Archaeopteryx
maff
http://news.independent.co.uk/world/science_technology/article330693.ece
By Steve Connor, Science Editor
Published: 02 December 2005
A perfectly preserved fossil of a feathered creature that lived 150
million years ago has provided further evidence to show that modern
birds are living dinosaurs.
The fossil is a complete skeleton of an Archaeopteryx and shows it had
features common to birds and a group of meat-eating dinosaurs called
therapods.
Archaeopteryx
http://news.google.com/news?num=100&hl=en&lr=&safe=off&q=Archaeopteryx&btnG=Search&sa=N&tab=gn
http://www.google.com/search?num=100&hl=en&lr=&q=Archaeopteryx&tab=nw&ie=UTF-8&sa=N
http://www.google.com/search?q=Archaeopteryx&btnG=Search+Directory&hl=en&cat=gwd%2FTop
http://groups.google.com/groups?q=Archaeopteryx&start=0&scoring=d&num=100&hl=en&lr=&safe=off&
allanm
That's the tenth "discovered". This hoaxer's been a busy little bee,
ain't he?
allanm
http://msnbc.msn.com/id/10283676/
The above came up while looking at the excellent MSN picture, by the
way. Tsk tsk, morally reprehensible, giving ammo to the Creationist
side that evos are all bigots etc etc. But dead funny.
Ash
> Fossil discovery supports theory that birds are living dinosaurs
> http://news.independent.co.uk/world/science_technology/article330693.ece
>
the foot is dinosaur like and not bird like is another nail in the
already firmly shut coffin of "It's just a bird with a couple of
uncommon features"
I suspect that there will be denial at first, then eventually
acceptance. Some sites will argue that this is still a feature found
rarely in modern birds, some will say it has mixed features but that
doesn't make it transitional as there is no evidence it developed from
dinosaurs or developed into modern birds and others will ignore it and
maintain that Archeopteryx is fully bird
Taoshan
> I wonder how the Creationists will deal with this. The discovery that
> the foot is dinosaur like and not bird like is another nail in the
> already firmly shut coffin of "It's just a bird with a couple of
> uncommon features"
Whatever they do, a lot of them are secretly wishing right about now
that their predecessors had flipped the coin the other way.
Creationists had their chance once to declare Archeopteryx is really
just a *dinosaur* with a few uncommon features. They blew it.
"Just a dinosaur" amounts to a lame defense for all the same reasons
'just a bird" does, but it would, anatomically at least, be a slightly
more credible thing to say. Switching now doesn't help them, though. A
flip-flop would only illustrate science's point: Archeopteryx is both.
Picking a single category over the other was always a game of coin
toss.
Taoshan
_
david ford
1991 Eldredge on _Archaeopteryx_
http://groups.google.com/groups?selm=Pine.SGI.3.95.970731000322.23196D-100000%40umbc10.umbc.edu
1979 Futuyma on gradualism, _Archaeopteryx_
http://groups.google.com/groups?selm=Pine.SGI.3.96A.990125233303.654987B-100000%40umbc8.umbc.edu
Stile4aly
How do quotations from 1979 and 1991 have any bearing on a discovery
that has just been unearthed?
Navillus
scientific community rejoices, but by reading the ID webpages and
opinions, all they ever talk about is how we're "embarassed by the
fossil record" etc. etc. Do they even read the articles or do they just
start bloviating right away?
an...@sci.sci
> 'just a bird" does, but it would, anatomically at least, be a slightly
> more credible thing to say. Switching now doesn't help them, though. A
> flip-flop would only illustrate science's point: Archeopteryx is both.
> Picking a single category over the other was always a game of coin
> toss.
I agree. Ignoring those aspects of the genome of Archeopteryx which
were common to both theropod dinosaurs and modern birds, such as
developing from a blastula, having a notochord in the embryo that
develops to a spinal column in the adult, etc., considering only traits
that are different between theropod dinosaur and modern bird, the
Archeopteryx is in some sense (if you weight the differences just
right) halfway between theropod dinosaur and modern bird. In a sense,
along the line of descent from theropod dinosaurs through Archeopteryx
to modern bird, the genome is starting fully theropod dinosaur,
emptying of theropod dinosaur characters and replacing them with modern
bird characters, to end up fully modern bird. The Archeopteryx is at
the point along this evolutionary path where the genome is half emptied
of theropod dinosaur characters and half filled with modern bird
characters. It's stupid to argue whether the genome is half empty (of
theropod dinosaur characters) or half full (of modern bird characters)
when it's in fact both half empty and half full at the same time.
And claiming it's entirely 100% full either way is even stupider.
<nit>Genome contains genes which code for characters, not the
characters themselves of course. But if I had used the long phrase
throughout the above paragraph it would have become unwealdy.</nit>
.
r norman
<snip>
>... Ignoring those aspects of the genome of Archeopteryx which
>were common to both theropod dinosaurs and modern birds, such as
>developing from a blastula, having a notochord in the embryo that
>develops to a spinal column in the adult, etc.,
<snip>
This is pushes things just a little bit. Why not use characters
exclusive to theropods and birds, but not shared by all bilateral
animals (blastula), not shared by all chordates (notochord), not
shared by all vertebrates (spinal cord)?
The fact is that, presented with a series of intermediate
(transitional) forms between A and B, the deniers will always claim
that half of them are "just various forms of A" while the other half
are "just various forms of B".
Taoshan
right. That is one beautiful critter.
Taoshan
_
an...@sci.sci
> >were common to both theropod dinosaurs and modern birds, such as
> >developing from a blastula, having a notochord in the embryo that
> >develops to a spinal column in the adult, etc.,
> shared by all bilateral animals (blastula), not shared by all chordates
> (notochord), not shared by all vertebrates (spinal cord)?
Because that method would not achieve my goal.
Here's my goal: To establish a scale of variation between theropod
dinosaurs and modern birds, whereby per my scale a theropod dinosaur
scores 100% on the theropod dinosaur scale and 0% on the modern bird
scale, while a modern bird scores 100% on the modern bird scale and 0%
on the theropod dinosaur scale, and finally where Archeopteryx scores
50% on each scale. Ideally, for any individual, the score on the
theropod dinosaur scale plus the score on the modern bird scale add to
exactly 100%, so that we can use the metaphor of a glass partly filled
with theropod dinosaur and exactly the remainder filled with modern
bird (genome, or the characters coded by the genome).
Here's the first part of my methodology: For each individual character
where theropod dinosarus differ from modern birds, set up thresholds
such that all theropod dinosaurs score 100% theropod dinosaur and 0%
modern bird, while all modern birds score 100% modern bird and 0%
theropod dinosaur. For example, if all modern birds have thick feathers
over at least 90% of their bodies, for example vultures are bald over
head and neck but heavily feathered nearly entirely elsewhere, while
non-vultures are heavily feathered over all their bodies except eyes
beak and claws of feet, then >= 90% heavy feather cover would be a good
threshold at the modern-bird end of the scale. If theropod dinosaurs
*never* have heavy feather cover, then 0% heavy feather cover would be
a good threshold at the theropod-dinosaur end of the scale. Then any
other species could be rated as to how much heavy feather cover they
have between 0% and 90%, with 0% being rated pure theropod dinosaur,
90% being rated pure modern bird, and 45% being rated 50% of each.
Do the same for each such character where different thresholds exist
for theropod dinosaur and modern bird.
It does no good to consider a character that is the *same* between
modern bird and theropod dinosaur, or where there's a statistical
difference but the two groups overlap, because then it's impossible to
establish a set of two thresholds that assure 0/100 or 100/0 rating for
the respective groups of animals. (This is the specific answer to your
quibble with my methodology.)
Now the second part of the methodology is to multiply each of those
particular scales by a weighting factor, initially assigned according
to our estimates as to how important each particular character is in
distinguishing modern birds from theropod dinosaurs, and also
considering the signal-to-noise ratio (giving very low weights to low
S/N ratio characters, high weights to nice clean signals with high S/N
ratio), i.e. a purely reasonable initial assignment of weights to the
individual characters that distinguish modern birds from theropod
dinosaurs.
Add up all those products of individual scores and weights, and divide
by the total weight, to again achieve a scale which rates all theropod
dinosaurs as 100/0 and all birds as 0/100. Then check where
Archeopteryx fits on that combination scale. If Archeopteryx isn't
exactly at 50/50 on that scale, fudge the weights to move Archeopteryx
closer to the 50/50 point. Keep doing that until Archeopteryx is
exactly at 50(theropodDinosaur)/50(modernBird) on that combination
scale. Now we have a combination scale which is a compromise between
absolute rationality and our forcing of Archeopteryx to sit exactly in
the middle at 50/50, halfway between theropod dinosaurs and modern
birds.
Based on over a hundred years of debate as to whether Archeopteryx is
"really" a bird or "really" a dinosaur, it's my intuition that it
wouldn't take much fudging of the weights to make Archeopteryx come out
exactly at 50/50 on the combination scale. It would already be close to
the 50/50 point at the start, based on totally rational assignment of
weights, so fudging weights to achieve the 50/50 point would be reasonble.
(Some other transitional fossils more recently discovered, which might
be at the 90/10 point or the 20/80 point initially, would require too
much fudging to make them 50/50, so the above methodology would not be
recommended for forcing them to the 50/50 point. But Archeopteryx is
close enough to the midpoint anyway that forcing it to the midpoint for
the sake of setting up a standard scale seems reasonble to me.)
Once we have such a standard scale, with therepod dinosaurs at one end,
modern birds at the other end, and Archeopteryx at the midpoint, we can
do two things: First, rate all other theropod/bird transitional fossils
on that scale, in order to see where the biggest gaps remain. But then
actually draw a multi-dimensional diagram showing the actual distances
between all nearby pairs of fossils, when all the various theropod/bird
distinguising characters are considered separately instead of weighted
into a single scale. See whether the actual characters really do fall
in a neat linear path from theropod dinosaur to modern bird, whereby
they really are all true transitional fossils (or very close to such),
or contraywise whether several of them are distant side branches not
along the actual evolutionary path from theropod dinosaurs to modern
birds.
Let me give an example, via metaphor with locations along a highway in
the USA, which connects Pittsburg to Denver. Any city East of Pittsburg
is analagous to theropod dinosaurs, while any city West of Denver is
analagous to modern birds. Consider this set of cities:
{Baltimore, Boston, Chicago, Dayton, Denver, Houston, Inidianapolis,
Kansas City, New York City, Pittsburg, Reno, Sacramento, St. Louis,
Topica} Try finding them on your map of the USA before reading the rest
below. See if you can guess what the correct sequence of transitional
cities are from Pittsburg to Denver, and what path I was using to
relate them, and how best to weight the various factors to get a scale
from Pittsburg to Denver.
The weighted scale I've chosen is mostly longitude, with very little of
any other factor considered, since longitude best represents the
graduation from Pittsburg to Denver.
Baltimore, New York City and Boston are all East of the Pittsburg
threshold, so count exactly the same as Pittsburg, 100% theropod dinosaur.
Sacramento and Reno are all West of the Denver threshold, so count
exactly the same as Denver, 100% modern bird.
All the rest of the cities (except Pittsburg and Denver themselves) fit
somewhere strictly between Pittsburg and Denver). So the list of
tentative transitional fossils on our one-dimensional scale would start
from Pittsburg and end at Denver. But when we consider latitude as well
as longitude, we discover that Chicago and Houston are far away from
the main line, not true transitionals between Pittsburg and Denver. The
rest of the cities are close enough to the main line that they can be
used as proxies for true transitionals, but Chicago and Houston need to
be crossed off the list of (near-)transitionals. Per the cup metaphor,
they might be only 40% full of theropod dinosaur and 40% full of modern
bird with 20% empty glass, even though the single weighted scale put
them at 50td/50mb.
.
John Harshman
>>"Just a dinosaur" amounts to a lame defense for all the same reasons
>>'just a bird" does, but it would, anatomically at least, be a slightly
>>more credible thing to say. Switching now doesn't help them, though. A
>>flip-flop would only illustrate science's point: Archeopteryx is both.
>>Picking a single category over the other was always a game of coin
>>toss.
>
>
> I agree. Ignoring those aspects of the genome of Archeopteryx which
> were common to both theropod dinosaurs and modern birds,
If you have an Archaeopteryx genome available for examination, I would
really like to talk to you. Same for any Mesozoic theropod.
> such as
> developing from a blastula, having a notochord in the embryo that
> develops to a spinal column in the adult, etc.,
Now I would have considered those phenotypic traits, not part of the genome.
> considering only traits
> that are different between theropod dinosaur and modern bird, the
> Archeopteryx is in some sense (if you weight the differences just
> right) halfway between theropod dinosaur and modern bird. In a sense,
> along the line of descent from theropod dinosaurs through Archeopteryx
> to modern bird, the genome is starting fully theropod dinosaur,
> emptying of theropod dinosaur characters and replacing them with modern
> bird characters, to end up fully modern bird. The Archeopteryx is at
> the point along this evolutionary path where the genome is half emptied
> of theropod dinosaur characters and half filled with modern bird
> characters. It's stupid to argue whether the genome is half empty (of
> theropod dinosaur characters) or half full (of modern bird characters)
> when it's in fact both half empty and half full at the same time.
> And claiming it's entirely 100% full either way is even stupider.
