Stem vertebrate fossil from 518 million years ago

[RonO]

https://phys.org/news/2022-07-reveals-yunnanozoans-oldest-stem-vertebrates.html

https://www.science.org/doi/10.1126/science.abm2708

https://www.science.org/doi/10.1126/science.adc9198

They claim to have identified features that would make yunnanozoans
examples of early stem vertebrates. Stem taxa are ancestral taxa. They
describe them as fish like, but they look like fugitives from the
amphioxus lineage.

Ron Okimoto

[erik simpson]

They've long been suggested as stem deuterostomes and possibly chordates. "Uncertainty affinities" appears often.
This is a pretty remarkable finding if it holds up.

[John Harshman]

Stem taxa aren't exactly ancestral. They're just taxa outside the
vertebrate crown group but more closely related to vertebrates than to
any other extant taxon.

[erik simpson]

I just got a look at the (paywalled) complete paper, and it looks quite convincing. The cartilaginous branchial arches
with extracellular microfibrils is good reason to think it closely related to vertebrates. Since Metaspriggina and Haikouichthys
are contemporaneous as well as Pikaia, the emergence of vertebrates from chordates must have taken place considerably earlier.

[John Harshman]

Does the paper contain a phylogenetic analysis?

[erik simpson]

The results of a Bayesian analysis is shown, so a qualified "yes". The data is in supplementary material that I don't have yet.
Later (or tomorrow I'll get it. The cladogram indicates yunnanozoans as sister group to all other vertebrates, with urochordates
as sister to total vertebrates.

[peter2...@gmail.com]

Erik Simpson did an OP in sci.bio.paleontology to call attention to this thread.

On Thursday, July 7, 2022 at 11:15:10 PM UTC-4, John Harshman wrote:
> On 7/7/22 7:23 PM, RonO wrote:
> > https://phys.org/news/2022-07-reveals-yunnanozoans-oldest-stem-vertebrates.html
> >
> >
> > https://www.science.org/doi/10.1126/science.abm2708
> >
> > https://www.science.org/doi/10.1126/science.adc9198
> >
> > They claim to have identified features that would make yunnanozoans
> > examples of early stem vertebrates.

That would imply that they have notochords, yet none of the three
linked articles uses the word; I get 0/0 just plugging in "noto".

You'd think at least one would comment on why their minute analysis did not detect one
in any of the 127 specimens studied. They look like larval tunicates, which have well developed
notochords, and not like adult tunicates, which have reabsorbed them.

> > Stem taxa are ancestral taxa. They
> > describe them as fish like, but they look like fugitives from the
> > amphioxus lineage.

By the favored hypothesis, this would make them stem {chordata, urochordata},
not stem chordata, much less stem vertebrata.

> Stem taxa aren't exactly ancestral. They're just taxa outside the
> vertebrate crown group but more closely related to vertebrates than to
> any other extant taxon.

...according to the fiction that no fossil animal is ancestral to any
other fossil animal. The fiction to which I refer is "cladistic fiction,"
which plays the same role in phylogenetic analysis as "legal fiction"
plays in the practice of law.

So strong is the force of this fiction that you, John, ostensibly speaking
for the reigning systematist orthodoxy, claim that it even prohibits
official use of "prime ancestor candidate," meaning a reasonably
complete skeleton that has no apomorphies which make it seem
unlikely that it is a direct ancestor.
["skeleton" because I have never tried to apply the concept outside Stegocephalia]

However, every time I try to tell you of various advantages of this designation,
you never refer to a scientific paper, but go into a broken record routine:

"I don't see any advantages to that."
...
"I don't see any advantages to that."
...
"I don't see any advantages to that."
...

Repeated, as often as necessary.


