well defined among protozoa and single celled
> fungi (e.g., yeast)?

Groups are fairly well defined. Phyla, however, are not. Few
systematists are using the term any more.

> I am conjecturing the situation is a bit clearer among eukaryotes,
> even the single celled ones. However, I am not sure.
> It does appear to me to be true that phyla in animals and
> divisions in plants are close to objective among extant animals. In
> fact, I think that the major phyla of animals seem well defined all
> the way back to the Cambrian explosion.

That's because you don't understand what Wilkins means. How does a
phylum differ from a class or from a kingdom? What objective measure can
you use to decide that group X should be ranked as a phylum rather than
something else or, as is the common recent practice, not ranked at all?

> There is little doubt there were "transitional" phyla previous to
> the Cambrian. There are a few fossil animals (?) in the Edicaran
> period, just before the Cambrian. I think that many of those animals
> have either ambiguous or extinct phyla.

Many of the Cambrian animals belong to extinct phyla, depending on just
which nodes you give the name to.

[John S. Wilkins]

Darwin123 <drose...@yahoo.com> wrote:

> On Nov 4, 12:45 am, j...@wilkins.id.au (John S. Wilkins) wrote:
> > Paul Ciszek <nos...@nospam.com> wrote:
> > > There are about 13 phyla with no fossil record. There are phyla with
> > > very few surviving members--Priapulida are down to only 18 known species.
> > > Onchyphora lost all their non-terrestrial members at some point. I am
> > > thinking that there are probably many exitinct taxa that would deserve
> > > the rank of "phylum" if we knew about them. And that humans may just
> > > manage to extinguish a phylum or two.
> >
> > If you think "phylum" is an objective rank, go look at what are phyla
> > amongst bacteria and single celled organisms...
> I agree that the rankings among bacteria and viruses are far more
> subjective than the rankings in eukaryotes.

Uh, no my point is that phylum is always subjective. What is salient to
us'n vertebrates is why we think phyla are qualitatively different, but
close examination doesn't support that for animals *or* bacteria. There
are just bodies. The plans exist solely in the heads of biologists.

> Among bacteria and viruses, there is a large amount of lateral
> gene transfer. The "hybridization barriers" in lateral gene transfer
> is much weaker for lateral gene transfer in meiotic gene exchange
> (i.e., true sex).

I am unsure there is any evidence to support this. For a start, the rate
of hybridisation between animal species is pretty high, much higher than
we expected. Second, it is not clear that the per-generation rate of LGT
in bacteria is all that high either.

> I don't know anything about the rankings in single celled
> eukaryotes. Are phyla well defined among protozoa and single celled
> fungi (e.g., yeast)?

They aren't well defined in any group. They are phenomenally salient due
to human perceptual dispositions and some similarity set metrics that
are historically contingent.

> I am conjecturing the situation is a bit clearer among eukaryotes,
> even the single celled ones. However, I am not sure.
> It does appear to me to be true that phyla in animals and
> divisions in plants are close to objective among extant animals. In
> fact, I think that the major phyla of animals seem well defined all
> the way back to the Cambrian explosion.
> There is little doubt there were "transitional" phyla previous to
> the Cambrian. There are a few fossil animals (?) in the Edicaran
> period, just before the Cambrian. I think that many of those animals
> have either ambiguous or extinct phyla.

The distinction is between *the* phyla, which are named groups, and the
*rank* of phylum, which is an abstraction. Individual phyla are probably
real groups (not all), but the rank is subjectively important.

[Darwin123]

On Nov 6, 4:30 pm, j...@wilkins.id.au (John S. Wilkins) wrote:

> The distinction is between *the* phyla, which are named groups, and the
> *rank* of phylum, which is an abstraction. Individual phyla are probably
> real groups (not all), but the rank is subjectively important.

Okay, I agree with that.
I also agree that the ranking of the different taxons can be
ambiguous, even when the particular taxon is not.
Plants, fungi and animals are considered kingdoms. Sponges,
cnidaria and chordates are phyla in the animal kingdom. Jawless fish,
bony fish and cartilaginous fish are classes.
I suspect that these different taxons are fairly well defined. One
can distinguish between plants, fungi and animals somewhat. The other
taxons are also well defined.
However, their ranking is not well defined. Is fungi a real
kingdom, or is it a superphylum of animals? Maybe it is a division of
plants?
Are sponges really a phylum of animals? Maybe we should classify
them as a kingdom? Somewhere between animals and fungi?
Maybe fish should be one class. Jawless fish, bony fish and
cartilaginous fish should be subclasses within fish. Or maybe they
should place jawless fish, bony fish and cartiliginous fish into
different phyla of animals.
The ranking is more ambiguous than the actual classification. I
didn't understand that point when I first read your post.
Thank you for clearing that up.

[Arkalen]

The funny thing is to look at the classification of Animalia on
Wikipedia. On one page birds will be a class along with mammals and
reptiles, and on the next they'll be nested within the order Dinosauria
which is part of the reptile class. And that's not even getting into the
fishes. Talk about confusing.

[Paul Ciszek]

In article <a99116e5-f8a8-48f3...@ek5g2000vbb.googlegroups.com>,

Darwin123 <drose...@yahoo.com> wrote:
> There are fossil eggs with embryos from preCambrian times. There
>are eggs from about 750 MYA. E can tell that they were triblastic
>animals. One can distinguish between deutrerosomes and protosomes in
>these eggs. However, there is no way to tell what phylum these guys
>were in. That doesn't mean that they were or weren't from extant
>phyla. However, scientists don't yet know.

Whoa, that is the first I have heard of 750 million year old embryos.
Cite? I thought fossil sponges go back only 650 million years.

--
"Remember when teachers, public employees, Planned Parenthood, NPR and PBS
crashed the stock market, wiped out half of our 401Ks, took trillions in
TARP money, spilled oil in the Gulf of Mexico, gave themselves billions in
bonuses, and paid no taxes? Yeah, me neither."

[John Harshman]

Paul Ciszek wrote:
> In article <a99116e5-f8a8-48f3...@ek5g2000vbb.googlegroups.com>,
> Darwin123 <drose...@yahoo.com> wrote:
>> There are fossil eggs with embryos from preCambrian times. There
>> are eggs from about 750 MYA. E can tell that they were triblastic
>> animals. One can distinguish between deutrerosomes and protosomes in
>> these eggs. However, there is no way to tell what phylum these guys
>> were in. That doesn't mean that they were or weren't from extant
>> phyla. However, scientists don't yet know.
>
> Whoa, that is the first I have heard of 750 million year old embryos.
> Cite? I thought fossil sponges go back only 650 million years.
>

He's referring to the Doushantuo. But I don't know where he gets his age
estimate.

[pnyikos]

> fishes. Talk about confusing.-

Yes, what you are seeing is two different classification schemes in
action. The one you name first uses the traditional Linnean
classification, the other is a cladistic classification [John Harshman
swears by them, hence I call him a cladophile.] which hasn't yet
divested itself of ranks like "order".

Cladistic classification works great for extant species, especially if
you are sure that the species did not give rise to another species, as
in the case of the polar bear being descended from the brown bear, the
two being species in the genus *Ursus*.

However, the further back in time you go and the more extinct groups
you use, the cladistic classification is less and less satisfactory
for anything except reconstructing phylogenies. At some point you
have to sacrifice the idea of every extinct creature having its own
subfamily, family, superfamily, suborder, order, superorder, etc. For
instance, the smallest clade containing *Archaeopteryx* and anything
else is the whole class Aves, when last I checked.

So, once the cladophiles drive all other systematists to extinction,
anyone wanting to know what birds resemble "Archie" the most is going
to have to try to worm the information out of the cladophiles most
knowledgeable about Aves. I don't know how well informed of John
Harshman is about Aves, for example.

Peter Nyikos

[pnyikos]

On Nov 6, 2:28 pm, John Harshman <jharsh...@pacbell.net> wrote:
> Darwin123 wrote:
> > On Nov 4, 12:45 am, j...@wilkins.id.au (John S. Wilkins) wrote:
> >> Paul Ciszek <nos...@nospam.com> wrote:
> >>> There are about 13 phyla with no fossil record.  There are phyla with
> >>> very few surviving members--Priapulida are down to only 18 known species.
> >>> Onchyphora lost all their non-terrestrial members at some point.  I am
> >>> thinking that there are probably many exitinct taxa that would deserve
> >>> the rank of "phylum" if we knew about them.

Receptaculites may be an example. They were believed to be some sort
of invertebrate at some point, but according to the miserably skimpy
Wikipedia reference, they are apparently believed to be algae
nowadays.

Conulariids, too -- the Wikipedia entry says the prevailing theory is
that they are cnidarians ["coelenterates"] but remarks at the end that
there is some controversy about this.

Or from a family? See my reply to Arkalen, for example, or my
comments at the end of this post.

>What objective measure can
> you use to decide that group X should be ranked as a phylum rather than
> something else or, as is the common recent practice, not ranked at all?

I suspect "subjective" is a dirty word to you when talking about
natural history, but there are VERY sound subjective rules of thumb
for these decisions that paleontologists were making until the
cladophiles upset the apple cart.

Did you choose your wife for objective reasons only? :-)

> >     There is little doubt there were "transitional" phyla previous to
> > the Cambrian. There are a few fossil animals (?) in the Edicaran
> > period,

Isn't the period called the Vendian, and Ediacara the name of the
known fauna from that time from Australia?

> >just before the Cambrian. I think that many of those animals
> > have either ambiguous or extinct phyla.

There is a theory that the Ediacara fauna left no descendants, and
that all resemblances between them and such things as jellyfish and
sea pens is due purely to convergence.

A few years ago, I read a book that hypothesized that they got all
their nutrition thru symbiosis with microorganisms, the way some tube
worms that live near deep sea vents do. Once evolution gave rise to a
predatory animal, these peaceful creatures were doomed according to
that theory.

> Many of the Cambrian animals belong to extinct phyla, depending on just
> which nodes you give the name to.

This last sentence of yours is priceless. Elaborating on what I told
Arkalen, I wonder whether a traditional systematist might well group
lots of Cambrian "phyla" into one family, simply because they don't
seem to have diverged from a common ancestor all that long ago, and
they still have a lot more in common than we have with hagfishes, or
even with lemurs.

The point is, if you call Chordata, Mollusca, Annelida etc. phyla, and
you are a cladistic systematist, then the logical thing to do is to
refer to all clades that diverged from other clades at the same time
these did as "phyla." The only exception would be if they are parts
of larger groups that you want to call phyla.

Peter Nyikos

[John S. Wilkins]

pnyikos <nyi...@bellsouth.net> wrote:

> I suspect "subjective" is a dirty word to you when talking about
> natural history, but there are VERY sound subjective rules of thumb
> for these decisions that paleontologists were making until the
> cladophiles upset the apple cart.
>
> Did you choose your wife for objective reasons only? :-)

If it were about choosing wives and husbands, then subjective is best,
subjectively speaking.

But classification is about finding real patterns and groupings in the
world, and so should not be held hostage to the whims and vagaries of
individuals, no matter how infuential they are. That is, unless
classification is just library science, in which case any system is as
good as any other, objectively, and it's all about what is most
convenient.