>
> <nit>Genome contains genes which code for characters, not the
> characters themselves of course. But if I had used the long phrase
> throughout the above paragraph it would have become unwealdy.</nit>
As it is, you just used a bizarre and incorrect turn of phrase. What's
wrong with just saying "characters" or "morphology" and forgetting about
the genome, which is forever inaccessible unless you have a Jurassic
Park moment?
Augray
>> >... Ignoring those aspects of the genome of Archeopteryx which
>> >were common to both theropod dinosaurs and modern birds, such as
>> >developing from a blastula, having a notochord in the embryo that
>> >develops to a spinal column in the adult, etc.,
>> Why not use characters exclusive to theropods and birds, but not
>> shared by all bilateral animals (blastula), not shared by all chordates
>> (notochord), not shared by all vertebrates (spinal cord)?
>
>Because that method would not achieve my goal.
>
>Here's my goal: To establish a scale of variation between theropod
>dinosaurs and modern birds, whereby per my scale a theropod dinosaur
>scores 100% on the theropod dinosaur scale and 0% on the modern bird
>scale, while a modern bird scores 100% on the modern bird scale and 0%
>on the theropod dinosaur scale, and finally where Archeopteryx scores
>50% on each scale. Ideally, for any individual, the score on the
>theropod dinosaur scale plus the score on the modern bird scale add to
>exactly 100%, so that we can use the metaphor of a glass partly filled
>with theropod dinosaur and exactly the remainder filled with modern
>bird (genome, or the characters coded by the genome).
Unfortunately, evolution doesn't work like that. Traits don't have to
evolve in conjunction with each other.
>Here's the first part of my methodology: For each individual character
>where theropod dinosarus differ from modern birds, set up thresholds
>such that all theropod dinosaurs score 100% theropod dinosaur and 0%
>modern bird, while all modern birds score 100% modern bird and 0%
>theropod dinosaur.
Under a cladistic interpretation, all modern birds are 100% theropod
dinosaurs.
>For example, if all modern birds have thick feathers
>over at least 90% of their bodies, for example vultures are bald over
>head and neck but heavily feathered nearly entirely elsewhere, while
>non-vultures are heavily feathered over all their bodies except eyes
>beak and claws of feet, then >= 90% heavy feather cover would be a good
>threshold at the modern-bird end of the scale. If theropod dinosaurs
>*never* have heavy feather cover, then 0% heavy feather cover would be
>a good threshold at the theropod-dinosaur end of the scale. Then any
>other species could be rated as to how much heavy feather cover they
>have between 0% and 90%, with 0% being rated pure theropod dinosaur,
>90% being rated pure modern bird, and 45% being rated 50% of each.
The above depends upon your definition of feathers.
>Do the same for each such character where different thresholds exist
>for theropod dinosaur and modern bird.
>
>It does no good to consider a character that is the *same* between
>modern bird and theropod dinosaur, or where there's a statistical
>difference but the two groups overlap, because then it's impossible to
>establish a set of two thresholds that assure 0/100 or 100/0 rating for
>the respective groups of animals. (This is the specific answer to your
>quibble with my methodology.)
>
>Now the second part of the methodology is to multiply each of those
>particular scales by a weighting factor, initially assigned according
>to our estimates as to how important each particular character is in
>distinguishing modern birds from theropod dinosaurs, and also
>considering the signal-to-noise ratio (giving very low weights to low
>S/N ratio characters, high weights to nice clean signals with high S/N
>ratio), i.e. a purely reasonable initial assignment of weights to the
>individual characters that distinguish modern birds from theropod
>dinosaurs.
How does one recognize a clean signal?
>Add up all those products of individual scores and weights, and divide
>by the total weight, to again achieve a scale which rates all theropod
>dinosaurs as 100/0 and all birds as 0/100. Then check where
>Archeopteryx fits on that combination scale. If Archeopteryx isn't
>exactly at 50/50 on that scale, fudge the weights to move Archeopteryx
>closer to the 50/50 point.
I'm hard pressed to think of a character possessed by Archaeopteryx
that *isn't* possessed by some theropod dinosaur.
>Keep doing that until Archeopteryx is
>exactly at 50(theropodDinosaur)/50(modernBird) on that combination
>scale. Now we have a combination scale which is a compromise between
>absolute rationality and our forcing of Archeopteryx to sit exactly in
>the middle at 50/50, halfway between theropod dinosaurs and modern
>birds.
This is kind of like trying to find a creature halfway between
primates and humans.
>Based on over a hundred years of debate as to whether Archeopteryx is
>"really" a bird or "really" a dinosaur,
It's both.
>it's my intuition that it
>wouldn't take much fudging of the weights to make Archeopteryx come out
>exactly at 50/50 on the combination scale.
I'm not optimistic.
>It would already be close to
>the 50/50 point at the start, based on totally rational assignment of
>weights, so fudging weights to achieve the 50/50 point would be reasonble.
>
>(Some other transitional fossils more recently discovered, which might
>be at the 90/10 point or the 20/80 point initially, would require too
>much fudging to make them 50/50, so the above methodology would not be
>recommended for forcing them to the 50/50 point. But Archeopteryx is
>close enough to the midpoint anyway that forcing it to the midpoint for
>the sake of setting up a standard scale seems reasonble to me.)
How did you come to the conclusion that Archaeopteryx is close to the
midpoint?
Ron O
>From the first article:
QUOTE:
The foot is better preserved than any previous Archaeopteryx specimen.
It shows a hyperextendible second toe, rather like the killer claw of
Velociraptor, the vicious therapod dinosaur depicted in the film
Jurassic Park. "These observations provide further evidence for the
theropod ancestry of birds," the scientists say. The hyperextendible
second toe blurs the distinction between this ancient bird and the
therapod dinosaurs, which brings both closer together as a single
group, they say.
END QUOTE:
I just put out this quote to show the lurkers that didn't go to the
article what the fuss is about. You should read the lead article.
People have to remember Jurassic Park and the scene with the killing
claw. This is a claw on the second toe of the bird/dino foot. Modern
birds do not have this killing claw, but apparently birds like Archae
retained the trait that a lot of theropod dinosaurs also had. They
also had feet more like chickens and ducks with three forward facing
toes than perching feet (two forward and two back) like song birds.
I recall that other theropods with feathers also had the killing claw,
but Archeopteryx had a full set of flight feathered wings even if it
didn't fly as well as modern birds.
I wonder what Archeopteryx ate. My guess would be small reptiles and
mammals because I don't think that you need a killing claw for insects.
Since it is found in marsh like areas it could have been a frog
hunter. I know from personal experience that bird legs are extremely
powerful. A kill strike from even a magpie sized bird would probably
do sufficient damage to your run of the mill creeping lifeform.
Ron Okimoto
Augray
Actually, they didn't. All toes pointed forward.
>I recall that other theropods with feathers also had the killing claw,
>but Archeopteryx had a full set of flight feathered wings even if it
>didn't fly as well as modern birds.
>
>I wonder what Archeopteryx ate. My guess would be small reptiles and
>mammals because I don't think that you need a killing claw for insects.
It probably wouldn't turn it's nose up at them.
> Since it is found in marsh like areas it could have been a frog
>hunter.
What makes you think it lived in a marsh-like area? The area at that
time was more like an atoll.
John Harshman
Well, one was sort of off to the side. Anyway, songbirds have one toe
back and three forward.
[snip]
Augray
<jharshman....@pacbell.net> wrote:
>Augray wrote:
>
>> On 4 Dec 2005 07:02:06 -0800, "Ron O" <roki...@cox.net> wrote:
[snip]
>>>People have to remember Jurassic Park and the scene with the killing
>>>claw. This is a claw on the second toe of the bird/dino foot. Modern
>>>birds do not have this killing claw, but apparently birds like Archae
>>>retained the trait that a lot of theropod dinosaurs also had. They
>>>also had feet more like chickens and ducks with three forward facing
>>>toes than perching feet (two forward and two back) like song birds.
>>
>>
>> Actually, they didn't. All toes pointed forward.
>
>Well, one was sort of off to the side. Anyway, songbirds have one toe
>back and three forward.
I was under the impression that Ron's use of the term "theropod
dinosaurs" was to refer to the infamous "non-avian theropods", which
apparently had all toes pointing forward.
>[snip]
John Harshman
You see various different reconstructions. The hallux is short and does
not support any weight in most, and this gives a lot of freedom to the
reconstruction. I've seen reconstructions of Deinonychus in which the
hallux points straight back, but most put it off to the side somewhere.
It's not reversed but neither is it parallel to the other digits. I
haven't made a real study of any of the evidence for hallux life
position in any theropods, and if you have, you know more than I do.
Ron O
I messed up on that one. I've only raise parakeets and chicken, quail
and pheasants. I'd seen finches and canaries in pet stores grabing the
cage wire like parakeets and I thought that they had the same
arrangement with two toes back, but they were probably using their
single back facing toe. The back toe in chickens seems to be pretty
worthless. It barely touches the ground, if at all, in a normal gait.
There is a five toe mutation where the fifth toe is even higher up the
shank above the rear facing toe.
I haven't seen any fossil raptor tracks, but I've seen a claimed T. rex
track and it only has three forward facing toes. There is no evidence
for a fourth toe that I can recall facing forward or back, it is sort
of like a chicken track.
Ron Okimoto
Augray
That may depend on the breed, but it looks like it couldn't help but
touch the ground here: http://www.earthlife.net/birds/anatomy.html#2
Also, it seems to assist in deceleration during walking in other bird
species.
>There is a five toe mutation where the fifth toe is even higher up the
>shank above the rear facing toe.
I'd be interested in hearing more about that. Is there anything in the
literature?
>I haven't seen any fossil raptor tracks, but I've seen a claimed T. rex
>track and it only has three forward facing toes. There is no evidence
>for a fourth toe that I can recall facing forward or back, it is sort
>of like a chicken track.
That toe didn't reach the ground, but it was still present. See
http://www.ucmp.berkeley.edu/trex/trexpo.html and
http://www.ucmp.berkeley.edu/trex/images/rexslide9_bg.jpg
>Ron Okimoto
Ron O
> >
> >I messed up on that one. I've only raise parakeets and chicken, quail
> >and pheasants. I'd seen finches and canaries in pet stores grabing the
> >cage wire like parakeets and I thought that they had the same
> >arrangement with two toes back, but they were probably using their
> >single back facing toe. The back toe in chickens seems to be pretty
> >worthless. It barely touches the ground, if at all, in a normal gait.
>
> That may depend on the breed, but it looks like it couldn't help but
> touch the ground here: http://www.earthlife.net/birds/anatomy.html#2
> Also, it seems to assist in deceleration during walking in other bird
> species.
That picture is taken at the wrong angle to see how high the rear
facing toe is. The picture is taken from an angle above the foot with
the leg angled towards the camera, and the toe is not usually extended
straight out. If the foot was angled away from the camera you would
see that the rear toe is over a centimeter above where the other toes
contact the ground. The rear facing toe points down and inwards in a
normal placement. There is a point of contact at times and the tip of
the toe has a relatively large pad on it, but like you say it may be
used for stopping. Sort of like when the roadrunner stops with its
feet forward, but it doesn't seem to be a very substantial toe. It
could be used for mating. The male has to get a good grip on the
females back because chickens do not have a penis and must come into
direct contact with the female vent to transfer semen. They have to
lean back to do this. The chicken normally walks by stepping directly
infront of the bird with the foot angled in slightly. The outer toe of
the foot is larger and thicker than the inner toe because the bird
tends to roll over it like we use our big inner toe, but in humans the
feet are angled out when we walk.
>
>
>
> >There is a five toe mutation where the fifth toe is even higher up the
> >shank above the rear facing toe.
>
> I'd be interested in hearing more about that. Is there anything in the
> literature?
There is a description of Po (polydactyly) in Poultry Breeding and
Genetics, 1990 edited by R.D. Crawford Elsvier press.
If you go to a poultry show look at the Silkies, Faverolles, and
Houdans have five toes as breed characteristics. There are other
breeds like Dorkings that have five toes, but you don't see them in the
US too often.
There is probably going to be a paper out on it because at a meeting
last year someone used it to try and figure out the way that the toes
wrapped around and up the shank during development. What looks like
the birds reversed knee is preety much our ankle, so the carples
(metacarples?) are sort of wrapped around the foot even though they
look like they are moving up the leg.
an...@sci.sci
> >dinosaurs and modern birds, whereby per my scale a theropod dinosaur
> >scores 100% on the theropod dinosaur scale and 0% on the modern bird
> >scale, while a modern bird scores 100% on the modern bird scale and 0%
> >on the theropod dinosaur scale, and finally where Archeopteryx scores
> >50% on each scale. Ideally, for any individual, the score on the
> >theropod dinosaur scale plus the score on the modern bird scale add to
> >exactly 100%, so that we can use the metaphor of a glass partly filled
> >with theropod dinosaur and exactly the remainder filled with modern
> >bird (genome, or the characters coded by the genome).
> Unfortunately, evolution doesn't work like that. Traits don't have to
> evolve in conjunction with each other.
Who said anything about traits evolving "in conjunction"?? You measure
each trait separately on scale from theropod threshold to modern bird
threshold, then average the traits together to obtain an overall scale.
For example, you might have traits: teeth, feathers, bones (with air
inside). Suppose in the first intermediary there are feathers 90% like
modern birds, but no change to anything else, so that species rates
(0%, 90%, 0%) which averages (with uniform weighting for this example)
to 30% bird. The next intermediary has no further change to feathers,
but teeth are 30% like modern birds, so that species rates (30%, 90%,
0%) which averages to 40% like modern birds. The next intermediary
finishes getting fully modern-bird feathers, and gets 20% aerated
bones, so it rates (30%, 100%, 20%) with average 50% like modern birds.