Peter Nyikos
Professor, Dept. of Mathematics -- standard disclaimer--
University of South Carolina
http://people.math.sc.edu/nyikos

[Lawyer Daggett]

It is an irresponsibly false characterization to assert that John's position is that no fossil
animal __is__ ancestral. The assertion is very distinct. It is that one cannot reasonably assert
that fossil X is ancestral to Y. A mathematician ought to understand the distinction
easily enough, and I assert that you do but make this distorted claim despite knowing
better. That's very poor form for someone who then brags about his academic position.
You and John have been back and forth on this often enough. You have no reasonable excuse
for the dishonest distortion.

[John Harshman]

On 7/12/22 1:10 PM, peter2...@gmail.com wrote:
> Erik Simpson did an OP in sci.bio.paleontology to call attention to this thread.
>
> On Thursday, July 7, 2022 at 11:15:10 PM UTC-4, John Harshman wrote:
>> On 7/7/22 7:23 PM, RonO wrote:
>>> https://phys.org/news/2022-07-reveals-yunnanozoans-oldest-stem-vertebrates.html
>>>
>>>
>>> https://www.science.org/doi/10.1126/science.abm2708
>>>
>>> https://www.science.org/doi/10.1126/science.adc9198
>>>
>>> They claim to have identified features that would make yunnanozoans
>>> examples of early stem vertebrates.
>
> That would imply that they have notochords, yet none of the three
> linked articles uses the word; I get 0/0 just plugging in "noto".
>
> You'd think at least one would comment on why their minute analysis did not detect one
> in any of the 127 specimens studied. They look like larval tunicates, which have well developed
> notochords, and not like adult tunicates, which have reabsorbed them.

Since a notochord is a diagnostic character of Chordata, not Vertebrata,
it would seem irrelevant to the purpose of the paper, which was about
the branchial apparatus, to mention it if they had found one. And of
course there are also taphonomic considerations. Why assume that a
notochord would be preserved had it been present?

>>> Stem taxa are ancestral taxa. They
>>> describe them as fish like, but they look like fugitives from the
>>> amphioxus lineage.
>
> By the favored hypothesis, this would make them stem {chordata, urochordata},
> not stem chordata, much less stem vertebrata.

This statement is unclear. From what I can tell (it's paywalled), the
authors do consider yunnanozoans to be stem vertebrates, i.e. closer to
crown vertebrates than urochordates. Or was it someone else's favored
hypothesis you had in mind?

>> Stem taxa aren't exactly ancestral. They're just taxa outside the
>> vertebrate crown group but more closely related to vertebrates than to
>> any other extant taxon.
>
> ...according to the fiction that no fossil animal is ancestral to any
> other fossil animal.

Not exactly. The claim, which is not fiction, is that no fossil animal
can be *known* (or reasonably believed) to be ancestral to anything,
fossil or otherwise. We just don't have the necessary information.

The fiction to which I refer is "cladistic fiction,"
> which plays the same role in phylogenetic analysis as "legal fiction"
> plays in the practice of law.

> So strong is the force of this fiction that you, John, ostensibly speaking
> for the reigning systematist orthodoxy, claim that it even prohibits
> official use of "prime ancestor candidate," meaning a reasonably
> complete skeleton that has no apomorphies which make it seem
> unlikely that it is a direct ancestor.
> ["skeleton" because I have never tried to apply the concept outside Stegocephalia]
>
> However, every time I try to tell you of various advantages of this designation,
> you never refer to a scientific paper, but go into a broken record routine:
>
> "I don't see any advantages to that."
> ...
> "I don't see any advantages to that."
> ...
> "I don't see any advantages to that."
> ...
>
> Repeated, as often as necessary.

Trying to hijack the thread, eh? Feel free to start a thread on the
advantages of "ancestor candidates". State them, and I'll tell you why
they aren't useful.

[erik simpson]

For the record, Fig 4 of https://www.science.org/doi/10.1126/science.abm2708 does indicate "?nt" (possible notochord)
in Yunnaozoans. In addition, the Supplementary Data indicates presence/absence scoring for Yunnanozoans.
Single-character phylogenetics isn't the norm these days.