The soundness of pre-phylogenetic classifications depended upon the
individual learning sufficient about the group to identify real
groupings within it. But the ranks and definienda are subjective or
conventional and mean nothing.

That said, some cladistic classifications seem to me to be an attempt to
do science by algorithm, and I do not think they license the claims and
inferences made from them.

[John Harshman]

pnyikos wrote:
> On Nov 6, 2:28 pm, John Harshman <jharsh...@pacbell.net> wrote:

>> What objective measure can
>> you use to decide that group X should be ranked as a phylum rather than
>> something else or, as is the common recent practice, not ranked at all?
>
> I suspect "subjective" is a dirty word to you when talking about
> natural history, but there are VERY sound subjective rules of thumb
> for these decisions that paleontologists were making until the
> cladophiles upset the apple cart.

An interesting claim. Can you state any of those rules?

>> Many of the Cambrian animals belong to extinct phyla, depending on just
>> which nodes you give the name to.
>
> This last sentence of yours is priceless. Elaborating on what I told
> Arkalen, I wonder whether a traditional systematist might well group
> lots of Cambrian "phyla" into one family, simply because they don't
> seem to have diverged from a common ancestor all that long ago, and
> they still have a lot more in common than we have with hagfishes, or
> even with lemurs.

You are free to wonder all you like, but I don't see much point in it.
Do you have any clear idea what my priceless sentence meant?

> The point is, if you call Chordata, Mollusca, Annelida etc. phyla, and
> you are a cladistic systematist, then the logical thing to do is to
> refer to all clades that diverged from other clades at the same time
> these did as "phyla."

I'm curious why you think that's "the logical thing to do". Basing ranks
on clade age has been suggested a few times, but has generally foundered
on the difficulty of deciding ages accurately.

> The only exception would be if they are parts
> of larger groups that you want to call phyla.

In other words, the only exceptions would be if you want to do something
else.

[John Harshman]

Sentimental attachment.

> Cladistic classification works great for extant species, especially if
> you are sure that the species did not give rise to another species, as
> in the case of the polar bear being descended from the brown bear, the
> two being species in the genus *Ursus*.

Yes indeed it does, even if species did give rise to other species.

> However, the further back in time you go and the more extinct groups
> you use, the cladistic classification is less and less satisfactory
> for anything except reconstructing phylogenies.

We obviously disagree on this point.

> At some point you
> have to sacrifice the idea of every extinct creature having its own
> subfamily, family, superfamily, suborder, order, superorder, etc.

No you don't. You just have to multiply redundant ranks. Or you could
forget about ranks, which is of course what most systematists do these days.

> For
> instance, the smallest clade containing *Archaeopteryx* and anything
> else is the whole class Aves, when last I checked.

We don't actually know, since the phylogeny around the base of Aves (I
would prefer calling it Avialae and save Aves for another group) is
currently in contention. I fail to see, however, any way in which this
is an example of the difficulties you allude to earlier.

> So, once the cladophiles drive all other systematists to extinction,
> anyone wanting to know what birds resemble "Archie" the most is going
> to have to try to worm the information out of the cladophiles most
> knowledgeable about Aves.

So here's the nub: You want the classification to tell you how similar
organisms are, but classifications are not well suited to that task. If
you remember, that's why you created your bizarre multiply-overlapping
groups system.

> I don't know how well informed of John
> Harshman is about Aves, for example.

There are so many things you don't know.

[Arkalen]

What else would you use cladistics for ?
Including extinct groups is a problem for any classification scheme; a
non-cladistic trait-based system of classification stumbles on how to
classify transitional forms. As far as I can tell cladistics (and using
phylogenies instead of ranked classes) is the only way to deal with that.

> At some point you
> have to sacrifice the idea of every extinct creature having its own
> subfamily, family, superfamily, suborder, order, superorder, etc. For
> instance, the smallest clade containing *Archaeopteryx* and anything
> else is the whole class Aves, when last I checked.

Archaeopteryx keeps being reclassified; inside Aves, outside Aves but
inside Avialae, with the dromaerosaurs...

Also, what about archaeopterygids ? That's a smaller clade than Aves,
whatever it's nested in.

>
> So, once the cladophiles drive all other systematists to extinction,
> anyone wanting to know what birds resemble "Archie" the most is going
> to have to try to worm the information out of the cladophiles most
> knowledgeable about Aves.

As opposed to trying to worm the information out of other systematists
who are most knowledgeable about birds...

Wait, are you suggesting that there are some modern birds that resemble
Archaeopteryx more than they resemble other modern birds, and that they
should be classified with it so that we can tell ? But wouldn't that be
completely misleading given all modern birds resemble each other more
than any of them resemble Archaeopteryx ?

If you are referring to extinct birds, then just looking at the
phylogeny gives the answer. The only uncertainties that exist reflect
actual disagreements among paleontologists over which fossils resemble
which, which would be an issue whatever the classification system used.

[pnyikos]

On Nov 7, 9:21 pm, John Harshman <jharsh...@pacbell.net> wrote:
> pnyikos wrote:
> > On Nov 6, 2:28 pm, John Harshman <jharsh...@pacbell.net> wrote:
> >> What objective measure can
> >> you use to decide that group X should be ranked as a phylum rather than
> >> something else or, as is the common recent practice, not ranked at all?
>
> > I suspect "subjective" is a dirty word to you when talking about
> > natural history, but there are VERY sound subjective rules of thumb
> > for these decisions that paleontologists were making until the
> > cladophiles upset the apple cart.
>
> An interesting claim. Can you state any of those rules?

Does Giganews no longer recognize sci.bio.systematics? That would be
the logical place to discuss this. Google Groups still recognizes it;
it isn't one of those archive-only newsgroups that really died.

It's moribund, but not dead like you claimed. I checked a minute ago,
and even posted successfully on it.

> >> Many of the Cambrian animals belong to extinct phyla, depending on just
> >> which nodes you give the name to.
>
> > This last sentence of yours is priceless. Elaborating on what I told
> > Arkalen, I wonder whether a traditional systematist might well group
> > lots of Cambrian "phyla" into one family, simply because they don't
> > seem to have diverged from a common ancestor all that long ago, and
> > they still have a lot more in common than we have with hagfishes, or
> > even with lemurs.
>
> You are free to wonder all you like, but I don't see much point in it.

As usual, you react like a theoretical systematist. Are you as
theoretical as John Wilkins, who sees no other use for sysematics
except the reconstruction of phylogenetic trees?

> Do you have any clear idea what my priceless sentence meant?

Sure. It meant that there are a bunch of nodes in the Tree of Life
that are dated to Cambrian times and you might call the clade that
originates at one of them a phylum, or you might just move a tad up or
down and make your division there. For instance, you might include
Onychophora and Arthropoda in one phylum that starts a few nodes
further down from each, or you might do things the way they are done
now.

An even better example is Nematoda/Nemathelminthes, Nematophora and a
couple of other taxa sometimes called phyla being lumped together in
Aschelminthes. That's what they did when I was a teenager, and there
might have been all kinds of compromises between the splitters and
lumpers since then.

Your turn: do you have any idea why I called it priceless?

> > The point is, if you call Chordata, Mollusca, Annelida etc. phyla, and
> > you are a cladistic systematist, then the  logical thing to do is to
> > refer to all clades that diverged from other clades at the same time
> > these did as "phyla."
>
> I'm curious why you think that's "the logical thing to do". Basing ranks
> on clade age has been suggested a few times, but has generally foundered
> on the difficulty of deciding ages accurately.

Is't not logical to call clades "phyla" if they branched off from each
other at nodes that are very close to being the same?

Anyway, I suppose you haven't got much use for the term "phyla" and so
for you the whole point is nearly moot.

> > The only exception would be if  they are parts
> > of larger groups that you want to call phyla.
>
> In other words, the only exceptions would be if you want to do something
> else.

Stop oversimplifying. I doubt that anyone would want to get rid of
Chordata, Mollusca, and Annelida as phyla (even though the last may
well be paraphyletic) but that is what would happen if you tried to
trim the number of extinct phyla excessively.

Peter Nyikos

[John Harshman]

pnyikos wrote:
> On Nov 7, 9:21 pm, John Harshman <jharsh...@pacbell.net> wrote:
>> pnyikos wrote:
>>> On Nov 6, 2:28 pm, John Harshman <jharsh...@pacbell.net> wrote:
>>>> What objective measure can
>>>> you use to decide that group X should be ranked as a phylum rather than
>>>> something else or, as is the common recent practice, not ranked at all?
>>> I suspect "subjective" is a dirty word to you when talking about
>>> natural history, but there are VERY sound subjective rules of thumb
>>> for these decisions that paleontologists were making until the
>>> cladophiles upset the apple cart.
>> An interesting claim. Can you state any of those rules?
>
> Does Giganews no longer recognize sci.bio.systematics?

Apparently. It causes an error every time I try it.

> That would be
> the logical place to discuss this. Google Groups still recognizes it;
> it isn't one of those archive-only newsgroups that really died.
>
> It's moribund, but not dead like you claimed. I checked a minute ago,
> and even posted successfully on it.

By "dead" I mean that it gets no legitimate traffic, and nobody reads
it. Or didn't the last I looked, which was some time ago.

>>>> Many of the Cambrian animals belong to extinct phyla, depending on just
>>>> which nodes you give the name to.
>>> This last sentence of yours is priceless. Elaborating on what I told
>>> Arkalen, I wonder whether a traditional systematist might well group
>>> lots of Cambrian "phyla" into one family, simply because they don't
>>> seem to have diverged from a common ancestor all that long ago, and
>>> they still have a lot more in common than we have with hagfishes, or
>>> even with lemurs.
>> You are free to wonder all you like, but I don't see much point in it.
>
> As usual, you react like a theoretical systematist. Are you as
> theoretical as John Wilkins, who sees no other use for sysematics
> except the reconstruction of phylogenetic trees?

I don't know how to react to that. I assure you that I deal with real
organisms, though I admit that my main obsession if figuring out just
how they're related.

>> Do you have any clear idea what my priceless sentence meant?
>
> Sure. It meant that there are a bunch of nodes in the Tree of Life
> that are dated to Cambrian times and you might call the clade that
> originates at one of them a phylum, or you might just move a tad up or
> down and make your division there. For instance, you might include
> Onychophora and Arthropoda in one phylum that starts a few nodes
> further down from each, or you might do things the way they are done
> now.

Even better than that, most of the Cambrian taxa lack one or more of the
characters of the crown group. The Cambrian "onychophorans", for
example, aside from being entirely marine, lack the specialized feeding
apparatus of extant onychophorans. It's been suggested that in most
cases we should consider phyla to be crown groups, and thus many of the
Cambrian fossils would not belong to those groups.

> An even better example is Nematoda/Nemathelminthes, Nematophora and a
> couple of other taxa sometimes called phyla being lumped together in
> Aschelminthes. That's what they did when I was a teenager, and there
> might have been all kinds of compromises between the splitters and
> lumpers since then.
>
> Your turn: do you have any idea why I called it priceless?