You get the idea now?
> >Here's the first part of my methodology: For each individual character
> >where theropod dinosars differ from modern birds, set up thresholds
> >such that all theropod dinosaurs score 100% theropod dinosaur and 0%
> >modern bird, while all modern birds score 100% modern bird and 0%
> >theropod dinosaur.
> Under a cladistic interpretation, all modern birds are 100% theropod
> dinosaurs.
Correct. I was using the term "theropod dinosaurs" to mean any animal
which was traditionally classified as such before the bird link was
established. Now I should perhaps say either "non-avian theropod
dinosaurs" or "early-Triassic theropod dinosaurs" or something else
like that to specifically include only the ancestors before they
started evolving along the bird-specific branch. Do you have a
suggestion as to the best group of ancient theropod dinosaurs to use as
the starting point for such lineage-to-birds analysis, and the jargon
to use to specify that particular group? We don't yet know exactly
which Triassic species was the last-common-ancestor of modern birds and
non-avian Jurassic/Cretaceous theropods such as Tyrannosaurus, and even
if we could pinpoint that LCA, we might not have complete fossils, so
we might not be able to rate that single species on all characters. So
that's why I'd prefer a group of closely related non-avian dinosaurs
somewhere near the LCA, none of which had yet diverged toward birds,
presumably somewhere around the basal group for Coelurosauria or
Maniraptora, rather than a single species, as the starting point of the
rating system.
Regarding my example of development of feathers from the near-basal/LCA
group toward modern birds:
> The above depends upon your definition of feathers.
Yes. I'm leaving it open to others (experts) to specify precisely what
character differences are well known between the near-basal/LCA group
and modern birds such that a good scale on that one feature or bunch of
features can be defined. As to feathers, I don't personally know how
sure the experts are that near-basal dinosaurs really didn't have
anything like modern bird-feathers so that such a distinction can be
obtained. I've read about recent disputes as to whether the bulk of the
dinosaurs really were fully-scaled and non-feathered as we always
assumed. If they were fully-feathered just like modern birds, but there
were subtle differences in structure of feathers, there might not be
good data to tell the difference, because feathers only rarely
fossilize well. So even though we know Archy was fully bird-feathered,
we might not be able to say for sure how pre-Archy LCA was or wasn't,
and we might have to totally disregard featheredness as a part of our
weighted scale for LCA-to-bird evolution. We know pretty much for sure
that the LCA of birds and modern crocodiles didn't have feathers,
right? So maybe we'll have to go back to that LCA as a starting point,
which would rate most non-avian dinosaurs as partly-birdlike, a few
percent anyway. I could live with that.
> How does one recognize a clean signal?
At least two factors to consider:
- Deviation within a clade, as opposed to deviation between two clades.
If they mostly overlap, that character can't be used to distinguish
them. But if they are grossly separate, the character *can* be used,
and the nearest points across the chasm can be used as the thresholds.
- Sufficiently well-preserved fossils showing the particular character
clearly over all the species we wish to rate. Adding more characters to
our weighted mean causes some species to be excluded from the rating
system but gives better S/N ratio for the ones that remain, so there's
a tradeoff between inclusiveness and accuracy. We might want to use a
very accurate weighted average of many characters that places only a
few very well known species in precise positions along the axis from
dinosaur-LCA to modern birds, then use a smaller set of characters to
estimate less accurately where a larger set of lesser-preserved species
might lie along that same axis.
> I'm hard pressed to think of a character possessed by Archaeopteryx
> that *isn't* possessed by some theropod dinosaur.
From a cladistic view, Archaeopteryx *is* a sub-type of theropod
dinosaur, so that statement doesn't make sense. I assume you mean some
non-avian theropod i.e. some theropod which was after the LCA along the
non-avian branch from the LCA? By the way, I like the way it's put here:
<http://www.ucmp.berkeley.edu/diapsids/saurischia/theropoda.html>
But with terms defined as I intended (near-LCA-theropods vs. modern
birds), surely there is some character unique to modern birds?
Regarding my scale from ancient theropods to modern birds, I'd be
willing to admit some retrograde evolution. Some modern birds might
have adapted to their environment in a way that causes some character
to revert partway back toward ancestral characters, in essence a weird
form of convergent evolution not synchronous in time. Likewise some
ancient theropod dinosaurs might have evolved, after the LCA with
birds, via true convergent evolution, toward characters partway toward
modern birds. Accordingly I might allow a relaxed scale whereby the LCA
with all theropods (including birds) is at 100/0, every lineage to
modern birds passes through a point of 0/100, but some modern birds are
currently back a little bit, such as 10/90, and some alternate branches
of Jurassic/Cretaceous theropod dinosaurs are forward a bit, such as
90/10. That would allow calculation of distance along the evolutionary
path from ancient theropod dinosaurs toward modern birds, allowing us
to counter Creationists' "missing link" claims with numbers showing a
clear progression along that path, even though it'd give "incorrect"
values for species off that path.
One way to fix the "incorrect value off path" problem, as well as be
more honest about the main path, is to rate each character with more
than one degree of freedom. Compare a line segment from A to B with a
triangle connecting A B and C. Along the line segment (as I explained
above), at the point A we have 100% A and 0% B, while at the point B we
have 0% A and 100% B. Between A and B we have other proportions,
neither all A nor all B, but where the two proportions add to 100%. But
in the triangle, we also have point C, which is 0% A and 0% B, because
it's 100% C. Inside the triangle the three proportions add to 100%,
although no two of them, in particular A and B, add to 100% except
along an edge. So if we have a character that evolved one way to get
from ancient theropods to modern birds, but which evolved a different
way among the not-so-ancient non-avian dinosaurs, or among *very*
modern birds, then these species would show some C, detracting from the
total of A and B. Also if some side branches along the way from ancient
theropods to modern birds evolved in a different direction, they
likewise would show some C, detracting from total A and B. (We might
have several different C for each character if different side branches
evolved these characters differently, so we have a higher dimensional
polyhedron instead of just a triangle.) So then it would be clear by
calculating the various characters which fossils showed a species right
on or very close to the main line of evolution (toward modern birds) or
significantly off to the side from it.
> This is kind of like trying to find a creature halfway between
> primates and humans.
No. Halfway between the LCA of all primates, and humans which are one
modern primate species. Depending on how you weight the various
factors, Chiuman LCA might be 90% human or 50% human or 10% human
compared to that primate LCA. The idea is to get a set of characters
that each distinguish end-to-end 0-100%, then weight them any way you
want to achieve some desired result for one of the intermediate CAs.
With N different characters, you have N-1 degrees of freedom for
weighting, so you might be able to specify in advance N-1 points in
advance if you're lucky. (You can directly solve N equations in N
unknowns, the N-1 specified points, plus the total weights equals 1 as
the other one equation, but if you're unlucky some of the weights will
be negative, which is not allowed.)
> >Based on over a hundred years of debate as to whether Archeopteryx is
> >"really" a bird or "really" a dinosaur,
> It's both.
It depends entirely on your definition. If by "dinosaur" you mean
"non-avian dinosaur", and by "bird" you mean "like a modern bird in all
respects", then it's only partly like each. But if you define
"dinosaur" as LCA of all dinosaurs, then that includes all birds and
Archeopteryx, and if you define "bird" as anything whatsoever in the
maximal clade that includes birds but doesn't include
Jurassic/Cretaceous dinosaurs such as Tyrannosaurus, then "bird"
includes Archeopteryx, so per that definition Archeopteryx is both
within the larger "dinosaur" clade and the "bird" sub-clade of it. It's
all a game of word definitions. The point is that the people doing the
arguing had some early idea of what the words meant, and tried to fit
reality into those definitions, hence endless debate, mostly
meaningless. So now we know better, and we prefer to define our terms
rather than debate semantics. "We" doesn't include Creationists and
IDiots of course. "They" still use stupid definitions as before.
By the way, when "exactly" was LCA of all currently-living modern
birds? I know the LCA of theropod dinosaurs, including birds, was
during the Triassic, right? But from that point, was there early
Triassic divergence among birds, with a LCA way back during Triassic
also? Or was there only a single line of evolution from the
theropod-LCA toward modern birds which has left any current
descendents, such that the LCA of *only* the surviving modern birds
would be much later even though the LCA of modern birds and extinct
bird-like side branches (mostly descendents of something close to
Archeopteryx) would still be during the Triassic?
> >it's my intuition that it
> >wouldn't take much fudging of the weights to make Archeopteryx come out
> >exactly at 50/50 on the combination scale.
> I'm not optimistic.
OK, here's a simple way to do it: The mouth of Archeopteryx had lots of
teeth, just like ancestral non-avian theropod dinosaurs, right? So if
you weight that 100% you get Archeopteryx very high on the
ancient-theropod scale and very low on the modern-bird scale? On the
other hand, whatever weighting of characters were used by just about
all experts to declare that Archeopteryx was a true bird, would cause
it to be rated more than 50% on the modern-bird scale and less than 50%
on the ancient-dinosaur scale. Now by the intermediate value theorem,
there's some value of weighting between 100% teeth and traditional
experts such that the result comes out exactly 50/50.
> How did you come to the conclusion that Archaeopteryx is close to the
> midpoint?
No, you misunderstand what I'm proposing. With N-1 degrees of freedom
for weighting the N characters, I'm simply adding an extra constraint
that Archaeopteryx will come out exactly at the midpoint, then seeing
where the rest of the fossils lie along either half of that line
segment. I picked Archaeopteryx as the canonical midpoint because of
the debate whether it's more one side or the other side of the
midpoint. If people can't agree which side of midpoint it lies, then
there's no reason not to place it exactly at the midpoint and then
calculate all the other transitional fossils per that assumption.
.
John Harshman
>>>Here's my goal: To establish a scale of variation between theropod
>>>dinosaurs and modern birds, whereby per my scale a theropod dinosaur
>>>scores 100% on the theropod dinosaur scale and 0% on the modern bird
>>>scale, while a modern bird scores 100% on the modern bird scale and 0%
>>>on the theropod dinosaur scale, and finally where Archeopteryx scores
>>>50% on each scale. Ideally, for any individual, the score on the
>>>theropod dinosaur scale plus the score on the modern bird scale add to
>>>exactly 100%, so that we can use the metaphor of a glass partly filled
>>>with theropod dinosaur and exactly the remainder filled with modern
>>>bird (genome, or the characters coded by the genome).
Back to the beginning. Why would we want to do this? What do we get out
of it?
[snip]
Augray
Yes, but good luck getting Archaeopteryx to match the 50% average.
>> >Here's the first part of my methodology: For each individual character
>> >where theropod dinosars differ from modern birds, set up thresholds
>> >such that all theropod dinosaurs score 100% theropod dinosaur and 0%
>> >modern bird, while all modern birds score 100% modern bird and 0%
>> >theropod dinosaur.
>>
>> Under a cladistic interpretation, all modern birds are 100% theropod
>> dinosaurs.
>
>Correct. I was using the term "theropod dinosaurs" to mean any animal
>which was traditionally classified as such before the bird link was
>established. Now I should perhaps say either "non-avian theropod
>dinosaurs" or "early-Triassic theropod dinosaurs" or something else
>like that to specifically include only the ancestors before they
>started evolving along the bird-specific branch.
Unfortunately, there isn't really a "bird-specific branch". You can
envision the entire evolutionary tree of theropods as offshoots of a
"bird-specific" trunk.
>Do you have a
>suggestion as to the best group of ancient theropod dinosaurs to use as
>the starting point for such lineage-to-birds analysis, and the jargon
>to use to specify that particular group? We don't yet know exactly
>which Triassic species was the last-common-ancestor of modern birds and
>non-avian Jurassic/Cretaceous theropods such as Tyrannosaurus, and even
>if we could pinpoint that LCA, we might not have complete fossils, so
>we might not be able to rate that single species on all characters. So
>that's why I'd prefer a group of closely related non-avian dinosaurs
>somewhere near the LCA, none of which had yet diverged toward birds,
>presumably somewhere around the basal group for Coelurosauria or
>Maniraptora, rather than a single species, as the starting point of the
>rating system.
If you're interested in including Tyrannosaurus, you should start at
the base of Coelurosauria. Paul Sereno places Ornitholestes in that
area, so that might be a good example to start with.
>Regarding my example of development of feathers from the near-basal/LCA
>group toward modern birds:
>> The above depends upon your definition of feathers.
>
>Yes. I'm leaving it open to others (experts) to specify precisely what
>character differences are well known between the near-basal/LCA group
>and modern birds such that a good scale on that one feature or bunch of
>features can be defined. As to feathers, I don't personally know how
>sure the experts are that near-basal dinosaurs really didn't have
>anything like modern bird-feathers so that such a distinction can be
>obtained. I've read about recent disputes as to whether the bulk of the
>dinosaurs really were fully-scaled and non-feathered as we always
>assumed. If they were fully-feathered just like modern birds, but there
>were subtle differences in structure of feathers, there might not be
>good data to tell the difference, because feathers only rarely
>fossilize well.
I very much doubt that the majority of dinosaurs had feathers
specialized for flight.