[peter2...@gmail.com]

Thanks, but the article is paywalled, and your link does not show Fig.4 without penetration of paywall.

> In addition, the Supplementary Data indicates presence/absence scoring for Yunnanozoans.

As opposed to...?

> Single-character phylogenetics isn't the norm these days.

I don't know why you bothered to write this. I merely talked about the curious case of the dog
that didn't bark in the night, without trying to solve the case.

Allegorically speaking, of course.

Peter Nyikos
Professor, Dept. of Mathematics

University of South Carolina -- standard disclaimer--
http://people.math.sc.edu/nyikos

[Lawyer Daggett]

On Tuesday, July 12, 2022 at 6:00:15 PM UTC-4, peter2...@gmail.com wrote:
> On Tuesday, July 12, 2022 at 5:40:15 PM UTC-4, erik simpson wrote:

> > For the record, Fig 4 of https://www.science.org/doi/10.1126/science.abm2708 does indicate "?nt" (possible notochord)
> > in Yunnaozoans.
> Thanks, but the article is paywalled, and your link does not show Fig.4 without penetration of paywall.

https://sc.edu/about/offices_and_divisions/university_libraries/find_services/off-campus_access/index.php

[erik simpson]

I'm sure your university has a subscription. Otherwise, it's here:
https://www.idrive.com/idrive/sh/sh?k=u1g2q6f1u0

[peter2...@gmail.com]

Yes, but that is not anything I ever claimed. Do I need to explain the meaning of "legal fiction" to you?

> The assertion is very distinct. It is that one cannot reasonably assert
> that fossil X is ancestral to Y.

It's not an assertion, it's a no-brainer that nobody but you has bothered to spell out here.


I admit, I should have distinguished between "fossil" and "species or genus represented by a fossil."

In particular, the term "prime ancestral candidate" refers to the species or genus that the fossil represents.

For example, in Kathleen Hunt's excellent article in the Talk.origins horse FAQ...

http://www.talkorigins.org/faqs/horses/horse_evol.html

...you can see a lineage going directly from Hyracotherium all the way to Equus,
consisting of prime candidates [see above] for direct ancestry of
each genus to the next. In the text, she even talks about a lineage within
Merychippus, with gradual evolution from one species to the next.

Now that I've told you this, I've mercifully deleted the rest of what you wrote,
and await your reaction.

Peter Nyikos
Professor, Dept. of Mathematics -- standard disclaimer--

Univ. of South Carolina at Columbia
http://people.math.sc.edu/nyikos

[Lawyer Daggett]

[Peter Nyikos]

> > > ...according to the fiction that no fossil animal is ancestral to any
> > > other fossil animal. The fiction to which I refer is "cladistic fiction,"
> > > which plays the same role in phylogenetic analysis as "legal fiction"
> > > plays in the practice of law.

> > The assertion is very distinct. It is that one cannot reasonably assert
> > that fossil X is ancestral to Y.

> It's not an assertion, it's a no-brainer that nobody but you has bothered to spell out here.
>
> I admit, I should have distinguished between "fossil" and "species or genus represented by a fossil."
>
> In particular, the term "prime ancestral candidate" refers to the species or genus that the fossil represents.
>
> For example, in Kathleen Hunt's excellent article in the Talk.origins horse FAQ...
>
> http://www.talkorigins.org/faqs/horses/horse_evol.html
>
> ...you can see a lineage going directly from Hyracotherium all the way to Equus,
> consisting of prime candidates [see above] for direct ancestry of
> each genus to the next. In the text, she even talks about a lineage within
> Merychippus, with gradual evolution from one species to the next.
>
> Now that I've told you this, I've mercifully deleted the rest of what you wrote,
> and await your reaction.
> Peter Nyikos
> Professor, Dept. of Mathematics -- standard disclaimer--
> Univ. of South Carolina at Columbia
> http://people.math.sc.edu/nyikos

My reaction is that your deletion is only merciful to you and your dishonesty.
You asserted a position to John that he has never promoted. You distorted
his position, egregiously.