No. Why?

>>> The point is, if you call Chordata, Mollusca, Annelida etc. phyla, and
>>> you are a cladistic systematist, then the logical thing to do is to
>>> refer to all clades that diverged from other clades at the same time
>>> these did as "phyla."
>> I'm curious why you think that's "the logical thing to do". Basing ranks
>> on clade age has been suggested a few times, but has generally foundered
>> on the difficulty of deciding ages accurately.
>
> Is't not logical to call clades "phyla" if they branched off from each
> other at nodes that are very close to being the same?

Sure, if you adopt the convention of basing ranks on age. And have the
data to back it up. There will of course be arbitrary dividing lines,
but so are the divisions of the stratigraphic record. However, it isn't
"the logical thing to do", merely one of many potentially convenient
options.

> Anyway, I suppose you haven't got much use for the term "phyla" and so
> for you the whole point is nearly moot.

Sentimental attachment only.

>>> The only exception would be if they are parts
>>> of larger groups that you want to call phyla.
>> In other words, the only exceptions would be if you want to do something
>> else.
>
> Stop oversimplifying. I doubt that anyone would want to get rid of
> Chordata, Mollusca, and Annelida as phyla (even though the last may
> well be paraphyletic) but that is what would happen if you tried to
> trim the number of extinct phyla excessively.

As far as I know, Annelida is monophyletic if you submerge Pogonophora.
(And Echiura too? I forget.)

[John S. Wilkins]

pnyikos <nyi...@bellsouth.net> wrote:

> As usual, you react like a theoretical systematist. Are you as
> theoretical as John Wilkins, who sees no other use for sysematics
> except the reconstruction of phylogenetic trees?

I do what now?

[pnyikos]

On Nov 9, 9:36 pm, j...@wilkins.id.au (John S. Wilkins) wrote:

> pnyikos <nyik...@bellsouth.net> wrote:
> > As usual, you react like a theoretical systematist.  Are you as
> > theoretical as John Wilkins, who sees no other use for sysematics
> > except the reconstruction of phylogenetic trees?
>
> I do what now?

Forgive me if I misinterpreted the following statement:

"The soundness of pre-phylogenetic classifications depended upon the
individual learning sufficient about the group to identify real
groupings within it."

To Harshman, "real groupings" means clades. And what are clades good
for, besides the reconstruction of phylogenetic trees?

[Don't confuse clades or phylogenetic trees with cladograms.
Traditional systematists can appreciate cladograms, from which they
can learn not only the phylogeny but also all kinds of things about
characters. That aids in such things as figuring out where the bases
of Mammalia and Aves lie, the better to delineate the paraphyletic
group Reptilia.]

Peter Nyikos

[John S. Wilkins]

pnyikos <nyi...@bellsouth.net> wrote:

> On Nov 9, 9:36 pm, j...@wilkins.id.au (John S. Wilkins) wrote:
> > pnyikos <nyik...@bellsouth.net> wrote:
> > > As usual, you react like a theoretical systematist. Are you as
> > > theoretical as John Wilkins, who sees no other use for sysematics
> > > except the reconstruction of phylogenetic trees?
> >
> > I do what now?
>
> Forgive me if I misinterpreted the following statement:
>
> "The soundness of pre-phylogenetic classifications depended upon the
> individual learning sufficient about the group to identify real
> groupings within it."
>
> To Harshman, "real groupings" means clades. And what are clades good
> for, besides the reconstruction of phylogenetic trees?

1. Clades <> evolutionary histories in my view (they are indicative, and
test evolutionary histories, but they are not themselves identical).

2. Real groupings are not in and of themselves evolutionary - they are
what they are and evolutionary history is an *explanation* of those
gorupings.

3. "Systematics" is not identical to "phylogenetic systematics"
(cladistics) as the additional qulaifying adjective suggests.

4. I have no idea whatsoever how you went from my comment to "no other
use than"...

>
> [Don't confuse clades or phylogenetic trees with cladograms.

Nor do I.

> Traditional systematists can appreciate cladograms, from which they
> can learn not only the phylogeny but also all kinds of things about
> characters. That aids in such things as figuring out where the bases
> of Mammalia and Aves lie, the better to delineate the paraphyletic
> group Reptilia.]

Here I disagree with you entire. Paraphyly is a statement about the
observer, not the groups observed.

[Ernest Major]

In message <Frudnby2EMp...@giganews.com>, John Harshman
<jhar...@pacbell.net> writes

You have to sink Pogonophora, Echiura and Vestimentifera to maintain
annelid monophyly.
--
alias Ernest Major

[pnyikos]

On Nov 10, 2:37 am, j...@wilkins.id.au (John S. Wilkins) wrote:
> pnyikos <nyik...@bellsouth.net> wrote:
> > On Nov 9, 9:36 pm, j...@wilkins.id.au (John S. Wilkins) wrote:
> > > pnyikos <nyik...@bellsouth.net> wrote:
> > > > As usual, you react like a theoretical systematist.  Are you as
> > > > theoretical as John Wilkins, who sees no other use for sysematics
> > > > except the reconstruction of phylogenetic trees?
>
> > > I do what now?
>
> > Forgive me if I misinterpreted the following statement:
>
> > "The soundness of pre-phylogenetic classifications depended upon the
> > individual learning sufficient about the group to identify real
> > groupings within it."
>
> > To Harshman, "real groupings" means clades.  And what are clades good
> > for, besides the reconstruction of phylogenetic trees?
>
> 1. Clades <> evolutionary histories in my view (they are indicative, and
> test evolutionary histories, but they are not themselves identical).

Does "figuring out evolutionary histories" mean anything in addition
to reconstructing phylogenetic trees?

> 2. Real groupings are not in and of themselves evolutionary - they are
> what they are and evolutionary history is an *explanation* of those
> gorupings.
>
> 3. "Systematics" is not identical to "phylogenetic systematics"
> (cladistics) as the additional qulaifying adjective suggests.

Sure, but if Harshman had his way, they would become identical. What
about you?

> 4. I have no idea whatsoever how you went from my comment to "no other
> use than"...

I went from "identifying real groupings" to "identifying clades" and
the two seemed synonymous to both you and Harshman, and you haven't
done anything to dispel that notion.

So, unless you can give a good answer to my question in response to
your point 1., the only real use for identifying clades is to
reconstruct phylogenetic trees.

>
>
> > [Don't confuse clades or phylogenetic trees with cladograms.
>
> Nor do I.
>
> > Traditional systematists can appreciate cladograms, from which they
> > can learn not only the phylogeny but also all kinds of things about
> > characters.  That aids in such things as figuring out where the bases
> > of Mammalia and Aves lie, the better to delineate the paraphyletic
> > group Reptilia.]
>
> Here I disagree with you entire. Paraphyly is a statement about the
> observer, not the groups observed.

It is a statement about the paraphyletic groups observed. Are you
trying to shoehorn the word "groups" into the definition "clades and
only clades"?

To go from the abstract to the concrete: *Hominini* is universally
considered a "group", because the conventional wisdom is that it is a
clade. But if one of its currently accepted members turned out to be
more closely related to *Pan* than it is to *Homo,* then the label
"Hominini" would be reassigned to a clade which excluded the offending
member.

I prefer looking upon the current members of *Hominini* as a nice
group, even if it does turn out to be paraphyletic.

But that's because I am secondarily a set theorist, and set theorists
look upon sets as being defined by their elements. So the current
*Hominini* is a nice set, and don't think a discovery that it is
paraphyletic would be a good reason for switching the label from it to
another set.

Now, if it were discovered to be polyphyletic, that would be a good
reason for either switching the label or doing away with it
altogether.

Peter Nyikos
Professor, Dept. of Mathematics -- standard disclaimer--
University of South Carolina
http://www.math.sc.edu/~nyikos/
nyikos @ math.sc.edu
Specialty: set-theoretic topology

[John Harshman]

pnyikos wrote:
> On Nov 10, 2:37 am, j...@wilkins.id.au (John S. Wilkins) wrote:
>> pnyikos <nyik...@bellsouth.net> wrote:
>>> On Nov 9, 9:36 pm, j...@wilkins.id.au (John S. Wilkins) wrote:
>>>> pnyikos <nyik...@bellsouth.net> wrote:
>>>>> As usual, you react like a theoretical systematist. Are you as
>>>>> theoretical as John Wilkins, who sees no other use for sysematics
>>>>> except the reconstruction of phylogenetic trees?
>>>> I do what now?
>>> Forgive me if I misinterpreted the following statement:
>>> "The soundness of pre-phylogenetic classifications depended upon the
>>> individual learning sufficient about the group to identify real
>>> groupings within it."
>>> To Harshman, "real groupings" means clades. And what are clades good
>>> for, besides the reconstruction of phylogenetic trees?
>> 1. Clades <> evolutionary histories in my view (they are indicative, and
>> test evolutionary histories, but they are not themselves identical).
>
> Does "figuring out evolutionary histories" mean anything in addition
> to reconstructing phylogenetic trees?

Of course it does. It also refers to the histories of characters, which
must be mapped onto those trees.

>> 2. Real groupings are not in and of themselves evolutionary - they are
>> what they are and evolutionary history is an *explanation* of those
>> gorupings.
>>
>> 3. "Systematics" is not identical to "phylogenetic systematics"
>> (cladistics) as the additional qulaifying adjective suggests.
>
> Sure, but if Harshman had his way, they would become identical. What
> about you?
>
>> 4. I have no idea whatsoever how you went from my comment to "no other
>> use than"...
>
> I went from "identifying real groupings" to "identifying clades" and
> the two seemed synonymous to both you and Harshman, and you haven't
> done anything to dispel that notion.
>
> So, unless you can give a good answer to my question in response to
> your point 1., the only real use for identifying clades is to
> reconstruct phylogenetic trees.

But systematics isn't just identifying clades. That is certainly a major
obsession, but it isn't the only one. Systematics also involves
describing and circumscribing species and may also involve studying
evolution by making use of those trees (i.e. comparative biology).

>>> [Don't confuse clades or phylogenetic trees with cladograms.
>> Nor do I.
>>
>>> Traditional systematists can appreciate cladograms, from which they
>>> can learn not only the phylogeny but also all kinds of things about
>>> characters. That aids in such things as figuring out where the bases
>>> of Mammalia and Aves lie, the better to delineate the paraphyletic
>>> group Reptilia.]
>> Here I disagree with you entire. Paraphyly is a statement about the
>> observer, not the groups observed.
>
> It is a statement about the paraphyletic groups observed. Are you
> trying to shoehorn the word "groups" into the definition "clades and
> only clades"?

He is indeed. That's because when he says "groups" he means "real
groups", and paraphyletic groups aren't real. If they're observed, the
observation is in the mind of the observer.

> To go from the abstract to the concrete: *Hominini* is universally
> considered a "group", because the conventional wisdom is that it is a
> clade. But if one of its currently accepted members turned out to be
> more closely related to *Pan* than it is to *Homo,* then the label
> "Hominini" would be reassigned to a clade which excluded the offending
> member.