>So even though we know Archy was fully bird-feathered,
>we might not be able to say for sure how pre-Archy LCA was or wasn't,
>and we might have to totally disregard featheredness as a part of our
>weighted scale for LCA-to-bird evolution. We know pretty much for sure
>that the LCA of birds and modern crocodiles didn't have feathers,
>right? So maybe we'll have to go back to that LCA as a starting point,
>which would rate most non-avian dinosaurs as partly-birdlike, a few
>percent anyway. I could live with that.
That's probably the way it's going to turn out.
>> How does one recognize a clean signal?
>
>At least two factors to consider:
>- Deviation within a clade, as opposed to deviation between two clades.
>If they mostly overlap, that character can't be used to distinguish
>them. But if they are grossly separate, the character *can* be used,
>and the nearest points across the chasm can be used as the thresholds.
>- Sufficiently well-preserved fossils showing the particular character
>clearly over all the species we wish to rate. Adding more characters to
>our weighted mean causes some species to be excluded from the rating
>system but gives better S/N ratio for the ones that remain, so there's
>a tradeoff between inclusiveness and accuracy. We might want to use a
>very accurate weighted average of many characters that places only a
>few very well known species in precise positions along the axis from
>dinosaur-LCA to modern birds, then use a smaller set of characters to
>estimate less accurately where a larger set of lesser-preserved species
>might lie along that same axis.
So, "noise" would seem to be traits that you're not interested in.
That seems a very arbitrary method of attaining accuracy.
>> >Add up all those products of individual scores and weights, and divide
>> >by the total weight, to again achieve a scale which rates all theropod
>> >dinosaurs as 100/0 and all birds as 0/100. Then check where
>> >Archeopteryx fits on that combination scale. If Archeopteryx isn't
>> >exactly at 50/50 on that scale, fudge the weights to move Archeopteryx
>> >closer to the 50/50 point.
>>
>> I'm hard pressed to think of a character possessed by Archaeopteryx
>> that *isn't* possessed by some theropod dinosaur.
>
>From a cladistic view, Archaeopteryx *is* a sub-type of theropod
>dinosaur, so that statement doesn't make sense. I assume you mean some
>non-avian theropod i.e. some theropod which was after the LCA along the
>non-avian branch from the LCA? By the way, I like the way it's put here:
><http://www.ucmp.berkeley.edu/diapsids/saurischia/theropoda.html>
>
>But with terms defined as I intended (near-LCA-theropods vs. modern
>birds), surely there is some character unique to modern birds?
The problem is that Archaeopteryx isn't a modern bird.
>Regarding my scale from ancient theropods to modern birds, I'd be
>willing to admit some retrograde evolution. Some modern birds might
>have adapted to their environment in a way that causes some character
>to revert partway back toward ancestral characters, in essence a weird
>form of convergent evolution not synchronous in time. Likewise some
>ancient theropod dinosaurs might have evolved, after the LCA with
>birds, via true convergent evolution, toward characters partway toward
>modern birds. Accordingly I might allow a relaxed scale whereby the LCA
>with all theropods (including birds) is at 100/0, every lineage to
>modern birds passes through a point of 0/100, but some modern birds are
>currently back a little bit, such as 10/90, and some alternate branches
>of Jurassic/Cretaceous theropod dinosaurs are forward a bit, such as
>90/10. That would allow calculation of distance along the evolutionary
>path from ancient theropod dinosaurs toward modern birds, allowing us
>to counter Creationists' "missing link" claims with numbers showing a
>clear progression along that path, even though it'd give "incorrect"
>values for species off that path.
Creationist "missing link" arguments seem to be devoted to showing
that Archaeopteryx is the same as extant birds.
>One way to fix the "incorrect value off path" problem, as well as be
>more honest about the main path, is to rate each character with more
>than one degree of freedom. Compare a line segment from A to B with a
>triangle connecting A B and C. Along the line segment (as I explained
>above), at the point A we have 100% A and 0% B, while at the point B we
>have 0% A and 100% B. Between A and B we have other proportions,
>neither all A nor all B, but where the two proportions add to 100%. But
>in the triangle, we also have point C, which is 0% A and 0% B, because
>it's 100% C. Inside the triangle the three proportions add to 100%,
>although no two of them, in particular A and B, add to 100% except
>along an edge. So if we have a character that evolved one way to get
>from ancient theropods to modern birds, but which evolved a different
>way among the not-so-ancient non-avian dinosaurs, or among *very*
>modern birds, then these species would show some C, detracting from the
>total of A and B. Also if some side branches along the way from ancient
>theropods to modern birds evolved in a different direction, they
>likewise would show some C, detracting from total A and B. (We might
>have several different C for each character if different side branches
>evolved these characters differently, so we have a higher dimensional
>polyhedron instead of just a triangle.) So then it would be clear by
>calculating the various characters which fossils showed a species right
>on or very close to the main line of evolution (toward modern birds) or
>significantly off to the side from it.
How does one determine if some percentage of C indicates a side
branch, or a deviation from the path towards living birds?
>> This is kind of like trying to find a creature halfway between
>> primates and humans.
>
>No. Halfway between the LCA of all primates, and humans which are one
>modern primate species. Depending on how you weight the various
>factors, Chiuman LCA might be 90% human or 50% human or 10% human
>compared to that primate LCA. The idea is to get a set of characters
>that each distinguish end-to-end 0-100%, then weight them any way you
>want to achieve some desired result for one of the intermediate CAs.
>With N different characters, you have N-1 degrees of freedom for
>weighting, so you might be able to specify in advance N-1 points in
>advance if you're lucky. (You can directly solve N equations in N
>unknowns, the N-1 specified points, plus the total weights equals 1 as
>the other one equation, but if you're unlucky some of the weights will
>be negative, which is not allowed.)
Once again, this sounds arbitrary.
>> >Based on over a hundred years of debate as to whether Archeopteryx is
>> >"really" a bird or "really" a dinosaur,
>> It's both.
>
>It depends entirely on your definition. If by "dinosaur" you mean
>"non-avian dinosaur", and by "bird" you mean "like a modern bird in all
>respects", then it's only partly like each. But if you define
>"dinosaur" as LCA of all dinosaurs, then that includes all birds and
>Archeopteryx, and if you define "bird" as anything whatsoever in the
>maximal clade that includes birds but doesn't include
>Jurassic/Cretaceous dinosaurs such as Tyrannosaurus, then "bird"
>includes Archeopteryx, so per that definition Archeopteryx is both
>within the larger "dinosaur" clade and the "bird" sub-clade of it. It's
>all a game of word definitions. The point is that the people doing the
>arguing had some early idea of what the words meant, and tried to fit
>reality into those definitions, hence endless debate, mostly
>meaningless. So now we know better, and we prefer to define our terms
>rather than debate semantics. "We" doesn't include Creationists and
>IDiots of course. "They" still use stupid definitions as before.
I would suggest that you drop the term "bird", at least in this
thread, precisely because of the uncertainty of the term.
Archaeopteryx was classified as a "bird" because of the presence of
feathers, but in other aspects it's unlike the feathered animals alive
today. I'd suggest that you use the name "Neornithes" for the clade
consisting of the last common ancestor of all feathered animals alive
today, and all its descendents, and the name "Avialae" for the clade
consisting of the last common ancestor of Archaeopteryx and
Neornithes, and all its descendents. These names are unambiguous and
in relatively common usage.
>By the way, when "exactly" was LCA of all currently-living modern
>birds? I know the LCA of theropod dinosaurs, including birds, was
>during the Triassic, right?
To be more specific, probably in the Late Triassic.
> But from that point, was there early
>Triassic divergence among birds, with a LCA way back during Triassic
>also?
That depends on your definition of "birds". The Avialae probably
appeared in the Middle to Late Jurassic.
>Or was there only a single line of evolution from the
>theropod-LCA toward modern birds which has left any current
>descendents, such that the LCA of *only* the surviving modern birds
>would be much later even though the LCA of modern birds and extinct
>bird-like side branches (mostly descendents of something close to
>Archeopteryx) would still be during the Triassic?
If I understand your question correctly, the answer is "no". The
closest living relatives of Neornithes are crocodiles and alligators.
>> >it's my intuition that it
>> >wouldn't take much fudging of the weights to make Archeopteryx come out
>> >exactly at 50/50 on the combination scale.
>> I'm not optimistic.
>
>OK, here's a simple way to do it: The mouth of Archeopteryx had lots of
>teeth, just like ancestral non-avian theropod dinosaurs, right? So if
>you weight that 100% you get Archeopteryx very high on the
>ancient-theropod scale and very low on the modern-bird scale? On the
>other hand, whatever weighting of characters were used by just about
>all experts to declare that Archeopteryx was a true bird, would cause
>it to be rated more than 50% on the modern-bird scale and less than 50%
>on the ancient-dinosaur scale.
The characters used to label Archaeopteryx a "true bird" were the
presence of feathers, a wishbone, a retroverted pubis, and a reversed
hallux (big toe). That's it. As far as I know, the feathers are
indistinguishable from those of Neornithes, and the wishbone is
probably diagnostic of Coelurosauria. Some theropod dinosaurs had a
retroverted pubis, and the reversed hallux has been revealed to be an
artifact of preservation.
>Now by the intermediate value theorem,
>there's some value of weighting between 100% teeth and traditional
>experts such that the result comes out exactly 50/50.
>
>> How did you come to the conclusion that Archaeopteryx is close to the
>> midpoint?
>
>No, you misunderstand what I'm proposing. With N-1 degrees of freedom
>for weighting the N characters, I'm simply adding an extra constraint
>that Archaeopteryx will come out exactly at the midpoint, then seeing
>where the rest of the fossils lie along either half of that line
>segment. I picked Archaeopteryx as the canonical midpoint because of
>the debate whether it's more one side or the other side of the
>midpoint. If people can't agree which side of midpoint it lies, then
>there's no reason not to place it exactly at the midpoint and then
>calculate all the other transitional fossils per that assumption.
So, you're going to force the data to match your preconceptions? Why
could a creationist use the same technique to place Archaeopteryx
within living birds?
an...@sci.sci
I posted an example elsewhere in the thread. If we give full weight to
teeth, Archaeopteryx would rate very close to full archaic-theropod. If
we give full weight to feathers, Archaeopteryx might score better than
50% like a modern bird. By the intermediate value theorem, there's some
intermediate weighting that yields exactly 50% each bird/archtheropod.
So I think it's easily possible.
> Unfortunately, there isn't really a "bird-specific branch".
Consider the following definition of two clades:
The aside-bird clade: Largest clade containing Tyranosaurus but not any
modern birds.
The bird-specific clade: Largest clade containing all modern birds but
not containing Tyranosaurus.
The only way that definition could fail is if the last common ancestor
of all modern birds also included Tyranosaurus, in which case it'd be
impossible that any clade includes all of that while excluding
Tyranosaurus.
Those two clades as defined have a most recent common ancestor, which
is a single species, which I'll call 'A' below. (It's probably
Coelurosauria.)
Let 'B' denote the last common ancestor of modern birds.
The clade started from B is a sub-clade of the bird-specific clade
defined above.
The oldest member of the bird-specific clade was an immediate
descendent of A. I'll label it AB1, i.e. first species on the line of
descent from A to B. The entire line of descent, except not including A
itself, i.e. the line of descent from AB1 to B, is what I mean by the
"bird-specific branch" assuming Theropod as the outgroup. Note that
Maniraptora
is within the bird-specific branch and Dromaeosauridae is
a side branch from within the bird-specific branch.
Alternately we could use Dromaeosauridae as the out group, which would
make Maniraptora = A, and make the "bird-specific branch" slightly
shorter than before.
> You can envision the entire evolutionary tree of theropods as offshoots of a
> "bird-specific" trunk.
Only if the outgroup is something further from birds than Tyranosaurus.
For example with Sauropodomorpha as the outgroup, all of Theropoda
would be within the "bird-specific clade", and the path Theropoda -
Tetanurae - Avetheropoda - Coelurosauria - Maniraptora would all be
part of the "bird-specific branch". But since Tyranosaurus is a
well-known dinosaur, not bird, it's unreasonable to include it within
the bird clade, so that's why I would make it an outgroup.
> If you're interested in including Tyrannosaurus, you should start at
> the base of Coelurosauria.
I want to *exclude* Tyrannosaurus from the "bird-specific clade", have
it as an outgroup. So I would probably want Coelurosauria to be the
common ancestor of the bird and not-bird clades, rather than the common
ancestor of the bird clade by itself.
> I very much doubt that the majority of dinosaurs had feathers
> specialized for flight.
I agree. Presumably there were be different characters for feathers per
se and feathers specialized for flight. Archaeopteryx apparently has
abundant feathers but they aren't adapted for flight, so just from
those two characters alone you can probably assign weights to make
Archaeopteryx sit at the 50-50 point along the path.
> So, "noise" would seem to be traits that you're not interested in.
> That seems a very arbitrary method of attaining accuracy.
That's actually the correct definition per Shannon's information
theory. You are trying to communicate a message, and anything other
than the message itself is noise, regardless of whether it is true
quantum or thermodyamic noise, or chaotic amplification&mixing thereof,
or cross-channel interference from other communication channels.
> The problem is that Archaeopteryx isn't a modern bird.
That sentence is meaningless in the modern context of evolution. It
makes sense only in the old Platonic/Lamarckian philosophy of fixed
ideal/metaphysical types which are only approximated by physical
objects. (Or it's trivially true simply because Archaeopteryx didn't
survive into modern times.)
> Creationist "missing link" arguments seem to be devoted to showing
> that Archaeopteryx is the same as extant birds.