Again, you asserted,
[Peter Nyikos]

> > > ...according to the fiction that no fossil animal is ancestral to any
> > > other fossil animal. The fiction to which I refer is "cladistic fiction,"
> > > which plays the same role in phylogenetic analysis as "legal fiction"
> > > plays in the practice of law.

There has not been an assertion that "no fossil animal is ancestral to any
other fossil animal". Your claim of such an assertion isn't ambiguous
It isn't qualified by "ifs", as you asserted elsewhere. It's simply dishonest.
And you don't even defend it, but rather you deflect and obfuscate.

Again, there exists a real and significant distinction between your nonsensical
claim that John or others assert that no fossil is ancestral to any other (fossil
or species), and the actual assertion that one cannot know if a particular fossil
is ancestral to some latter fossil or species. Again, you are absolutely capable
of appreciating the distinction. It has been expounded upon to you multiple
times. That you deliberately choose to obfuscate over the distinction is a
significant discredit to any reputation for honesty you want to maintain.

[Bob Casanova]

On Tue, 12 Jul 2022 16:08:36 -0700 (PDT), the following
appeared in talk.origins, posted by "peter2...@gmail.com"
<peter2...@gmail.com>:

<snip>
>
>...I've mercifully deleted the rest of what you wrote...
>
A-a-a-a-a-nd we're back to normal.
--

Bob C.

"The most exciting phrase to hear in science,
the one that heralds new discoveries, is not
'Eureka!' but 'That's funny...'"

- Isaac Asimov

[peter2...@gmail.com]

On Tuesday, July 12, 2022 at 7:45:15 PM UTC-4, Bob Casanova wrote:
> On Tue, 12 Jul 2022 16:08:36 -0700 (PDT), the following
> appeared in talk.origins, posted by "peter2...@gmail.com"
> <peter2...@gmail.com>:
>
> <snip>
> >
> >...I've mercifully deleted the rest of what you wrote...

> A-a-a-a-a-nd we're back to normal.
> --
>
> Bob C.

Not quite. Before I went on break in December, and going back at least a year,
you were not replying directly to anything I posted. Always you went through a middleman, someone who had replied to me.
It was what I long ago (1992, years before I heard about talk.origins), called a "de facto killfile".

So this time, you actually had a chance to read the rest of what I wrote before you snipped it. Did you?


Peter Nyikos

[Glenn]

Most of these people are anything but normal.
In the last year or so, though, it appears that Bob is slipping in the noodle department, not quite his usual loony self. I think he's pushing 80, so it may be good to patronize him somewhat.

[Bob Casanova]

On Wed, 13 Jul 2022 16:02:07 -0700 (PDT), the following

appeared in talk.origins, posted by "peter2...@gmail.com"
<peter2...@gmail.com>:

>On Tuesday, July 12, 2022 at 7:45:15 PM UTC-4, Bob Casanova wrote:
>> On Tue, 12 Jul 2022 16:08:36 -0700 (PDT), the following
>> appeared in talk.origins, posted by "peter2...@gmail.com"
>> <peter2...@gmail.com>:
>>
>> <snip>
>> >
>> >...I've mercifully deleted the rest of what you wrote...
>
>> A-a-a-a-a-nd we're back to normal.
>> --
>>
>> Bob C.
>
>Not quite. Before I went on break in December, and going back at least a year,
>you were not replying directly to anything I posted. Always you went through a middleman, someone who had replied to me.
>It was what I long ago (1992, years before I heard about talk.origins), called a "de facto killfile".
>
>So this time, you actually had a chance to read the rest of what I wrote before you snipped it. Did you?
>

I have no idea. Is that a problem?