No it wouldn't, since Hominini includes Pan, if you will remember.
You're talking about hypothetical relationships *within* Hominini.
Further, these days phylogenetic definitions are becoming popular. Names
aren't reassigned; taxa are moved in and out as estimates of the tree
change.

> I prefer looking upon the current members of *Hominini* as a nice
> group, even if it does turn out to be paraphyletic.

What do you imagine are the current members of Hominini?

> But that's because I am secondarily a set theorist, and set theorists
> look upon sets as being defined by their elements. So the current
> *Hominini* is a nice set, and don't think a discovery that it is
> paraphyletic would be a good reason for switching the label from it to
> another set.

Even if your example were correct, systematists wouldn't generally agree
with you. But that's OK. You don't care what we think, and we don't care
what you think.

> Now, if it were discovered to be polyphyletic, that would be a good
> reason for either switching the label or doing away with it
> altogether.

Why? Wouldn't it still be the same nice set? Your justification seems to
apply as well to polyphyletic as to paraphyletic groups. Perhaps you
meant to say something else.

[John S. Wilkins]

Pretty much what he said:

[pnyikos]

On Nov 28, 7:44 pm, John Harshman <jharsh...@pacbell.net> wrote:
> pnyikos wrote:
> > On Nov 10, 2:37 am, j...@wilkins.id.au (John S. Wilkins) wrote:
> >> pnyikos <nyik...@bellsouth.net> wrote:
> >>> On Nov 9, 9:36 pm, j...@wilkins.id.au (John S. Wilkins) wrote:
> >>>> pnyikos <nyik...@bellsouth.net> wrote:
> >>>>> As usual, you react like a theoretical systematist.  Are you as
> >>>>> theoretical as John Wilkins, who sees no other use for sysematics
> >>>>> except the reconstruction of phylogenetic trees?
> >>>> I do what now?
> >>> Forgive me if I misinterpreted the following statement:
> >>> "The soundness of pre-phylogenetic classifications depended upon the
> >>> individual learning sufficient about the group to identify real
> >>> groupings within it."
> >>> To Harshman, "real groupings" means clades.  And what are clades good
> >>> for, besides the reconstruction of phylogenetic trees?
> >> 1. Clades <> evolutionary histories in my view (they are indicative, and
> >> test evolutionary histories, but they are not themselves identical).
>
> > Does "figuring out evolutionary histories" mean anything in addition
> > to reconstructing phylogenetic trees?
>
> Of course it does. It also refers to the histories of characters, which
> must be mapped onto those trees.

That's a very useful thing to have. And I see that a cladistic
classification gives a bit more motivation for doing this than the
Linnean does.

And so you've given me another reason for recommending a dual system
(like Dewey decimal and Library of Congress for libraries) rather than
advocating a complete return to the Linnean classification.

[...]

> > So, unless you can give a good answer to my question in response to
> > your point 1., the only real use for identifying clades is to
> > reconstruct phylogenetic trees.
>
> But systematics isn't just identifying clades. That is certainly a major
> obsession, but it isn't the only one. Systematics also involves
> describing and circumscribing species

...and, in the Linnean classification, for circumscribing higher taxa.

> and may also involve studying
> evolution by making use of those trees (i.e. comparative biology).

> >>> [Don't confuse clades or phylogenetic trees with cladograms.
> >> Nor do I.

And don't confuse the making of cladograms OR the reconstruction of
phylogenetic trees with your cladophile philosophy of "clades and only
clades are allowed in taxonomy." See what I wrote earlier:

> >>> Traditional systematists can appreciate cladograms, from which they
> >>> can learn not only the phylogeny but also all kinds of things about
> >>> characters.  That aids in such things as figuring out where the bases
> >>> of Mammalia and Aves lie, the better to delineate the paraphyletic
> >>> group Reptilia.]
> >> Here I disagree with you entire. Paraphyly is a statement about the
> >> observer, not the groups observed.
>
> > It is a statement about the paraphyletic groups observed.  Are you
> > trying to shoehorn the word "groups" into the definition "clades and
> > only clades"?
>
> He is indeed. That's because when he says "groups" he means "real
> groups", and paraphyletic groups aren't real. If they're observed, the
> observation is in the mind of the observer.
>
> > To go from the abstract to the concrete:  *Hominini* is universally
> > considered a "group", because the conventional wisdom is that it is a
> > clade.  But if one of its currently accepted members turned out to be
> > more closely related to *Pan* than it is to *Homo,* then  the label
> > "Hominini" would be reassigned to a clade which excluded the offending
> > member.
>
> No it wouldn't, since Hominini includes Pan, if you will remember.

What I recall is that Pan is included in Hominidae. Has it even
graduated to Hominini?

In that case, substitute "Homininae" for "Hominini" in what I said.

> You're talking about hypothetical relationships *within* Hominini.
> Further, these days phylogenetic definitions are becoming popular. Names
> aren't reassigned; taxa are moved in and out as estimates of the tree
> change.

That really puts a burden on naming clades. With so many of them,
they'll soon have to start naming them after people, like the naming
of comets and moon craters, or giving them numbers which give no clue
as to what's in the clades.

> > I prefer looking upon the current members of *Hominini* as a nice
> > group, even if it does turn out to be paraphyletic.
>
> What do you imagine are the current members of Hominini?

I bow to your superior knowledge on this matter.

> > But that's because I am secondarily a set theorist, and set theorists
> > look upon sets as being defined by their elements.  So the current
> > *Hominini*

Substitute *Homininae* if necessary.

> is a nice set, and don't think a discovery that it is
> > paraphyletic would be a good reason for switching the label from it to
> > another set.
>
> Even if your example were correct, systematists wouldn't generally agree
> with you. But that's OK. You don't care what we think, and we don't care
> what you think.

I do care what you think, I just can't see any reason for it that
isn't counterbalanced by the inconveniences; and you've just suggested
another one -- a burden on names.

> > Now, if it were discovered to be polyphyletic, that would be a good
> > reason for either switching the label or doing away with it
> > altogether.
>
> Why? Wouldn't it still be the same nice set?

It would lose its niceness--the reasons for having made it a taxon in
the first place no longer apply.

Once you start admitting polyphyletic groupings, utter chaos could
reign because no one would know where to draw a reasonable line.
What's more, the hierarchical classification species - genus - family
- order - class - phylum - kingdom (with a number of intermediates I
didn't bother to list) could break down.

Paraphyletic groups more stable than the cladistic taxa that result
from them by listing all descendants. They localize the
difficulties better. If it turns out, for instance, that
Archaeopteryx is far less related to birds than has heretofore been
thought, you might have to expunge it from a lot of really big clades,
whereas it is rather isolated in the Linnean classification already.

And so, when paraphyletic groups are allowed, there is not so great a
need to keep up with the latest ideas as to what is descended from
what.

>Your justification seems to
> apply as well to polyphyletic as to paraphyletic groups. Perhaps you
> meant to say something else.

I had additional justifications in mind, including the above, but my
time is limited.

And now more so than ever. As I said in a talk.origins exclusive, I'll
soon be going on a posting break for the holidays.

What's more, I still have a while to go before the grading of exams is
done. This is probably my only Usenet post until some time this
afternoon; but I did want to revisit sci.bio.paleontology, where this
was the latest thread on which I have something I want to say.

Peter Nyikos

[John Harshman]

Not really. Classification, in fact, should have nothing to do with it.
It's the tree that counts, not how you name the pieces of it.

> And so you've given me another reason for recommending a dual system
> (like Dewey decimal and Library of Congress for libraries) rather than
> advocating a complete return to the Linnean classification.

No I haven't, as it turns out.

> [...]
>
>>> So, unless you can give a good answer to my question in response to
>>> your point 1., the only real use for identifying clades is to
>>> reconstruct phylogenetic trees.
>> But systematics isn't just identifying clades. That is certainly a major
>> obsession, but it isn't the only one. Systematics also involves
>> describing and circumscribing species
>
> ...and, in the Linnean classification, for circumscribing higher taxa.

In any classification, actually. But it's a small part, since any
reasonable person would agree that names and ranks are arbitrary.

>> and may also involve studying
>> evolution by making use of those trees (i.e. comparative biology).
>
>>>>> [Don't confuse clades or phylogenetic trees with cladograms.
>>>> Nor do I.
>
> And don't confuse the making of cladograms OR the reconstruction of
> phylogenetic trees with your cladophile philosophy of "clades and only
> clades are allowed in taxonomy." See what I wrote earlier:

Nobody does. But thanks for the advice.

Why would you think that?

>>> I prefer looking upon the current members of *Hominini* as a nice
>>> group, even if it does turn out to be paraphyletic.
>> What do you imagine are the current members of Hominini?
>
> I bow to your superior knowledge on this matter.
>
>>> But that's because I am secondarily a set theorist, and set theorists
>>> look upon sets as being defined by their elements. So the current
>>> *Hominini*
>
> Substitute *Homininae* if necessary.
>
>> is a nice set, and don't think a discovery that it is
>>> paraphyletic would be a good reason for switching the label from it to
>>> another set.
>> Even if your example were correct, systematists wouldn't generally agree
>> with you. But that's OK. You don't care what we think, and we don't care
>> what you think.
>
> I do care what you think, I just can't see any reason for it that
> isn't counterbalanced by the inconveniences; and you've just suggested
> another one -- a burden on names.

I have suggested no such thing.

>>> Now, if it were discovered to be polyphyletic, that would be a good
>>> reason for either switching the label or doing away with it
>>> altogether.
>> Why? Wouldn't it still be the same nice set?
>
> It would lose its niceness--the reasons for having made it a taxon in
> the first place no longer apply.

What are those reasons? I thought that niceness had to do with our
ability to define the group, according to you. Now you suddenly
introduce another criterion, which we might call Mayr-monophyly.

> Once you start admitting polyphyletic groupings, utter chaos could
> reign because no one would know where to draw a reasonable line.

As is the case with paraphyletic groups too.

> What's more, the hierarchical classification species - genus - family
> - order - class - phylum - kingdom (with a number of intermediates I
> didn't bother to list) could break down.

Why would it break down? All you need is an agreement not to have
overlapping sets. Polyphyletic groups are perfectly fine for that.

> Paraphyletic groups more stable than the cladistic taxa that result
> from them by listing all descendants.

How so?

> They localize the
> difficulties better. If it turns out, for instance, that
> Archaeopteryx is far less related to birds than has heretofore been
> thought, you might have to expunge it from a lot of really big clades,
> whereas it is rather isolated in the Linnean classification already.

I think this difficulty too is illusory. Stop viewing with alarm.

> And so, when paraphyletic groups are allowed, there is not so great a
> need to keep up with the latest ideas as to what is descended from
> what.

Sure, because your groups don't have to correspond to anything real, and
therefore don't have to be adjusted to reality. That's an advantage?

>> Your justification seems to
>> apply as well to polyphyletic as to paraphyletic groups. Perhaps you
>> meant to say something else.
>
> I had additional justifications in mind, including the above, but my
> time is limited.

Your time is too limited to allow you to say what you mean?