There's not a single species of modern bird wherein Archaeopteryx would
be identical to a member of that species. I don't believe any
Creationist has ever made such a claim. They claim something rather
different, that there is a fixed Platonic/Lamarckian type "bird" and
that all modern birds and also Archaeopteryx are samples of that type.
> How does one determine if some percentage of C indicates a side
> branch, or a deviation from the path towards living birds?
What do mean by "deviation" there? Do you mean that one of the
linear-evolution (2-cornered/ended) characters temporarily evolves
backward towards the archaic-dinosaurs and then later reverses
direction again to head toward modern bird characters again? Or do you
mean that there's a three-cornered characteristic, which happen to
start at A, move towards C for a while, and finally start moving toward
B, instead of following a straight path from A to B?
Regarding assigning arbitrary non-negative weights to the various
individual characters so that various constraints can be met, such as
the constraint that Archaeopteryx sits at 50-50 on the final scale:
> Once again, this sounds arbitrary.
Yes, it is. Task 1 is to verify that at least one set of non-negative
weightings will satisfy that desired ad hoc constraint. If that
succeeds, the next task is to consider the entire space of such
weightings (the interior+surface of a hyper-polyhedron) to decide which
seems the most reasonable weighting from a professional point of view.
> I would suggest that you drop the term "bird", at least in this
> thread, precisely because of the uncertainty of the term.
I agree the term "bird" by itself is ambiguous. But the term "modern
bird" has a precise meaning. There are certain critters currently alive
today which have feathers and wishbones and other features which make
them very competant at flying. (Bats, squirrels, conifer seeds, etc.
may float or fly but don't have feathers etc. so they are clearly not
within that set.) Then there are a few over-weight critters which can't
fly even though they descend from critters that were good flyers, such
as ostriches and penguins. This set defines "modern birds". Then we
can define "Aves" as the smallest clade containing all these "modern birds".
(I believe if we start from only the actual living feathered flyers,
don't force in the ostrich and penguin, and then generate the smallest
clade containing them all, it will also include ostriches and penguins.
Is that correct?)
If you wish, anywhere I mention my definition for "bird-specific clade"
etc., you may replace that by "modern-bird-specific clade".
> Archaeopteryx was classified as a "bird" because of the presence of
> feathers, but in other aspects it's unlike the feathered animals alive
> today.
Whoever classified it as a "bird" instead of as a transitional toward
modern birds was not correct, IMO.
> I'd suggest that you use the name "Neornithes" for the clade
> consisting of the last common ancestor of all feathered animals alive
> today, and all its descendents,
What's wrong with just calling it "Aves"??
> and the name "Avialae" for the clade consisting of the last common
> ancestor of Archaeopteryx and Neornithes, and all its descendents.
OK, here's part of the cladogram I found online a couple days ago
`--Coelurosauria
|--Tyrannosauroidea
`--Maniraptora
|--Dromaeosauridae
`--AVES
Before we go on, do you agree with that cladogram?
If so, it's not clear to me whether AVES refers to *all* members of
Maniraptora except not any of Dromaeosauridae and not the common
parents of Dromaeosauridae and AVES, i.e. each of Dromaeosauridae and
AVES are basically half-clades within Maniraptora, or whether AVES is
way down a branch there and not the whole half-clade within
Maniraptora? Or is it ambiguous, where Aves can mean either *only* the
modern birds or the entire half-clade, and the above diagram
deliberately leaves the iclusiveness of AVES ambiguous in that way?
If the term 'Aves' is truly ambiguous, then you say Neornithes is the
tiny clade containing *only* from the LCA of modern birds (living
feathered animals) forward? Then what is the standard term for the
half-clade just under Maniraptora containing Neornithes as a small part
of it but also containing all the extinct side-branches along the path
from Maniraptora to Neornithes?
> These names are unambiguous and in relatively common usage.
If they're unambiguous, wheras Aves is ambiguous, then I'd accept
abandoning Aves entirely.
> >By the way, when "exactly" was LCA of all currently-living modern
> >birds? I know the LCA of theropod dinosaurs, including birds, was
> >during the Triassic, right?
> To be more specific, probably in the Late Triassic.
Ah, I would have guessed a little earlier, so it's good that I asked
and that you knew the answer, thanks. I was pretty sure it was sometime
in the Triassic, and glad now to learn I was correct, and glad to be
able to narrow the range tighter now.
> >But from that point, was there early
> >Triassic divergence among birds, with a LCA way back during Triassic
> >also?
> That depends on your definition of "birds". The Avialae probably
> appeared in the Middle to Late Jurassic.
OK, I used that term to do a Google search and found this cladogram:
<http://www.dinosauria.com/dml/clado/avialae.html>
+ Avialae
o Rahonavis ostromi
o Unenlagia comahuensis
o Aves
# Archaeopteryx bavarica
# Archaeopteryx lithographica*
# Metornithes
@ Confuciusornis chuonzhous
@ Confuciusornis sanctus*
@ Confuciusornis suniae
@ Jibeinia luanhera [i. s.]
@ Alvarezsauria
@ ---
- ?Noguerornis
- Ornithothoraces
Before I go on, do you agree that looks correct by best evidence today?
If so, I see "Aves" is set as parent of Archaeopteryx, so my term
"pure bird" is not the same as "Aves" per that location in the tree.
Aves per that tree seems to go all the way from what I'd consider only
half-birdlike to fully-modern-birds. Doing another Google search for
Ornithothoraces I see this deeper narrower cladogram:
<http://www.dinosauria.com/dml/clado/ornithothoraces.html>
+ Ornithothoraces
o Platanavis [i. s.]
o Enantiornithes
# ... (many branches in here)
o Ornithurae
# ... (many branches in here)
I tried to find the definition of Ornithurae, but it seems to be in
dispute, either "all animals closer to modern birds than to
Archaeopteryx", or "common ancestor of Hesperornis and Carinatae".
As for the definition of Enantiornithes, they are considered "opposite
birds", actual birds but with scapula and related bone oriented
opposite from modern birds, and the entire sub-class extinct.
I'm guessing modern birds are within Ornithurae, so I searched again:
<http://www.dinosauria.com/dml/clado/ornithothoraces.html>
o Ornithurae
# ?Liaoningornis
# Songlingornis [i. s.]
# ---
@ ?Chaoyangia
@ ?Gansus
@ ?Kuszholia
@ ?Patagopteryx
@ ---
- Carinatae
= Ambiortimorphae
* Ambiortus
* Apatornis celer
* Ichthyornis dispar
= Neornithes
- Hesperornithiformes
... (10 subs, all from Cretaceous) ...
<http://en.wikipedia.org/wiki/Carinatae>
In phylogenetic taxonomy, the Carinatae are considered the last common
ancestor of Neornithes (living birds) and Ichthyornis (an extinct
seabird of the Cretaceous).
Ah, that's what I was looking for!
Here's a good expnanatory article about cladistics/taxonomy, including
explanation of the three primary ways to define a clade:
<http://palaeo.gly.bris.ac.uk/Essays/phylocode/biolrev.html>
Class Aves can be
given a node-based definition that makes it equivalent to the commonly
accepted taxon, namely 'Class Aves is the clade stemming from the most
recent common ancestor of Archaeopteryx and the sparrow.'
Sigh, so if "Aves" is synomous with "bird", then Archaeopteryx is
forever considered to be a bird, sigh, double sigh. Oh well, at least
the term Neornithes is nicely defined as applying to all "modern
birds", so from now on I'll use that taxon as the endpoint of my scale
from archaic theropod to modern bird. So thanks to you (Augray) for
telling me the keyword I needed for starting the above search, and
thanks to Google for letting me perform all those searches to find the
various cladograms and definitions/descriptions.
So scanning the cladograms above to glean the path, I see I have a gap
between Maniraptora and Avialae, so I do one more search, and find:
<http://town.morrison.co.us/dino-colo/taxonomy/dinoclad.php>
--Maniraptora
|--Oviraptorosauria
`--Eumaniraptora
|--Deinonychosauria
| ...
`--Avialae
So now I can finish building the path as follows:
Coelurosauria -> Maniraptora -> Eumaniraptora -> Avialae -> Aves ->
Metornithes -> --- -> Ornithothoraces -> Ornithurae -> --- -> --- ->
Carinatae -> Neornithes
.
Augray
>> good luck getting Archaeopteryx to match the 50% average.
>
>I posted an example elsewhere in the thread. If we give full weight to
>teeth, Archaeopteryx would rate very close to full archaic-theropod. If
>we give full weight to feathers, Archaeopteryx might score better than
>50% like a modern bird. By the intermediate value theorem, there's some
>intermediate weighting that yields exactly 50% each bird/archtheropod.
>So I think it's easily possible.
>
>> Unfortunately, there isn't really a "bird-specific branch".
>
>Consider the following definition of two clades:
>The aside-bird clade: Largest clade containing Tyranosaurus but not any
>modern birds.
What I think you mean is the clade composed of all animals closer to
Tyrannosaurus than to modern birds.
>The bird-specific clade: Largest clade containing all modern birds but
>not containing Tyranosaurus.
All animals closer to modern birds than to Tyrannosaurus.
>The only way that definition could fail is if the last common ancestor
>of all modern birds also included Tyranosaurus, in which case it'd be
>impossible that any clade includes all of that while excluding
>Tyranosaurus.
>Those two clades as defined have a most recent common ancestor, which
>is a single species, which I'll call 'A' below. (It's probably
>Coelurosauria.)
Actually, it might instead be Maniraptoriformes (see below).
>Let 'B' denote the last common ancestor of modern birds.
>The clade started from B is a sub-clade of the bird-specific clade
>defined above.
>The oldest member of the bird-specific clade was an immediate
>descendent of A. I'll label it AB1, i.e. first species on the line of
>descent from A to B. The entire line of descent, except not including A
>itself, i.e. the line of descent from AB1 to B, is what I mean by the
>"bird-specific branch" assuming Theropod as the outgroup. Note that
>Maniraptora
>is within the bird-specific branch and Dromaeosauridae is
>a side branch from within the bird-specific branch.
>Alternately we could use Dromaeosauridae as the out group, which would
>make Maniraptora = A, and make the "bird-specific branch" slightly
>shorter than before.
Isn't this rather arbitrary? Why use Tyrannosaurus rather than
Allosaurus, or Coelophysis?
>> You can envision the entire evolutionary tree of theropods as offshoots of a
>> "bird-specific" trunk.
>
>Only if the outgroup is something further from birds than Tyranosaurus.
>For example with Sauropodomorpha as the outgroup, all of Theropoda
>would be within the "bird-specific clade", and the path Theropoda -
>Tetanurae - Avetheropoda - Coelurosauria - Maniraptora would all be
>part of the "bird-specific branch". But since Tyranosaurus is a
>well-known dinosaur, not bird, it's unreasonable to include it within
>the bird clade, so that's why I would make it an outgroup.
Why is it unreasonable?
>> If you're interested in including Tyrannosaurus, you should start at
>> the base of Coelurosauria.
>
>I want to *exclude* Tyrannosaurus from the "bird-specific clade", have
>it as an outgroup.
This seems an arbitrary decision.
>So I would probably want Coelurosauria to be the
>common ancestor of the bird and not-bird clades, rather than the common
>ancestor of the bird clade by itself.
>
>> I very much doubt that the majority of dinosaurs had feathers
>> specialized for flight.
>
>I agree. Presumably there were be different characters for feathers per
>se and feathers specialized for flight. Archaeopteryx apparently has
>abundant feathers but they aren't adapted for flight,
Actually, they *are* adapted for flight.
>so just from
>those two characters alone you can probably assign weights to make
>Archaeopteryx sit at the 50-50 point along the path.
Then you'll need some trait other than feathers.
>> So, "noise" would seem to be traits that you're not interested in.
>> That seems a very arbitrary method of attaining accuracy.
>
>That's actually the correct definition per Shannon's information
>theory. You are trying to communicate a message, and anything other
>than the message itself is noise, regardless of whether it is true
>quantum or thermodyamic noise, or chaotic amplification&mixing thereof,
>or cross-channel interference from other communication channels.
But then, a creationist could exclude the characters that you consider
important, on the basis that they're "noise".
>> The problem is that Archaeopteryx isn't a modern bird.
>
>That sentence is meaningless in the modern context of evolution. It
>makes sense only in the old Platonic/Lamarckian philosophy of fixed
>ideal/metaphysical types which are only approximated by physical
>objects. (Or it's trivially true simply because Archaeopteryx didn't
>survive into modern times.)
Actually, "Archaeopteryx isn't a modern bird" makes a *lot* of sense,
in that it lacked many traits possessed by the birds alive today.
Unless you want to debate how one recognizes a bird.
>> Creationist "missing link" arguments seem to be devoted to showing
>> that Archaeopteryx is the same as extant birds.
>
>There's not a single species of modern bird wherein Archaeopteryx would
>be identical to a member of that species.
So? You can say that about virtually any species of bird alive today
when compared to the rest. Imagine saying that "there's not a single
species of primate wherein humans would be identical to a member of
that species". This would demonstrate that humans aren't primates.
>I don't believe any
>Creationist has ever made such a claim. They claim something rather
>different, that there is a fixed Platonic/Lamarckian type "bird" and
>that all modern birds and also Archaeopteryx are samples of that type.
Which is basically what I said.
>> How does one determine if some percentage of C indicates a side
>> branch, or a deviation from the path towards living birds?