> And now more so than ever. As I said in a talk.origins exclusive, I'll
> soon be going on a posting break for the holidays.

Merry Christmas etc.

[pnyikos]

On Dec 12, 10:49 am, John Harshman <jharsh...@pacbell.net> wrote:

...a bunch of sound bites, while leaving some of my questions
unanswered.

> pnyikos wrote:
> > On Nov 28, 7:44 pm, John Harshman <jharsh...@pacbell.net> wrote:
> >> pnyikos wrote:

[snip lots of shop talk, more suitable for sci.bio.systematics if only
Giganews still carried it]

> >>> To go from the abstract to the concrete:  *Hominini* is universally
> >>> considered a "group", because the conventional wisdom is that it is a
> >>> clade.  But if one of its currently accepted members turned out to be
> >>> more closely related to *Pan* than it is to *Homo,* then  the label
> >>> "Hominini" would be reassigned to a clade which excluded the offending
> >>> member.
> >> No it wouldn't, since Hominini includes Pan, if you will remember.
>
> > What I recall is that Pan is included in Hominidae.  Has it even
> > graduated to Hominini?

Well, which of us recalled correctly?

Here's an even more radical thought, motivated by your reminding me
that *Paranthropus* (a.k.a. "Zinjanthropus," a.k.a. Australopictheus
boisei) had the teeth of an extreme vegetarian. Is it conceivable
that its sister taxon is *Gorilla* rather than any of the Hominini?

> > In that case, substitute "Homininae" for "Hominini" in what I said.

Am I correct in assuming that *Pan* has not been put in the subfamily
*Homininae*, at least not yet?

Now we return to systematics:

> >> You're talking about hypothetical relationships *within* Hominini.
> >> Further, these days phylogenetic definitions are becoming popular. Names
> >> aren't reassigned; taxa are moved in and out as estimates of the tree
> >> change.
>
> > That really puts a burden on naming clades. With so many of them,
> > they'll soon have to start naming them after people, like the naming
> > of comets and moon craters, or giving them numbers which give no clue
> > as to what's in the clades.
>
> Why would you think that?

Because each new taxon gives rise to a new clade, and unless it
belongs to a known genus the clade seems to need a name. After all, if
taxa are moved in or out of it, it is the name that provides
continuity.
[...]

> >>>I am secondarily a set theorist, and set theorists
> >>> look upon sets as being defined by their elements. So the current
> >>> *Hominini*
>
> > Substitute *Homininae* if necessary.
>
> >> is a nice set, and don't think a discovery that it is
> >>> paraphyletic would be a good reason for switching the label from it to
> >>> another set.

[...]

> >>> Now, if it were discovered to be polyphyletic, that would be a good
> >>> reason for either switching the label or doing away with it
> >>> altogether.
> >> Why? Wouldn't it still be the same nice set?

[...]

> >> Your justification seems to
> >> apply as well to polyphyletic as to paraphyletic groups. Perhaps you
> >> meant to say something else.
>
> > I had additional justifications in mind, including the above, but my
> > time is limited.
>
> Your time is too limited to allow you to say what you mean?

In this case, "what you mean" is an expression which corresponds to
data better described by "various arguments for your POV" and they
could go on for thousands of lines, so yes, my time is too limited.

You somehow decided from the way I worded something that it was my
ONLY justification for liking the group *Homininae* even if it is
paraphyletic. But it isn't. And I can't be constantly trying to read
your mind as to the effect various wordings will have on you.

Peter Nyikos

[pnyikos]

They do have to be adjusted to reality. If a new dinosaur, easily
seen not to be a bird, is found, it has to be put into the
paraphyletic group Archosauria under the traditional classification.

I think you are getting too carried away with your artificial use of
the word "unreal", which is a pejorative term for "not a clade" and
nothing more.

[big snip. too short on time]

> Merry Christmas etc.

Happy Winter Solstice and New Year.

Did I find the wish that corresponds best to your view of the holiday
season?

Peter Nyikos

[John S. Wilkins]

pnyikos <nyi...@bellsouth.net> wrote:

> > > What I recall is that Pan is included in Hominidae. Has it even
> > > graduated to Hominini?
>
> Well, which of us recalled correctly?

Which ever one said Pan had been included in Hominini

http://en.wikipedia.org/wiki/Hominini

[John Harshman]

pnyikos wrote:
> On Dec 12, 10:49 am, John Harshman <jharsh...@pacbell.net> wrote:
>
> ...a bunch of sound bites, while leaving some of my questions
> unanswered.
>
>> pnyikos wrote:
>>> On Nov 28, 7:44 pm, John Harshman <jharsh...@pacbell.net> wrote:
>>>> pnyikos wrote:
>
> [snip lots of shop talk, more suitable for sci.bio.systematics if only
> Giganews still carried it]
>
>
>>>>> To go from the abstract to the concrete: *Hominini* is universally
>>>>> considered a "group", because the conventional wisdom is that it is a
>>>>> clade. But if one of its currently accepted members turned out to be
>>>>> more closely related to *Pan* than it is to *Homo,* then the label
>>>>> "Hominini" would be reassigned to a clade which excluded the offending
>>>>> member.
>>>> No it wouldn't, since Hominini includes Pan, if you will remember.
>>> What I recall is that Pan is included in Hominidae. Has it even
>>> graduated to Hominini?
>
> Well, which of us recalled correctly?

I did. (But remember that there is no such thing as the official
classification. There are several definitions of these various names out
there. I merely use the currently most common ones.) And what do you
mean by "graduated"? Are you falling into the scala naturae fallacy?

> Here's an even more radical thought, motivated by your reminding me
> that *Paranthropus* (a.k.a. "Zinjanthropus," a.k.a. Australopictheus
> boisei) had the teeth of an extreme vegetarian. Is it conceivable
> that its sister taxon is *Gorilla* rather than any of the Hominini?

It's certainly conceivable. But is it at all likely? No.

>>> In that case, substitute "Homininae" for "Hominini" in what I said.
>
> Am I correct in assuming that *Pan* has not been put in the subfamily
> *Homininae*, at least not yet?

No.

> Now we return to systematics:
>
>>>> You're talking about hypothetical relationships *within* Hominini.
>>>> Further, these days phylogenetic definitions are becoming popular. Names
>>>> aren't reassigned; taxa are moved in and out as estimates of the tree
>>>> change.
>>> That really puts a burden on naming clades. With so many of them,
>>> they'll soon have to start naming them after people, like the naming
>>> of comets and moon craters, or giving them numbers which give no clue
>>> as to what's in the clades.
>> Why would you think that?
>
> Because each new taxon gives rise to a new clade, and unless it
> belongs to a known genus the clade seems to need a name. After all, if
> taxa are moved in or out of it, it is the name that provides
> continuity.

Does not compute. Why should every possible cladistic taxon need to be
named? If anything, this is an argument against your system, because if
every taxon needs a name, there are many more names required for
paraphyletic taxa than for monophyletic taxa.

So say what you mean. If you state your criterion for some preference, I
should not be required to read your mind and fail to object just because
you might have other unexpressed criteria.

[John Harshman]

pnyikos wrote:
> On Dec 12, 10:49 am, John Harshman <jharsh...@pacbell.net> wrote:
>> pnyikos wrote:
>
>>> when paraphyletic groups are allowed, there is not so great a
>>> need to keep up with the latest ideas as to what is descended from
>>> what.
>> Sure, because your groups don't have to correspond to anything real, and
>> therefore don't have to be adjusted to reality.
>
> They do have to be adjusted to reality. If a new dinosaur, easily
> seen not to be a bird, is found, it has to be put into the
> paraphyletic group Archosauria under the traditional classification.

No it doesn't. Suppose somebody thought it had some major advanced
character that required it to be removed from Archosauria, because it
was just too distinct from the standard archosaur.

> I think you are getting too carried away with your artificial use of
> the word "unreal", which is a pejorative term for "not a clade" and
> nothing more.

It isn't pejorative. It's just that only clades are real. Other groups
are arbitrary.

> [big snip. too short on time]
>
>> Merry Christmas etc.
>
> Happy Winter Solstice and New Year.
>
> Did I find the wish that corresponds best to your view of the holiday
> season?

Christmas is a secular holiday. Let's face it.

[Steven L.]

"pnyikos" <nyi...@bellsouth.net> wrote in message
news:e409069b-d68f-45df...@w1g2000vba.googlegroups.com:

My wish that corresponds best to my view of this time of the year is as
follows:

"I hope I recover swiftly from this Pseudomonas aeroginosa sinus
infection."




-- Steven L.

[Michael Ige]

if you are looking for a fast and quick way to cure the sinus
infection then .I can give you a detailed break down and analyses of
how you can do that, the symptoms and how it can be transmitted
check out this website below for more info
http://sinusinfectionremediesinfo.com/


On Dec 15, 9:11 am, "Steven L." <sdlit...@earthlink.net> wrote:
> "pnyikos" <nyik...@bellsouth.net> wrote in message

[Friar Broccoli]

On 2011-12-23 11:56, Michael Ige wrote:
> if you are looking for a fast and quick way to cure the sinus
> infection then .I can give you a detailed break down and analyses of
> how you can do that, the symptoms and how it can be transmitted
> check out this website below for more info
> http://sinusinfectionremediesinfo.com/

As a lifelong suffer from chronic sinus infections, I must say I was
very far from impressed by the above. (Not false, just not effective)

What works for me is an effective decongestant applied (reasonably)
aggressively as soon as I become aware of a problem. Recently I have
discovered this:

http://www.drugs.com/mmx/xylometazoline-hydrochloride.html
(just Sudafed in a spray)

which is very effective as a first response, although because it causes
secondary reactions when overused, you need to switch to a less
aggressive sinus (pill) decongestant sometimes for a period of weeks
thereafter.

--
Friar Broccoli (Robert Keith Elias), Quebec Canada
I consider ALL arguments in support of my views

[pnyikos]

On Dec 13 2011, 1:26 pm, John Harshman <jharsh...@pacbell.net> wrote:
> pnyikos wrote:
> > On Dec 12, 10:49 am, John Harshman <jharsh...@pacbell.net> wrote:
> >> pnyikos wrote:
>
> >>> when paraphyletic groups are allowed, there is not so great a
> >>> need to keep up with the latest ideas as to what is descended from
> >>> what.
> >> Sure, because your groups don't have to correspond to anything real, and
> >> therefore don't have to be adjusted to reality.
>
> > They do have to be adjusted to reality.  If a new dinosaur, easily
> > seen not to be a bird, is found, it has to be put into the
> > paraphyletic group Archosauria under the traditional classification.
>
> No it doesn't. Suppose somebody thought it had some major advanced
> character that required it to be removed from Archosauria, because it
> was just too distinct from the standard archosaur.

In that case, one would either push Aves further down in the tree than
it is at present, to incorporate the critter, or if the MAJOR advanced
character was so stupendous that it merited a new class, it would be
given one.

However, the person proposing it would have to be willing to do a lot
of explanations to skeptics, if he wanted the new class to be adopted.
Failing that, it would simply be dumped into Archosauria.