>
>What do mean by "deviation" there? Do you mean that one of the
>linear-evolution (2-cornered/ended) characters temporarily evolves
>backward towards the archaic-dinosaurs and then later reverses
>direction again to head toward modern bird characters again?
No.
>Or do you
>mean that there's a three-cornered characteristic, which happen to
>start at A, move towards C for a while, and finally start moving toward
>B, instead of following a straight path from A to B?
Yes.
>Regarding assigning arbitrary non-negative weights to the various
>individual characters so that various constraints can be met, such as
>the constraint that Archaeopteryx sits at 50-50 on the final scale:
>
>> Once again, this sounds arbitrary.
>
>Yes, it is. Task 1 is to verify that at least one set of non-negative
>weightings will satisfy that desired ad hoc constraint. If that
>succeeds, the next task is to consider the entire space of such
>weightings (the interior+surface of a hyper-polyhedron) to decide which
>seems the most reasonable weighting from a professional point of view.
Do you consider ad hoc conclusions to be professional?
>> I would suggest that you drop the term "bird", at least in this
>> thread, precisely because of the uncertainty of the term.
>
>I agree the term "bird" by itself is ambiguous. But the term "modern
>bird" has a precise meaning. There are certain critters currently alive
>today which have feathers and wishbones and other features which make
>them very competant at flying. (Bats, squirrels, conifer seeds, etc.
>may float or fly but don't have feathers etc. so they are clearly not
>within that set.) Then there are a few over-weight critters which can't
>fly even though they descend from critters that were good flyers, such
>as ostriches and penguins. This set defines "modern birds". Then we
>can define "Aves" as the smallest clade containing all these "modern birds".
>(I believe if we start from only the actual living feathered flyers,
>don't force in the ostrich and penguin, and then generate the smallest
>clade containing them all, it will also include ostriches and penguins.
>Is that correct?)
Maybe, maybe not. It depends upon where you place the Tinamiformes.
>If you wish, anywhere I mention my definition for "bird-specific clade"
>etc., you may replace that by "modern-bird-specific clade".
>
>> Archaeopteryx was classified as a "bird" because of the presence of
>> feathers, but in other aspects it's unlike the feathered animals alive
>> today.
>
>Whoever classified it as a "bird" instead of as a transitional toward
>modern birds was not correct, IMO.
They tend to do both at the same time. A creature can be a "bird" and
still be a transitional towards modern birds. It all depends on what
you mean by "bird". And you yourself refer above to "bird and not-bird
clades" separating at the base of Coelurosauria.
>> I'd suggest that you use the name "Neornithes" for the clade
>> consisting of the last common ancestor of all feathered animals alive
>> today, and all its descendents,
>
>What's wrong with just calling it "Aves"??
Because "Aves" has different meanings to different people. For some,
Aves=Avialae, and for others, Aves=Neornithes. Heck, someone has tried
to define Aves=Theropoda, and another has tried to define it as all
animals closer to living birds than to crocodiles, which would make
Triceratops a bird. Suffice it to say, its meaning is ambiguous when
discussing extinct animals.
>> and the name "Avialae" for the clade consisting of the last common
>> ancestor of Archaeopteryx and Neornithes, and all its descendents.
I'd like to correct myself here. "Avialae" is in fact all animals
closer to living birds than to Deinonychosauria (the clade composed of
troodonts and dromaeosaurs).
>OK, here's part of the cladogram I found online a couple days ago
> `--Coelurosauria
> |--Tyrannosauroidea
> `--Maniraptora
> |--Dromaeosauridae
> `--AVES
>Before we go on, do you agree with that cladogram?
There's a lot missing. The base of the Coelurosauria tree is
unresolved, and I've seen three different versions:
--Coelurosauria
`--Maniraptoriformes
|--Ornithomimosauria
`--+--Tyrannosauroidea
`--Maniraptora
--Coelurosauria
|--Tyrannosauroidea
`--Maniraptoriformes
|--Ornithomimosauria
`--Maniraptora
--Coelurosauria
`--Maniraptoriformes
|--+--Ornithomimosauria
| `--Tyrannosauroidea
`--Maniraptora
However, we have a much better understanding of Maniraptora:
--Maniraptora
|--+--Oviraptorosauria
| `--Therizinosauroidea
`--Eumaniraptora
|--Deinonychosauria
| |--Dromaeosauridae
| `--Troodontidae
`--Avialae
`--Pygostylia
`--Ornithothoraces
|--Enantiornithes
`--Euornithes
`--Neornithes
>If so, it's not clear to me whether AVES refers to *all* members of
>Maniraptora except not any of Dromaeosauridae and not the common
>parents of Dromaeosauridae and AVES, i.e. each of Dromaeosauridae and
>AVES are basically half-clades within Maniraptora,
Which would make Aves "paraphyletic", which is not a desirable state.
>or whether AVES is
>way down a branch there and not the whole half-clade within
>Maniraptora?
I think that's what it's trying to portray. Unfortunately, it doesn't
reflect anyone else's opinions on the matter.
>Or is it ambiguous, where Aves can mean either *only* the
>modern birds or the entire half-clade, and the above diagram
>deliberately leaves the iclusiveness of AVES ambiguous in that way?
I'm not sure it's deliberate, but it certainly is ambiguous.
>If the term 'Aves' is truly ambiguous, then you say Neornithes is the
>tiny clade containing *only* from the LCA of modern birds (living
>feathered animals) forward? Then what is the standard term for the
>half-clade just under Maniraptora containing Neornithes as a small part
>of it but also containing all the extinct side-branches along the path
>from Maniraptora to Neornithes?
Given the incompleteness of the cladogram, that question can't be
answered.
>> These names are unambiguous and in relatively common usage.
>
>If they're unambiguous, wheras Aves is ambiguous, then I'd accept
>abandoning Aves entirely.
Aves has been used to label clades synonymous with both Avialae and
Neornithes, depending on the researcher, so yes, I'd suggest
abandoning it, at least for this thread.
Heck no! This page is almost seven years out of date. The Metornithes
clade was originally erected to include the last common ancestor of
living birds and Mononykus (which is a member of Alvarezsauria) and
all its descendents. Mononykus was thought to be a member of Avialae
based on several characters it seemed to have in common with that
group, but more recent analyses place the Alvarezsauria outside of
Avialae.
>If so, I see "Aves" is set as parent of Archaeopteryx, so my term
>"pure bird" is not the same as "Aves" per that location in the tree.
I'm not sure what you mean by '"Aves" is set as parent of
Archaeopteryx'. I would suggest the phrase "Aves is the clade
containing Archaeopteryx".
>Aves per that tree seems to go all the way from what I'd consider only
>half-birdlike to fully-modern-birds. Doing another Google search for
>Ornithothoraces I see this deeper narrower cladogram:
><http://www.dinosauria.com/dml/clado/ornithothoraces.html>
> + Ornithothoraces
> o Platanavis [i. s.]
> o Enantiornithes
> # ... (many branches in here)
> o Ornithurae
> # ... (many branches in here)
>
>I tried to find the definition of Ornithurae, but it seems to be in
>dispute, either "all animals closer to modern birds than to
>Archaeopteryx", or "common ancestor of Hesperornis and Carinatae".
"Ornithurae" (bird tail) describes a clade of feathered animals that
possess a pygostyle, the fusion of tail vertebrate. When the group was
named, the only fossil bird known was Archaeopteryx. Since we now have
more fossils, it's become obvious that neither of the two definitions
above are very descriptive. For more info, see
http://www.taxonsearch.org/dev/taxon_edit.php?Action=View&tax_id=233
>As for the definition of Enantiornithes, they are considered "opposite
>birds", actual birds but with scapula and related bone oriented
>opposite from modern birds, and the entire sub-class extinct.
The formal definition is all animals closer to Sinornis than to
Neornithes.
>I'm guessing modern birds are within Ornithurae, so I searched again:
><http://www.dinosauria.com/dml/clado/ornithothoraces.html>
> o Ornithurae
> # ?Liaoningornis
> # Songlingornis [i. s.]
> # ---
> @ ?Chaoyangia
> @ ?Gansus
> @ ?Kuszholia
> @ ?Patagopteryx
> @ ---
> - Carinatae
> = Ambiortimorphae
> * Ambiortus
> * Apatornis celer
> * Ichthyornis dispar
> = Neornithes
> - Hesperornithiformes
> ... (10 subs, all from Cretaceous) ...
>
><http://en.wikipedia.org/wiki/Carinatae>
> In phylogenetic taxonomy, the Carinatae are considered the last common
> ancestor of Neornithes (living birds) and Ichthyornis (an extinct
> seabird of the Cretaceous).
>Ah, that's what I was looking for!
Unfortunately, this clade name has problems similar to that of
Ornithurae. See
http://www.taxonsearch.org/dev/taxon_edit.php?Action=View&tax_id=71
>Here's a good expnanatory article about cladistics/taxonomy, including
>explanation of the three primary ways to define a clade:
><http://palaeo.gly.bris.ac.uk/Essays/phylocode/biolrev.html>
> Class Aves can be
> given a node-based definition that makes it equivalent to the commonly
> accepted taxon, namely 'Class Aves is the clade stemming from the most
> recent common ancestor of Archaeopteryx and the sparrow.'
>Sigh, so if "Aves" is synomous with "bird",
Is it?
>then Archaeopteryx is
>forever considered to be a bird, sigh, double sigh. Oh well, at least
>the term Neornithes is nicely defined as applying to all "modern
>birds", so from now on I'll use that taxon as the endpoint of my scale
>from archaic theropod to modern bird. So thanks to you (Augray) for
>telling me the keyword I needed for starting the above search, and
>thanks to Google for letting me perform all those searches to find the
>various cladograms and definitions/descriptions.
>
>So scanning the cladograms above to glean the path, I see I have a gap
>between Maniraptora and Avialae, so I do one more search, and find:
><http://town.morrison.co.us/dino-colo/taxonomy/dinoclad.php>
> --Maniraptora
> |--Oviraptorosauria
> `--Eumaniraptora
> |--Deinonychosauria
> | ...
> `--Avialae
>
>So now I can finish building the path as follows:
>Coelurosauria -> Maniraptora -> Eumaniraptora -> Avialae -> Aves ->
> Metornithes -> --- -> Ornithothoraces -> Ornithurae -> --- -> --- ->
> Carinatae -> Neornithes
I'd suggest: Coelurosauria -> Maniraptoriformes -> Maniraptora ->
Eumaniraptora -> Avialae -> Pygostylia -> Ornithothoraces ->
Euornithes -> Neornithes
John Harshman
> On Tue, 13 Dec 2005 00:45:03 -0800, an...@sci.sci wrote:
>
>
>>>good luck getting Archaeopteryx to match the 50% average.
>>
>>I posted an example elsewhere in the thread. If we give full weight to
>>teeth, Archaeopteryx would rate very close to full archaic-theropod. If
>>we give full weight to feathers, Archaeopteryx might score better than
>>50% like a modern bird. By the intermediate value theorem, there's some
>>intermediate weighting that yields exactly 50% each bird/archtheropod.
>>So I think it's easily possible.
>>
>>
>>>Unfortunately, there isn't really a "bird-specific branch".
>>
>>Consider the following definition of two clades:
>>The aside-bird clade: Largest clade containing Tyranosaurus but not any
>>modern birds.
>
>
> What I think you mean is the clade composed of all animals closer to
> Tyrannosaurus than to modern birds.
Is there a difference? I found it entirely clear. It had the form of
some stem-based definitions in common use, except they substitute "most
inclusive" for "largest".
>>The bird-specific clade: Largest clade containing all modern birds but
>>not containing Tyranosaurus.
>
>
> All animals closer to modern birds than to Tyrannosaurus.
Ditto. Let's not be obsessive.
>>The only way that definition could fail is if the last common ancestor
>>of all modern birds also included Tyranosaurus, in which case it'd be
>>impossible that any clade includes all of that while excluding
>>Tyranosaurus.
>>Those two clades as defined have a most recent common ancestor, which
>>is a single species, which I'll call 'A' below. (It's probably
>>Coelurosauria.)
>
> Actually, it might instead be Maniraptoriformes (see below).
There are apparently multiple definitions of Coelurosauria.
>>Let 'B' denote the last common ancestor of modern birds.
>>The clade started from B is a sub-clade of the bird-specific clade
>>defined above.
>>The oldest member of the bird-specific clade was an immediate
>>descendent of A. I'll label it AB1, i.e. first species on the line of
>>descent from A to B. The entire line of descent, except not including A
>>itself, i.e. the line of descent from AB1 to B, is what I mean by the
>>"bird-specific branch" assuming Theropod as the outgroup. Note that
>>Maniraptora
>>is within the bird-specific branch and Dromaeosauridae is
>>a side branch from within the bird-specific branch.
>>Alternately we could use Dromaeosauridae as the out group, which would
>>make Maniraptora = A, and make the "bird-specific branch" slightly
>>shorter than before.
>
> Isn't this rather arbitrary? Why use Tyrannosaurus rather than
> Allosaurus, or Coelophysis?
Or Oviraptor, or Deinonychus, or Archaeopteryx, or Hesperornis.
>>>You can envision the entire evolutionary tree of theropods as offshoots of a
>>>"bird-specific" trunk.
>>
>>Only if the outgroup is something further from birds than Tyranosaurus.
>>For example with Sauropodomorpha as the outgroup, all of Theropoda
>>would be within the "bird-specific clade", and the path Theropoda -
>>Tetanurae - Avetheropoda - Coelurosauria - Maniraptora would all be
>>part of the "bird-specific branch". But since Tyranosaurus is a
>>well-known dinosaur, not bird, it's unreasonable to include it within
>>the bird clade, so that's why I would make it an outgroup.
>
> Why is it unreasonable?
Better question: What is the point of defining a "bird-specific branch"?
>>>If you're interested in including Tyrannosaurus, you should start at
>>>the base of Coelurosauria.
>>
>>I want to *exclude* Tyrannosaurus from the "bird-specific clade", have
>>it as an outgroup.
>
> This seems an arbitrary decision.
Yeah, like that.
[snip]
>>>I would suggest that you drop the term "bird", at least in this
>>>thread, precisely because of the uncertainty of the term.
>>
>>I agree the term "bird" by itself is ambiguous. But the term "modern
>>bird" has a precise meaning. There are certain critters currently alive
>>today which have feathers and wishbones and other features which make
>>them very competant at flying. (Bats, squirrels, conifer seeds, etc.
>>may float or fly but don't have feathers etc. so they are clearly not
>>within that set.) Then there are a few over-weight critters which can't
>>fly even though they descend from critters that were good flyers, such
>>as ostriches and penguins. This set defines "modern birds". Then we
>>can define "Aves" as the smallest clade containing all these "modern birds".
That's one definition. The other common definition includes
Archaeopteryx. The group you define, in technical terms the avian crown
group, is also known as Neornithes.
>>(I believe if we start from only the actual living feathered flyers,
>>don't force in the ostrich and penguin, and then generate the smallest
>>clade containing them all, it will also include ostriches and penguins.
>>Is that correct?)
>
> Maybe, maybe not. It depends upon where you place the Tinamiformes.
There is no real ambiguity on that score. Tinamous are paleognaths, and
paleognaths are monophyletic. You will still find the odd
single-character taxonomy that disputes this, but that's it.
[snip]
>>>I'd suggest that you use the name "Neornithes" for the clade
>>>consisting of the last common ancestor of all feathered animals alive
>>>today, and all its descendents,
>>
>>What's wrong with just calling it "Aves"??
>
> Because "Aves" has different meanings to different people. For some,
> Aves=Avialae, and for others, Aves=Neornithes. Heck, someone has tried
> to define Aves=Theropoda, and another has tried to define it as all
> animals closer to living birds than to crocodiles, which would make
> Triceratops a bird. Suffice it to say, its meaning is ambiguous when
> discussing extinct animals.
It looks as if Aves may be on the way out because of inability to agree
on its definition. I just read a paper that used Neornithes *and*
Avialae, with Aves nowhere in sight.
>>>and the name "Avialae" for the clade consisting of the last common
>>>ancestor of Archaeopteryx and Neornithes, and all its descendents.
>
> I'd like to correct myself here. "Avialae" is in fact all animals
> closer to living birds than to Deinonychosauria (the clade composed of
> troodonts and dromaeosaurs).
Actually, "Avialae" has been defined three different ways (and all by
Gauthier!): stem-based, node-based, and apomorphy-based definitions.
I've never seen the apomorphy-based definition actually used, but I have
seen the stem- and node-based definitions frequently. Does it make a
practical difference at the current moment? That is, does anyone lately
propose any avialans (under the stem-based definition) that are basal to
Archaeopteryx?
[snip]
adam
John Harshman wrote:
> Augray wrote:
>
> >
> > I'd like to correct myself here. "Avialae" is in fact all animals
> > closer to living birds than to Deinonychosauria (the clade composed of
> > troodonts and dromaeosaurs).
>
> Actually, "Avialae" has been defined three different ways (and all by
> Gauthier!): stem-based, node-based, and apomorphy-based definitions.
> I've never seen the apomorphy-based definition actually used, but I have
> seen the stem- and node-based definitions frequently. Does it make a
> practical difference at the current moment? That is, does anyone lately
> propose any avialans (under the stem-based definition) that are basal to
> Archaeopteryx?
>
> [snip]
Its really hard to say since eumaniraptoran, or should that be
paravian, phylogeny seems so flaky at the moment.
Some have found Rahonavis (Holtz, 2000) to occupy that position and/or
Unenlagia (eg. Rauhut 2003). Epidendrosaurus has not been included in
any published analysis but fairly consistently comes out in that
position in Mickey Mortimers private work in progress (reported
frequently on the DML). Luis Chiappe (2003?, in the Mesozoic Birds
volume) thought the alvarezsaurs occupied that position but few others
find them there.
What is fairly clear that very minimal changes, inclusion/exclusion of
one taxon, or even the fusion/fission of a terminal (I'm referring to
Nequenraptor/Unenlagia) can have dramatic changes on the topology in
this part of the tree. I think we may be looking at the products of an
extremely rapid radiation, whose branches may simply be too short for
morphology to accurately resolve.
cheers
Adam
Augray
<jharshman....@pacbell.net> wrote:
>Augray wrote:
>
>> On Tue, 13 Dec 2005 00:45:03 -0800, an...@sci.sci wrote:
>>
>>
>>>>good luck getting Archaeopteryx to match the 50% average.
>>>
>>>I posted an example elsewhere in the thread. If we give full weight to
>>>teeth, Archaeopteryx would rate very close to full archaic-theropod. If
>>>we give full weight to feathers, Archaeopteryx might score better than
>>>50% like a modern bird. By the intermediate value theorem, there's some
>>>intermediate weighting that yields exactly 50% each bird/archtheropod.
>>>So I think it's easily possible.
>>>
>>>
>>>>Unfortunately, there isn't really a "bird-specific branch".
>>>
>>>Consider the following definition of two clades:
>>>The aside-bird clade: Largest clade containing Tyranosaurus but not any
>>>modern birds.
>>
>>
>> What I think you mean is the clade composed of all animals closer to
>> Tyrannosaurus than to modern birds.
>
>Is there a difference? I found it entirely clear. It had the form of
>some stem-based definitions in common use, except they substitute "most
>inclusive" for "largest".
I wasn't sure if he (?) was making reference to some paraphyletic
clade or not.
>>>The bird-specific clade: Largest clade containing all modern birds but
>>>not containing Tyranosaurus.
>>
>>
>> All animals closer to modern birds than to Tyrannosaurus.
>
>Ditto. Let's not be obsessive.
Just looking for clarity.
>>>The only way that definition could fail is if the last common ancestor
>>>of all modern birds also included Tyranosaurus, in which case it'd be
>>>impossible that any clade includes all of that while excluding
>>>Tyranosaurus.
>>>Those two clades as defined have a most recent common ancestor, which
>>>is a single species, which I'll call 'A' below. (It's probably
>>>Coelurosauria.)
>>
>> Actually, it might instead be Maniraptoriformes (see below).
>
>There are apparently multiple definitions of Coelurosauria.
Hmmm... that would seem to be true.
[snip]
>>>>You can envision the entire evolutionary tree of theropods as offshoots of a
>>>>"bird-specific" trunk.
>>>
>>>Only if the outgroup is something further from birds than Tyranosaurus.
>>>For example with Sauropodomorpha as the outgroup, all of Theropoda
>>>would be within the "bird-specific clade", and the path Theropoda -
>>>Tetanurae - Avetheropoda - Coelurosauria - Maniraptora would all be
>>>part of the "bird-specific branch". But since Tyranosaurus is a
>>>well-known dinosaur, not bird, it's unreasonable to include it within
>>>the bird clade, so that's why I would make it an outgroup.
>>
>> Why is it unreasonable?
>
>Better question: What is the point of defining a "bird-specific branch"?
So far, it seems to be a need to arrive at a predetermined outcome.
I'm just trying to peel back the layers.
>>>>If you're interested in including Tyrannosaurus, you should start at
>>>>the base of Coelurosauria.
>>>
>>>I want to *exclude* Tyrannosaurus from the "bird-specific clade", have
>>>it as an outgroup.
>>
>> This seems an arbitrary decision.
>
>Yeah, like that.
[snip]
>>>(I believe if we start from only the actual living feathered flyers,
>>>don't force in the ostrich and penguin, and then generate the smallest
>>>clade containing them all, it will also include ostriches and penguins.
>>>Is that correct?)
>>
>> Maybe, maybe not. It depends upon where you place the Tinamiformes.
>
>There is no real ambiguity on that score. Tinamous are paleognaths, and
>paleognaths are monophyletic. You will still find the odd
>single-character taxonomy that disputes this, but that's it.
I wasn't sure how certain this was.
>[snip]
>
>>>>I'd suggest that you use the name "Neornithes" for the clade
>>>>consisting of the last common ancestor of all feathered animals alive
>>>>today, and all its descendents,
>>>
>>>What's wrong with just calling it "Aves"??
>>
>> Because "Aves" has different meanings to different people. For some,
>> Aves=Avialae, and for others, Aves=Neornithes. Heck, someone has tried
>> to define Aves=Theropoda, and another has tried to define it as all
>> animals closer to living birds than to crocodiles, which would make
>> Triceratops a bird. Suffice it to say, its meaning is ambiguous when
>> discussing extinct animals.
>
>It looks as if Aves may be on the way out because of inability to agree
>on its definition.
That's too bad. I have to admit that I prefer Aves when referring to
the crown clade, as it matches Linnaeus's original intent.
>I just read a paper that used Neornithes *and*
>Avialae, with Aves nowhere in sight.
>
>>>>and the name "Avialae" for the clade consisting of the last common
>>>>ancestor of Archaeopteryx and Neornithes, and all its descendents.
>>
>> I'd like to correct myself here. "Avialae" is in fact all animals
>> closer to living birds than to Deinonychosauria (the clade composed of
>> troodonts and dromaeosaurs).
>
>Actually, "Avialae" has been defined three different ways (and all by
>Gauthier!): stem-based, node-based, and apomorphy-based definitions.
I wasn't aware of the node-based one. Where is it defined?
>I've never seen the apomorphy-based definition actually used, but I have
>seen the stem- and node-based definitions frequently. Does it make a
>practical difference at the current moment? That is, does anyone lately
>propose any avialans (under the stem-based definition) that are basal to
>Archaeopteryx?
Unenlagia has been placed there in more than one cladogram (although
not in the recently published Buitreraptor paper), and Rahonavis has
shown up there as well.
>[snip]
John Harshman
> On Thu, 15 Dec 2005 16:13:37 GMT, John Harshman
> <jharshman....@pacbell.net> wrote:
[snip]
>>
>>Actually, "Avialae" has been defined three different ways (and all by
>>Gauthier!): stem-based, node-based, and apomorphy-based definitions.
>
> I wasn't aware of the node-based one. Where is it defined?
You will find all three in Gauthier & de Queiroz, in the Ostrom volume.
Page 26 says, "Thus, though Gauthier (1986) defined "Avialiae" as the
name of a stem-based clade, he also used it as a name for the
Archaeopteryx node (for example, when listing the synapomorphies
[1986:12])...".
>>I've never seen the apomorphy-based definition actually used, but I have
>>seen the stem- and node-based definitions frequently. Does it make a
>>practical difference at the current moment? That is, does anyone lately
>>propose any avialans (under the stem-based definition) that are basal to
>>Archaeopteryx?
>
> Unenlagia has been placed there in more than one cladogram (although
> not in the recently published Buitreraptor paper), and Rahonavis has
> shown up there as well.
Thanks. I'm recently leaning toward the idea that formal clade
definitions are not all that useful when the tree is in flux.
Augray
wrote:
>John Harshman wrote:
>> Augray wrote:
>>
>[snip]
>
>> >
>> > I'd like to correct myself here. "Avialae" is in fact all animals
>> > closer to living birds than to Deinonychosauria (the clade composed of
>> > troodonts and dromaeosaurs).
>>
>> Actually, "Avialae" has been defined three different ways (and all by
>> Gauthier!): stem-based, node-based, and apomorphy-based definitions.
>> I've never seen the apomorphy-based definition actually used, but I have
>> seen the stem- and node-based definitions frequently. Does it make a
>> practical difference at the current moment? That is, does anyone lately
>> propose any avialans (under the stem-based definition) that are basal to
>> Archaeopteryx?
>>
>> [snip]
>
>Its really hard to say since eumaniraptoran, or should that be
>paravian, phylogeny seems so flaky at the moment.
Just to be pedantic, Paraves is stem-based, whereas Eumaniraptora is
node based, although at this point, I don't think it makes much
difference.
>Some have found Rahonavis (Holtz, 2000) to occupy that position and/or
>Unenlagia (eg. Rauhut 2003). Epidendrosaurus has not been included in
>any published analysis
Actually, it has, in:
Xu X., & Zhang F.-C. 2005. A new maniraptoran dinosaur from China with
long feathers on the metatarsus. Naturwissenschaften 92:173–177.
>but fairly consistently comes out in that
>position in Mickey Mortimers private work in progress (reported
>frequently on the DML).
The relevant portion of the Xu & Zhang cladogram is as follows:
Eumaniraptora
|--Pedopenna
|--+--Archaeopteryx
| |--Epidendrosaurus
| `--+--Confusiusornis
| |--Rahonavis
| `--Jeholornis
`--+--+--Sinovenator
| `--other troodonts
`--+--+--Sinornithosaurus
| |--Microraptor
| `--Graciliraptor
`--other dromaeosaurs
Pedopenna is the animal discussed in the paper.