If not just one, but many diverse species were found that displayed
the major character, the class-founder-wannabe would have a much
easier time in getting the idea adopted.

Trivia: there was once (1960's) an elaborate book about a fictitious
island [although the author played it straight, just like Jonathan
Swift did with his islands] where a family of mammals called
"snouters" exploded in a veritable orgy of evolution, to the point
where some species were far more different from non-snouters than even
bats are from all other known mammals.

At some point, the author could easily have said, "the descendants of
the last common ancestor of these species merit placement in a new
class".

And maybe he did, but the species in the book that merited such a
treatment were so outlandish, I closed the book and returned it to the
shelf of the bookstore where I was browsing through it.

> > I think you are getting too carried away with your artificial use of
> > the word "unreal", which is a pejorative term for "not a clade" and
> > nothing more.
>
> It isn't pejorative. It's just that only clades are real. Other groups
> are arbitrary.

They are real, but only certain groups were adopted by the
systematists of old, who included lots of old-fashioned naturalists.

Naturalists are a dying breed, which is a great shame. Their passing
can only widen the gulf between the two cultures that C P Snow wrote
about, the sciences and the humanities:

http://books.google.com/books/about/The_two_cultures.html?id=OyHm4sc6IPoC

> > [big snip.  too short on time]
>
> >> Merry Christmas etc.
>
> > Happy Winter Solstice and New Year.
>
> > Did I find the wish that corresponds best to your view of the holiday
> > season?
>
> Christmas is a secular holiday. Let's face it.

It is to most people, granted. But many atheists are clueless about
how to make it even more secular than it already is without stirring
up a lot of backlash.

Instead of in-your-face exhibits blasting Christianity (and sometimes
other religions) in competition with creches, menorahs, etc. they
could do some charming and informative Winter Solstice exhibits, like
a mock-up of Stonehenge, or a model of the earth and the sun showing
their relationship during the Winter Solstice in the Northern
Hemisphere.

Peter Nyikos

[John Harshman]

Which helps to illustrate how silly the idea of rank is in systematics.

> Trivia: there was once (1960's) an elaborate book about a fictitious
> island [although the author played it straight, just like Jonathan
> Swift did with his islands] where a family of mammals called
> "snouters" exploded in a veritable orgy of evolution, to the point
> where some species were far more different from non-snouters than even
> bats are from all other known mammals.

You're talking about the Rhinogradentia. It's famous, at least among
biologists.

> At some point, the author could easily have said, "the descendants of
> the last common ancestor of these species merit placement in a new
> class".
>
> And maybe he did, but the species in the book that merited such a
> treatment were so outlandish, I closed the book and returned it to the
> shelf of the bookstore where I was browsing through it.

Pity.

>>> I think you are getting too carried away with your artificial use of
>>> the word "unreal", which is a pejorative term for "not a clade" and
>>> nothing more.
>> It isn't pejorative. It's just that only clades are real. Other groups
>> are arbitrary.
>
> They are real, but only certain groups were adopted by the
> systematists of old, who included lots of old-fashioned naturalists.
>
> Naturalists are a dying breed, which is a great shame. Their passing
> can only widen the gulf between the two cultures that C P Snow wrote
> about, the sciences and the humanities:
>
> http://books.google.com/books/about/The_two_cultures.html?id=OyHm4sc6IPoC

You seem to have wandered into a digression without addressing my point.

>>> [big snip. too short on time]
>>>> Merry Christmas etc.
>>> Happy Winter Solstice and New Year.
>>> Did I find the wish that corresponds best to your view of the holiday
>>> season?
>> Christmas is a secular holiday. Let's face it.
>
> It is to most people, granted. But many atheists are clueless about
> how to make it even more secular than it already is without stirring
> up a lot of backlash.
>
> Instead of in-your-face exhibits blasting Christianity (and sometimes
> other religions) in competition with creches, menorahs, etc. they
> could do some charming and informative Winter Solstice exhibits, like
> a mock-up of Stonehenge, or a model of the earth and the sun showing
> their relationship during the Winter Solstice in the Northern
> Hemisphere.

You really do just say anything that comes into your head, don't you?

[pnyikos]

Dead wrong. You are failing to appreciate the role of the word
"diverse" in the sentence to which you are responding: the more
diverse, the better a case for making it a class rather than a family,
order, etc.

> > Trivia: there was once (1960's) an elaborate book about a fictitious
> > island [although the author played it straight, just like Jonathan
> > Swift did with his islands] where a family of mammals called
> > "snouters" exploded in  a veritable orgy of evolution, to the point
> > where some species were far more different from non-snouters than even
> > bats are from all other known mammals.
>
> You're talking about the Rhinogradentia. It's famous, at least among
> biologists.
>
> > At some point, the author could easily have said, "the descendants of
> > the last common ancestor of these species merit placement in a new
> > class".
>
> > And maybe he did, but the species in the book that merited such a
> > treatment were so outlandish, I closed the book and returned it to the
> > shelf of the bookstore where I was browsing through it.
>
> Pity.

You liked the book a lot? Why?

More on topic: don't you see how it illustrated the concept of some
creatures meriting a class distinct from the class mammalia to which
the primitive snouters still belonged?

> >>> I think you are getting too carried away with your artificial use of
> >>> the word "unreal", which is a pejorative term for "not a clade" and
> >>> nothing more.
> >> It isn't pejorative. It's just that only clades are real. Other groups
> >> are arbitrary.
>
> > They are real, but only certain groups were adopted by the
> > systematists of old, who included lots of old-fashioned naturalists.
>
> > Naturalists are a dying breed, which is a great shame.  Their passing
> > can only widen the gulf between the two cultures that C P Snow  wrote
> > about, the sciences and the humanities:
>
> >http://books.google.com/books/about/The_two_cultures.html

> You seem to have wandered into a digression without addressing my point.

On the contrary, it is the old fashioned naturalists who could best
appreciate how the concept of ranks helps people organize their
thinking about the biological world. Most science books for laymen
nowadays lack the naturalists' touch, and so it is hard for a layman
to get a feel for the big picture.

Peter Nyikos

[John Harshman]

Why?

>>> Trivia: there was once (1960's) an elaborate book about a fictitious
>>> island [although the author played it straight, just like Jonathan
>>> Swift did with his islands] where a family of mammals called
>>> "snouters" exploded in a veritable orgy of evolution, to the point
>>> where some species were far more different from non-snouters than even
>>> bats are from all other known mammals.
>> You're talking about the Rhinogradentia. It's famous, at least among
>> biologists.
>>
>>> At some point, the author could easily have said, "the descendants of
>>> the last common ancestor of these species merit placement in a new
>>> class".
>>> And maybe he did, but the species in the book that merited such a
>>> treatment were so outlandish, I closed the book and returned it to the
>>> shelf of the bookstore where I was browsing through it.
>> Pity.
>
> You liked the book a lot? Why?

Because it shows how bizarre evolution can be. Exaggerated, perhaps. But
not by much.

> More on topic: don't you see how it illustrated the concept of some
> creatures meriting a class distinct from the class mammalia to which
> the primitive snouters still belonged?

No. Why?

>>>>> I think you are getting too carried away with your artificial use of
>>>>> the word "unreal", which is a pejorative term for "not a clade" and
>>>>> nothing more.
>>>> It isn't pejorative. It's just that only clades are real. Other groups
>>>> are arbitrary.
>>> They are real, but only certain groups were adopted by the
>>> systematists of old, who included lots of old-fashioned naturalists.
>>> Naturalists are a dying breed, which is a great shame. Their passing
>>> can only widen the gulf between the two cultures that C P Snow wrote
>>> about, the sciences and the humanities:
>>> http://books.google.com/books/about/The_two_cultures.html
>
>> You seem to have wandered into a digression without addressing my point.
>
> On the contrary, it is the old fashioned naturalists who could best
> appreciate how the concept of ranks helps people organize their
> thinking about the biological world. Most science books for laymen
> nowadays lack the naturalists' touch, and so it is hard for a layman
> to get a feel for the big picture.

Personally, I think that your favorite grades only obscure evolution and
so disguise the big picture.

[pnyikos]

On Jan 31, 8:54 am, John Harshman <jharsh...@pacbell.net> wrote:
> pnyikos wrote:
> > On Jan 13, 7:10 pm, John Harshman <jharsh...@pacbell.net> wrote:
> >> pnyikos wrote:
> >>> On Dec 13 2011, 1:26 pm, John Harshman <jharsh...@pacbell.net> wrote:
> >>>> pnyikos wrote:

...about paraphyletic taxa:

> >>>>> They do have to be adjusted to reality.  If a new dinosaur, easily
> >>>>> seen not to be a bird, is found, it has to be put into the
> >>>>> paraphyletic group Archosauria under the traditional classification.
>
> >>>> No it doesn't. Suppose somebody thought it had some major advanced
> >>>> character that required it to be removed from Archosauria, because it
> >>>> was just too distinct from the standard archosaur.
>
> >>> In that case, one would either push Aves further down in the tree than
> >>> it is at present, to incorporate the critter, or if the MAJOR advanced
> >>> character was so stupendous that it merited a new class, it would be
> >>> given one.
> >>> However,  the person proposing it would have to be willing to do a lot
> >>> of explanations to skeptics, if he wanted the new class to be adopted.
> >>> Failing that,  it would simply be dumped into Archosauria.
> >>> If not just one, but many diverse species were found that displayed
> >>> the major character, the class-founder-wannabe  would have a much
> >>> easier time in getting the idea adopted.
>
> >> Which helps to illustrate how silly the idea of rank is in systematics.
>
> > Dead wrong.  You are failing to appreciate the role of the word
> > "diverse" in the sentence to which you are responding: the  more
> > diverse, the better a case for making it a class rather than a family,
> > order, etc.
>
> Why?

To better serve the general public, who can better appreciate a
diverse class of things than a whole big category of one or two
things, remote from all others.

Ivory tower types may see things differently, but if they leave the
public behind, they also risk losing public funding, without which
many branches of biology are in danger of withering away. Even with
all the popular excitement about dinosaurs [as distinct from fossil
Aves], vertebrate paleontologists are one such breed of biologist.

John Horner does well because he is a popularizer. Vertebrate
paleontologists studying less popular taxa have a hard time making a
career out of it.

> >>> Trivia: there was once (1960's) an elaborate book about a fictitious
> >>> island [although the author played it straight, just like Jonathan
> >>> Swift did with his islands] where a family of mammals called
> >>> "snouters" exploded in  a veritable orgy of evolution, to the point
> >>> where some species were far more different from non-snouters than even
> >>> bats are from all other known mammals.
> >> You're talking about the Rhinogradentia. It's famous, at least among
> >> biologists.
>
> >>> At some point, the author could easily have said, "the descendants of
> >>> the last common ancestor of these species merit placement in a new
> >>> class".
> >>> And maybe he did, but the species in the book that merited such a
> >>> treatment were so outlandish, I closed the book and returned it to the
> >>> shelf of the bookstore where I was browsing through it.
> >> Pity.
>
> > You liked the book a lot?  Why?
>
> Because it shows how bizarre evolution can be. Exaggerated, perhaps. But
> not by much.
>
> > More on topic: don't you see how it illustrated the concept of some
> > creatures meriting a class distinct from the class mammalia to which
> > the primitive snouters still belonged?
>
> No. Why?

I'm afraid you are too tone-deaf at present for me to make music here
for you.

[...]

>... it is the old fashioned naturalists who could best

> > appreciate how the concept of ranks helps people organize their
> > thinking about the biological world.  Most science books for laymen
> > nowadays lack the naturalists' touch, and so it is hard for a layman
> > to get a feel for the big picture.
>
> Personally, I think that your favorite grades only obscure evolution and
> so disguise the big picture.

And I think your cladophile ideas are the ones that obscure evolution
by maintaining what Gould called "the dirty little secret": you have
"real organisms" only at the tips of your trees, never the nodes.
With classical taxa, you can say "the species in X descended from some
member of Y" without having to specify which member.

We can zero in on a paraphyletic Maniraptora that excludes Aves and
say that we have very good evidence that all living birds descended
ultimately from a Maniraptoran.

Then if the evidence suddenly starts to show something else, we can
name some other taxon that we originally thought to be a clade, and
say that the evidence now shows that this taxon was paraphyletic,
because now the evidence points to Aves being descended from this
taxon.

That's just one example of many. One case where something like the
above has already taken place is that, at one time, the best available
evidence suggested that Mesonychia was paraphyletic and that Cetacea
is descended from one of its members. Now the evidence shows
otherwise.

The only difference is that, with Linnean taxa having fallen out of
fashion, nobody seems to have bothered to set up a paraphyletic taxon
from which whales are now believed to have descended. We have to
content ourselves with the miserably unhelpful "whales are
artiodactyls, and their nearest living relative is believed to be the
hippopotamus.."

Peter Nyikos

[John Harshman]

You are confusing names with ranks again. For the general public, or for
anyone, it's enough to give a name to a clade in order to draw attention
to it. Cladistic classification serves a major purpose: it highlights
cladistic relationships and so emphasizes evolution, i.e. common descent.

> Ivory tower types may see things differently, but if they leave the
> public behind, they also risk losing public funding, without which
> many branches of biology are in danger of withering away. Even with
> all the popular excitement about dinosaurs [as distinct from fossil
> Aves], vertebrate paleontologists are one such breed of biologist.

Ah, a little populism from the esteemed Prof. N. I think we can take
that risk. There is no ivory tower involved.

> John Horner does well because he is a popularizer. Vertebrate
> paleontologists studying less popular taxa have a hard time making a
> career out of it.

What a string of non sequiturs we have here. From naming clades only to
making things fun for hoi polloi to popularizing science to studying
popular groups.

>>>>> Trivia: there was once (1960's) an elaborate book about a fictitious
>>>>> island [although the author played it straight, just like Jonathan
>>>>> Swift did with his islands] where a family of mammals called
>>>>> "snouters" exploded in a veritable orgy of evolution, to the point
>>>>> where some species were far more different from non-snouters than even
>>>>> bats are from all other known mammals.
>>>> You're talking about the Rhinogradentia. It's famous, at least among
>>>> biologists.
>>>>> At some point, the author could easily have said, "the descendants of
>>>>> the last common ancestor of these species merit placement in a new
>>>>> class".
>>>>> And maybe he did, but the species in the book that merited such a
>>>>> treatment were so outlandish, I closed the book and returned it to the
>>>>> shelf of the bookstore where I was browsing through it.
>>>> Pity.
>>> You liked the book a lot? Why?
>> Because it shows how bizarre evolution can be. Exaggerated, perhaps. But
>> not by much.
>>
>>> More on topic: don't you see how it illustrated the concept of some
>>> creatures meriting a class distinct from the class mammalia to which
>>> the primitive snouters still belonged?
>> No. Why?
>
> I'm afraid you are too tone-deaf at present for me to make music here
> for you.

In other words, you have no argument except "it's just so obvious". Why
should bats be included in mammals while birds are separated from reptiles?

>> ... it is the old fashioned naturalists who could best
>>> appreciate how the concept of ranks helps people organize their
>>> thinking about the biological world. Most science books for laymen
>>> nowadays lack the naturalists' touch, and so it is hard for a layman
>>> to get a feel for the big picture.
>> Personally, I think that your favorite grades only obscure evolution and
>> so disguise the big picture.
>
> And I think your cladophile ideas are the ones that obscure evolution
> by maintaining what Gould called "the dirty little secret": you have
> "real organisms" only at the tips of your trees, never the nodes.
> With classical taxa, you can say "the species in X descended from some
> member of Y" without having to specify which member.

That has nothing to do with real organisms being at tips only. You are
very confused. There are no real organisms in "some member of Y". And
it's just as true that monkeys, for example, descended from some member
of Primates as that they descended from some paraphyletic group of
"non-monkeys", so cladistic classification incorporates your needs, just
making it clearer that it's a strange thing to say.

> We can zero in on a paraphyletic Maniraptora that excludes Aves and
> say that we have very good evidence that all living birds descended
> ultimately from a Maniraptoran.

Which says nothing other than that birds are maniraptorans. And while
this may be subjective, to my ear the latter contains more of the wonder
of evolution than the former. As in "humans are odd fish".

> Then if the evidence suddenly starts to show something else, we can
> name some other taxon that we originally thought to be a clade, and
> say that the evidence now shows that this taxon was paraphyletic,
> because now the evidence points to Aves being descended from this
> taxon.

Why would we want to do any such thing, and how does cladistic
classification differ from this in any significant way? If we find
Maniraptora not to contain birds, and some other group to contain birds,
we just tranfer birds to the new group.

> That's just one example of many.

Example of what? Special pleading, it seems to me.

> One case where something like the
> above has already taken place is that, at one time, the best available
> evidence suggested that Mesonychia was paraphyletic and that Cetacea
> is descended from one of its members. Now the evidence shows
> otherwise.
>
> The only difference is that, with Linnean taxa having fallen out of
> fashion, nobody seems to have bothered to set up a paraphyletic taxon
> from which whales are now believed to have descended.

That's because we don't need paraphyletic taxa. Why is that a bad thing?
Isn't it so much cooler to know that whales are artiodactyls and that
their living sister group is hippos?

> We have to
> content ourselves with the miserably unhelpful "whales are
> artiodactyls, and their nearest living relative is believed to be the
> hippopotamus.."

Why is that either miserable or unhelpful? It tell us a lot about
evolution, much more than erecting pointless paraphyletic groups would.
>
> Peter Nyikos
>

[pnyikos]

I'd lost track of this thread for a long time. There are still some
things in it that I think would benefit from a further discussion.

On Feb 21, 3:08 pm, John Harshman <jharsh...@pacbell.net> wrote:
> pnyikos wrote:
> > On Jan 31, 8:54 am, John Harshman <jharsh...@pacbell.net> wrote:
> >> pnyikos wrote:

> > Ivory tower types may see things differently, but if they leave the
> > public behind, they also risk losing public funding, without which
> > many branches of biology are in danger of withering away.  Even with
> > all the popular excitement about dinosaurs [as distinct from fossil
> > Aves], vertebrate paleontologists are one such breed of biologist.
>
> Ah, a little populism from the esteemed Prof. N. I think we can take
> that risk. There is no ivory tower involved.
>
> > John Horner does well because he is a popularizer.  Vertebrate
> > paleontologists studying less popular taxa have a hard time making a
> > career out of it.
>
> What a string of non sequiturs we have here. From naming clades only to
> making things fun for hoi polloi to popularizing science to studying
> popular groups.

These are all valid themes. Calling them "non sequiturs" does not
take away from their significance, only from my writing style.

[...]

> >>> More on topic: don't you see how it illustrated the concept of some
> >>> creatures meriting a class distinct from the class mammalia to which
> >>> the primitive snouters still belonged?
> >> No. Why?
>
> > I'm afraid you are too tone-deaf at present for me to make music here
> > for you.
>
> In other words, you have no argument except "it's just so obvious". Why
> should bats be included in mammals while birds are separated from reptiles?

Partly because there is a whole subclass, Dinosauria, intervening
between birds and reptiles, and nothing like it between bats and other
mammals.

In fact, the intervening ones between superficially shrew-like mammals
and full-fledged bats is "purely hypothetical" in your terminology
because we have none of their fossils.

> >> ...  it is the old fashioned naturalists who could best
> >>> appreciate how the concept of ranks helps people organize their
> >>> thinking about the biological world.  Most science books for laymen
> >>> nowadays lack the naturalists' touch, and so it is hard for a layman
> >>> to get a feel for the big picture.
>
> >> Personally, I think that your favorite grades only obscure evolution and
> >> so disguise the big picture.
>
> > And I think your cladophile ideas are the ones that obscure evolution
> > by maintaining what Gould called "the dirty little secret": you have
> > "real organisms" only at the tips of your trees, never the nodes.
> > With classical taxa, you can say "the species in X descended from some
> > member of Y" without having to specify which member.
>
> That has nothing to do with real organisms being at tips only. You are
> very confused.

Not at all; you just don't follow my line of thought, due to some
confusion of your own:

>There are no real organisms in "some member of Y".

There are real organisms in Y even by your standards. It is just that
old-fashioned taxonomists acknowledge the fact -- or as close to fact
as any talk of "X is the sister group of Y" can ever get-- that the
organisms in X are descended from some organisms which fit all the
criteria for Y.

e.g. Aves = X, Maniraptora = Y

and some day we may be able to narrow Y down more.

> And
> it's just as true that monkeys, for example, descended from some member
> of Primates

They are, themselves, members of Primates.

> as that they descended from some paraphyletic group of
> "non-monkeys", so cladistic classification incorporates your needs,

Correction: cladistics *per se* incorporates my needs, cladophile
classification obscures them.

> just making it clearer that it's a strange thing to say.

Not as strange as saying that monkeys are descended from primates, nor
as strange as saying that the sister group of John Harshman is John
Harshman's mother, if John Harshman was an only child.

[John Harshman's father is a close second, but he didn't contribute
any mitochondria to John.]

> > We can zero in on a paraphyletic Maniraptora that excludes Aves and
> > say that we have very good evidence that all living birds descended
> > ultimately from a Maniraptoran.
>
> Which says nothing other than that birds are maniraptorans.

...thereby being totally unhelpful to anyone who wants to see what a
good old-fashioned pre-aves maniraptoran looked like.

I can't even say "pre-aves maniraptoran" without being accused of
talking about a fictitious group.

> And while
> this may be subjective, to my ear the latter contains more of the wonder
> of evolution than the former. As in "humans are odd fish".

Poetry is one thing, baiting a hook for a fish is another.

> > Then if the evidence suddenly starts to show something else, we can
> > name some other taxon that we originally thought to be a clade, and
> > say that the evidence now shows that this taxon was paraphyletic,
> > because now the evidence points to Aves being descended from this
> > taxon.
>
> Why would we want to do any such thing, and how does cladistic
> classification differ from this in any significant way? If we find
> Maniraptora not to contain birds, and some other group to contain birds,
> we just tranfer birds to the new group.

...and thereby utterly change what the word "maniraptoran" refers to.
Old-fashioned taxonomists are blissfully free of having to indulge in
such wholesale rewriting of books and websites.

[...]

> > One case where something like the
> > above has already taken place is that, at one time, the best available
> > evidence suggested that Mesonychia was paraphyletic and that Cetacea
> > is descended from one of its members.  Now the evidence shows
> > otherwise.

And Mesonychia still means exactly what it meant before the change, to
a traditional cladist. The same animals were mesonychids that were
mesonychids before the change.

> > The only difference is that, with Linnean taxa having fallen out of
> > fashion, nobody seems to have bothered to set up a paraphyletic taxon
> > from which whales are now believed to have descended.
>
> That's because we don't need paraphyletic taxa. Why is that a bad thing?
> Isn't it so much cooler to know that whales are artiodactyls and that
> their living sister group is hippos?

Not any cooler than knowing that whales are descended from
artiodactyls, and that among *known* animals, their sister group is
hippos.

But neither fact says anything about how much hippos resemble the LCA
of *Pakicetus* and the hippo. An old fashioned Linnean would look at
what we'd be able to reconstruct of that LCA, and tell us whether it
and the hippo deserve to be in the same family, and whether
*Pakicetus* could safely be assigned to the genus of the LCA, leaving
the species of the LCA "purely hypothetical" as a sop to cladophiles
like you..

> > We have to
> > content ourselves with the miserably unhelpful "whales are
> > artiodactyls, and their nearest living relative is believed to be the
> > hippopotamus.."
>
> Why is that either miserable or unhelpful? It tell us a lot about
> evolution, much more than erecting pointless paraphyletic groups would.

Wrong, see above.

Peter Nyikos

[John Harshman]

pnyikos wrote:
> I'd lost track of this thread for a long time. There are still some
> things in it that I think would benefit from a further discussion.
>
> On Feb 21, 3:08 pm, John Harshman <jharsh...@pacbell.net> wrote:
>> pnyikos wrote:
>>> On Jan 31, 8:54 am, John Harshman <jharsh...@pacbell.net> wrote:
>>>> pnyikos wrote:
>
>>> Ivory tower types may see things differently, but if they leave the
>>> public behind, they also risk losing public funding, without which
>>> many branches of biology are in danger of withering away. Even with
>>> all the popular excitement about dinosaurs [as distinct from fossil
>>> Aves], vertebrate paleontologists are one such breed of biologist.
>> Ah, a little populism from the esteemed Prof. N. I think we can take
>> that risk. There is no ivory tower involved.
>>
>>> John Horner does well because he is a popularizer. Vertebrate
>>> paleontologists studying less popular taxa have a hard time making a
>>> career out of it.
>> What a string of non sequiturs we have here. From naming clades only to
>> making things fun for hoi polloi to popularizing science to studying
>> popular groups.
>
> These are all valid themes. Calling them "non sequiturs" does not
> take away from their significance, only from my writing style.

It isn't significance I'm complaining about, nor is it your writing
style per se. It's the lack of any valid connection from one topic to
another and the conflation of topics.

>>>>> More on topic: don't you see how it illustrated the concept of some
>>>>> creatures meriting a class distinct from the class mammalia to which
>>>>> the primitive snouters still belonged?
>>>> No. Why?
>>> I'm afraid you are too tone-deaf at present for me to make music here
>>> for you.
>> In other words, you have no argument except "it's just so obvious". Why
>> should bats be included in mammals while birds are separated from reptiles?
>
> Partly because there is a whole subclass, Dinosauria, intervening
> between birds and reptiles, and nothing like it between bats and other
> mammals.

So one arbitrary convention begets an objective reason for another? That
makes no sense. There is a subclass because you make one up. You could
easily make up a sublass separating bats from mammals; it's a bit more
difficult because we have only recently understood mammal phylogeny well
enough to attempt it, but it would still be possible. I hereby declare
that Laurasiatheria is a subclass, in which the Class Chiroptera is
embedded. Why can't I (or shouldn't I) do that?

> In fact, the intervening ones between superficially shrew-like mammals
> and full-fledged bats is "purely hypothetical" in your terminology
> because we have none of their fossils.

Why should that matter to us? How is either of these any sort of
objective reason to classify taxa differently?

>>>> ... it is the old fashioned naturalists who could best
>>>>> appreciate how the concept of ranks helps people organize their
>>>>> thinking about the biological world. Most science books for laymen
>>>>> nowadays lack the naturalists' touch, and so it is hard for a layman
>>>>> to get a feel for the big picture.
>>>> Personally, I think that your favorite grades only obscure evolution and
>>>> so disguise the big picture.
>>> And I think your cladophile ideas are the ones that obscure evolution
>>> by maintaining what Gould called "the dirty little secret": you have
>>> "real organisms" only at the tips of your trees, never the nodes.
>>> With classical taxa, you can say "the species in X descended from some
>>> member of Y" without having to specify which member.
>> That has nothing to do with real organisms being at tips only. You are
>> very confused.
>
> Not at all; you just don't follow my line of thought, due to some
> confusion of your own:
>
>> There are no real organisms in "some member of Y".
>
> There are real organisms in Y even by your standards.

But that's not "some member of Y". Which member of Y? You don't know.
You can't say. And it's quite likely that you have no actual
representative of the correct member to point to even if you could. It's
no different from my situation. You have merely created the illusion of
something to point to rather than anything real.

> It is just that
> old-fashioned taxonomists acknowledge the fact -- or as close to fact
> as any talk of "X is the sister group of Y" can ever get-- that the
> organisms in X are descended from some organisms which fit all the
> criteria for Y.

> e.g. Aves = X, Maniraptora = Y

Which has exactly the same information as saying that X is belongs to Y,
or that Aves belongs to Maniraptora. Paraphyletic taxa contribute nothing.

> and some day we may be able to narrow Y down more.

In which case the new Y is also a clade that X belongs to. No gain of
information by designating a paraphyletic group.

>> And
>> it's just as true that monkeys, for example, descended from some member
>> of Primates
>
> They are, themselves, members of Primates.

Exactly. So why do you demand a paraphyletic ancestral group?

>> as that they descended from some paraphyletic group of
>> "non-monkeys", so cladistic classification incorporates your needs,
>
> Correction: cladistics *per se* incorporates my needs, cladophile
> classification obscures them.

I'm not sure what you mean by cladistics here if not cladistic
classification. Are you talking about a set of methods for estimating
phylogeny? If so, how do those methods incorporate your needs? If not,
what are you talking about?

>> just making it clearer that it's a strange thing to say.
>
> Not as strange as saying that monkeys are descended from primates, nor
> as strange as saying that the sister group of John Harshman is John
> Harshman's mother, if John Harshman was an only child.

Nobody would say either of these things. I said the former only to show
you how silly your demand for paraphyly was. And nobody is claiming that
sister groups are ancestors, or vice versa.

> [John Harshman's father is a close second, but he didn't contribute
> any mitochondria to John.]
>
>>> We can zero in on a paraphyletic Maniraptora that excludes Aves and
>>> say that we have very good evidence that all living birds descended
>>> ultimately from a Maniraptoran.
>> Which says nothing other than that birds are maniraptorans.
>
> ...thereby being totally unhelpful to anyone who wants to see what a
> good old-fashioned pre-aves maniraptoran looked like.

> I can't even say "pre-aves maniraptoran" without being accused of
> talking about a fictitious group.

And indeed it is fictitious. But what do you mean by "pre-Aves
maniraptoran"? I don't think you quite know, and I don't either.

>> And while
>> this may be subjective, to my ear the latter contains more of the wonder
>> of evolution than the former. As in "humans are odd fish".
>
> Poetry is one thing, baiting a hook for a fish is another.

Pinochle is one thing, pineapple is another. If there's a point in that,
I don't see it. Your claim is that cladistic classification obscures
evolution; my claim is that on the contrary, it highlights evolution. In
other words, we are just arguing over impressions, i.e. poetry.

>>> Then if the evidence suddenly starts to show something else, we can
>>> name some other taxon that we originally thought to be a clade, and
>>> say that the evidence now shows that this taxon was paraphyletic,
>>> because now the evidence points to Aves being descended from this
>>> taxon.
>> Why would we want to do any such thing, and how does cladistic
>> classification differ from this in any significant way? If we find
>> Maniraptora not to contain birds, and some other group to contain birds,
>> we just tranfer birds to the new group.
>
> ...and thereby utterly change what the word "maniraptoran" refers to.
> Old-fashioned taxonomists are blissfully free of having to indulge in
> such wholesale rewriting of books and websites.

No, not utterly. Partially. But any change in the tree changes what any
taxon name means, if by meaning we refer to content. Your way would
change meanings just as much as mine would. It's just that you're
comfortable with your way.

> [...]
>>> One case where something like the
>>> above has already taken place is that, at one time, the best available
>>> evidence suggested that Mesonychia was paraphyletic and that Cetacea
>>> is descended from one of its members. Now the evidence shows
>>> otherwise.
>
> And Mesonychia still means exactly what it meant before the change, to
> a traditional cladist. The same animals were mesonychids that were
> mesonychids before the change.

Whatever is "a traditional cladist"?

>>> The only difference is that, with Linnean taxa having fallen out of
>>> fashion, nobody seems to have bothered to set up a paraphyletic taxon
>>> from which whales are now believed to have descended.
>> That's because we don't need paraphyletic taxa. Why is that a bad thing?
>> Isn't it so much cooler to know that whales are artiodactyls and that
>> their living sister group is hippos?
>
> Not any cooler than knowing that whales are descended from
> artiodactyls, and that among *known* animals, their sister group is
> hippos.

Whatever can be your second point (among *known* animals)? Literally, it
isn't true, as the sister group of whales isn't hippos; there are some
probably fossil members ("anthracotheres") that aren't hippos. That's
why I said "living sister group", not just "sister group".

As for your first point, I think it is in fact cooler.

> But neither fact says anything about how much hippos resemble the LCA
> of *Pakicetus* and the hippo.

Nor does anything useful in classification.

> An old fashioned Linnean would look at
> what we'd be able to reconstruct of that LCA, and tell us whether it
> and the hippo deserve to be in the same family, and whether
> *Pakicetus* could safely be assigned to the genus of the LCA, leaving
> the species of the LCA "purely hypothetical" as a sop to cladophiles
> like you..

And the point of that exercise would be what, exactly? Trying to
incorporate two discordant criteria (cladistic relationships and
similarity) into a single classification results in a bad mapping to both.

>>> We have to
>>> content ourselves with the miserably unhelpful "whales are
>>> artiodactyls, and their nearest living relative is believed to be the
>>> hippopotamus.."
>> Why is that either miserable or unhelpful? It tell us a lot about
>> evolution, much more than erecting pointless paraphyletic groups would.
>
> Wrong, see above.

I didn't see where above you answered this. You could of course add
information by taking your paraphyletic ancestral group -- whatever you
want it to be -- and making it monophyletic by adding whales. That's at
least as good as your desire to put hippos and hypothetical ancestors
into the same family.