AC
Ash <asha...@winterfell73.fsnet.co.uk> wrote:
> maff wrote:
>> Fossil discovery supports theory that birds are living dinosaurs
>> http://news.independent.co.uk/world/science_technology/article330693.ece
>>
Top Ten Creationist Reasons to Reject Archaeopteryx:
10. It's not a transitional.
9. It's a dinosaur.
8. It's a bird.
7. It's a hoax.
6. God hates you.
5. Fossil hunters are all atheists out to turn us into lesbians.
4. It must have fallen off the Ark and drowned.
3. There's a guy who saw one in New Mexico.
2. Satan must have put it there to fool you.
And the number one reason for a Creationist to reject Archaeopteryx:
1. Do you have any video tape of it? Huh huh? Didn't think so.
<snip>
--
Aaron Clausen
mightym...@hotmail.com
Augray
<jharshman....@pacbell.net> wrote:
>Augray wrote:
>
>> On Thu, 15 Dec 2005 16:13:37 GMT, John Harshman
>> <jharshman....@pacbell.net> wrote:
>
>[snip]
>
>>>
>>>Actually, "Avialae" has been defined three different ways (and all by
>>>Gauthier!): stem-based, node-based, and apomorphy-based definitions.
>>
>> I wasn't aware of the node-based one. Where is it defined?
>
>You will find all three in Gauthier & de Queiroz, in the Ostrom volume.
>Page 26 says, "Thus, though Gauthier (1986) defined "Avialiae" as the
>name of a stem-based clade, he also used it as a name for the
>Archaeopteryx node (for example, when listing the synapomorphies
>[1986:12])...".
I can't find a node-based definition on page 12, but I've just glanced
over it. I'll give it a more thorough look-over this evening.
>>>I've never seen the apomorphy-based definition actually used, but I have
>>>seen the stem- and node-based definitions frequently. Does it make a
>>>practical difference at the current moment? That is, does anyone lately
>>>propose any avialans (under the stem-based definition) that are basal to
>>>Archaeopteryx?
>>
>> Unenlagia has been placed there in more than one cladogram (although
>> not in the recently published Buitreraptor paper), and Rahonavis has
>> shown up there as well.
>
>Thanks. I'm recently leaning toward the idea that formal clade
>definitions are not all that useful when the tree is in flux.
I think that it really depends upon the definition of the clade
Apomorphy-based clades are the worst, and an example is Gauthier & de
Queiroz's redefinition of Avialae. I mean, how is the first animal
with feathered wings ever going to be determined with certainty?
"Avifilopluma" is even worse, as it *might* end up encompassing all of
Dinosauria.
John Harshman
> On Thu, 15 Dec 2005 18:12:03 GMT, John Harshman
> <jharshman....@pacbell.net> wrote:
>
>
>>Augray wrote:
>>
>>
>>>On Thu, 15 Dec 2005 16:13:37 GMT, John Harshman
>>><jharshman....@pacbell.net> wrote:
>>
>>[snip]
>>
>>
>>>>Actually, "Avialae" has been defined three different ways (and all by
>>>>Gauthier!): stem-based, node-based, and apomorphy-based definitions.
>>>
>>>I wasn't aware of the node-based one. Where is it defined?
>>
>>You will find all three in Gauthier & de Queiroz, in the Ostrom volume.
>>Page 26 says, "Thus, though Gauthier (1986) defined "Avialiae" as the
>>name of a stem-based clade, he also used it as a name for the
>>Archaeopteryx node (for example, when listing the synapomorphies
>>[1986:12])...".
>
> I can't find a node-based definition on page 12, but I've just glanced
> over it. I'll give it a more thorough look-over this evening.
I don't believe it's an explicit definition, just a usage that implies
that definition.
>>>>I've never seen the apomorphy-based definition actually used, but I have
>>>>seen the stem- and node-based definitions frequently. Does it make a
>>>>practical difference at the current moment? That is, does anyone lately
>>>>propose any avialans (under the stem-based definition) that are basal to
>>>>Archaeopteryx?
>>>
>>>Unenlagia has been placed there in more than one cladogram (although
>>>not in the recently published Buitreraptor paper), and Rahonavis has
>>>shown up there as well.
>>
>>Thanks. I'm recently leaning toward the idea that formal clade
>>definitions are not all that useful when the tree is in flux.
>
> I think that it really depends upon the definition of the clade
> Apomorphy-based clades are the worst, and an example is Gauthier & de
> Queiroz's redefinition of Avialae. I mean, how is the first animal
> with feathered wings ever going to be determined with certainty?
> "Avifilopluma" is even worse, as it *might* end up encompassing all of
> Dinosauria.
I don't like other definitions either. You may end up with something
quite different than you intended, with synonyms colliding en masse.
Things work better when you have at least some stable nodes, or a
backbone of stable taxa. That's why crown groups are so attractive.
an...@sci.sci
> Tyrannosaurus than to modern birds.
Closer (less distance between them) per what metric?
> All animals closer to modern birds than to Tyrannosaurus.
Same question.
> Actually, it might instead be Maniraptoriformes (see below).
I'm not interested in the details so much. I just want one clade that
contains two sub-clades, one containing all the modern birds whatsoever
(don't care whether Archy is in it or not), and one sub-clade
containing all the true dinsaurs ("giant lizards") within the whole
clade, which must be at least one such true dinosaur. Is that possible?
Well of course it's possible. Just start with the LCA of all modern
birds, then progress upward in the hierarchy adding one side-branch at
a time until at least one "giant lizard" is within the overall clade.
So that would be the ideal place to take the "road less traveled"
decision toward the modern birds or towards the "giant lizards" most
closely related to the modern birds. What would be the largest dinosaur
within the non-bird half-clade there? If it's only medium big, then
maybe go up another node until some really large dinosaur joins the
clade. Or maybe just go up to the LCA of Tyrannosaurus and modern birds
like I originally planned.
> Why use Tyrannosaurus rather than Allosaurus, or Coelophysis?
I really don't care, so long as that outgroup (compared to birds) is
some really large dinosaur everyone knows about and agrees is a true
dinosaur ("giant lizard"). (Actually it's "terrible lizard", but that
means big and scary-looking lizard, and all big dinosaurs really are
quite scary looking, so the two terms are equivalent.)
> >part of the "bird-specific branch". But since Tyranosaurus is a
> >well-known dinosaur, not bird, it's unreasonable to include it within
> >the bird clade, so that's why I would make it an outgroup.
> Why is it unreasonable?
Because I want to show the difference between the big dinosaurs and the
modern birds, so one clade must not have any big dinosaurs, and the
other clade must not have any modern birds. It's ok if either clade has
some archaic bird-like critters and/or some small dinosaurs. I only
care about the modern birds and the big dinosaurs for my fork in the
road.
> But then, a creationist could exclude the characters that you consider
> important, on the basis that they're "noise".
If a Creationist wants to make up a list of characters, such that they
distinguish modern birds from large dinosaurs, and such that for some
characters Archy is more like modern birds and for other characers
Archy is more unlike modern birds, so that by appropriate weighting I
can accomplish my goal of placing Archy at the half-and-half point in
the weighted scale, that's fine with me.
> Actually, "Archaeopteryx isn't a modern bird" makes a *lot* of sense,
> in that it lacked many traits possessed by the birds alive today.
We're in agreement. If you list some such traits, where Archy is on the
non-bird side of the 50% point, and some other traits, where Archy is
bird side of the 50% point, that will be fine.
> Unless you want to debate how one recognizes a bird.
Let us agree that in the context of ancient fossils, the word "bird" by
itself is ambiguous and useless.
> >There's not a single species of modern bird wherein Archaeopteryx would
> >be identical to a member of that species.
> So? You can say that about virtually any species of bird alive today
> when compared to the rest. Imagine saying that "there's not a single
> species of primate wherein humans would be identical to a member of
> that species". This would demonstrate that humans aren't primates.
No, invalid argument. The smallest clade containing all modern birds
does *not* contain Archy. The smallest clade containing all modern
non-human apes *does* include humans. So we can say that from a
cladistic point of view, Archy shoudn't be considered a modern bird,
but humans *should* be considered a modern ape.
> And you yourself refer above to "bird and not-bird
> clades" separating at the base of Coelurosauria.
Any place I wrote such words, please understand that I meant to say
"modern bird" and "not modern bird" clades, as the two sub-clades of
the LCA of modern birds and some outgroup terrible lizard such as
Tyranosaurus.
> Because "Aves" has different meanings to different people. For some,
> Aves=Avialae, and for others, Aves=Neornithes.
I was assuming the cladograms I found on the Web were commonly agreed
upon, so that when I spliced them together into one large cladogram,
and traced the descendency chain from Coelurosauria to Neornithes,
what I got (below) would be agreed-upon.
Coelurosauria -> Maniraptora -> Eumaniraptora -> Avialae -> Aves ->
Metornithes -> --- -> Ornithothoraces -> Ornithurae -> --- -> --- ->
Carinatae -> Neornithes
If not, then I'd have to replace that particular "Aves" node in the
tree, and in the descendency chain, with a "---" (unnamed) node.
> There's a lot missing. The base of the Coelurosauria tree is
> unresolved, and I've seen three different versions:
> --Coelurosauria
> `--Maniraptoriformes
> |--Ornithomimosauria
> `--+--Tyrannosauroidea
> `--Maniraptora
> --Coelurosauria
> |--Tyrannosauroidea
> `--Maniraptoriformes
> |--Ornithomimosauria
> `--Maniraptora
> --Coelurosauria
> `--Maniraptoriformes
> |--+--Ornithomimosauria
> | `--Tyrannosauroidea
> `--Maniraptora
All of those have the descendency chain:
Coelurosauria -> Maniraptoriformes -> Maniraptora
so I take it that is agreed upon. Inserting that middle node, and taking
away the name "Aves", I now have:
Coelurosauria -> Maniraptoriformes -> Maniraptora -> Eumaniraptora ->
Avialae -> --- -> Metornithes -> --- -> Ornithothoraces ->
Ornithurae -> --- -> --- -> Carinatae -> Neornithes
> However, we have a much better understanding of Maniraptora:
> --Maniraptora
> |--+--Oviraptorosauria
> | `--Therizinosauroidea
> `--Eumaniraptora
> |--Deinonychosauria
> | |--Dromaeosauridae
> | `--Troodontidae
> `--Avialae
> `--Pygostylia
> `--Ornithothoraces
> |--Enantiornithes
> `--Euornithes
> `--Neornithes
That has the path:
Maniraptora -> Eumaniraptora -> Avialae -> Pygostylia ->
Ornithothoraces -> Euornithes -> Neornithes
Aligning that with what I have, anchoring them together where yours and
mine have nodes with the same name, we have two sub-chains of dispute:
Avialae -> Pygostylia -> Ornithothoraces
Avialae -> --- -> Metornithes -> --- -> Ornithothoraces
Ornithothoraces -> Euornithes -> Neornithes
Ornithothoraces -> Ornithurae -> --- -> --- -> Carinatae -> Neornithes
So in the first dispute, are Pygostylia and Metornithes two different
names for the same node, or is one an ancestor of the other, or is one
of them *not* on the direct line as shown?
In the second dispute, does Euornithes sit ancestral to Ornithurae, or
synonym to Ornithurae, or between Ornithurae and Carinatae, or synonym
to Carinatae, or descendent of Carinatae ?
> >If so, it's not clear to me whether AVES refers to *all* members of
> >Maniraptora except not any of Dromaeosauridae and not the common
> >parents of Dromaeosauridae and AVES, i.e. each of Dromaeosauridae and
> >AVES are basically half-clades within Maniraptora,
> Which would make Aves "paraphyletic", which is not a desirable state.
No, by half-clade I mean first-level sub-clade, assuming the node has
three branches (one up and two down).
--Ornithothoraces
|--Enantiornithes
`--Euornithes
each of Enantiornithes and Euornithes is a half-clade with respect to
Ornithothoraces. The entire clade of Ornithothoraces, minus the single
toplevel node, is now two separate clades, each "half" of the total.
> Aves has been used to label clades synonymous with both Avialae and
> Neornithes, depending on the researcher, so yes, I'd suggest
> abandoning it, at least for this thread.
Done, up near the start of this particular followup article.
<snip a full screen I copied from a Web site>
> >Before I go on, do you agree that looks correct by best evidence today?
> Heck no! This page is almost seven years out of date.
No way for me to know without your help, so thanks!!
Anyway, all I need for the purpose of this discussion is the single
descendency chain from Coelurosauria to Neornithes. Should I merge what
I found with what you found, with your help, i.e. fix the 1/1 and the
1/2 discrepancies I asked about 40-45 lines above here? Or should I
just discard the path I constructed myself and use the one you gave me?
> I'd suggest:
> Coelurosauria -> Maniraptoriformes -> Maniraptora -> Eumaniraptora ->
> Avialae -> Pygostylia -> Ornithothoraces -> Euornithes -> Neornithes
OK, that's basically taking what I had, except for those two sub-chains
where there was a conflict, discarding my one named inner node from the
first conflict and my two from the second conflict, and putting in your
one named node into each instead, and omitting all my unnamed nodes
completely. In cases where we have a clear branching, with known
species along side branches from there, but no name for the branch
point itself, I sort of like the idea of keeping the unnamed
branch-nodes in the tree, hence also in the single descendency chain.
With the latest-info you are working from, have all the unnamed nodes
between Coelurosauria and Neornithes either gone away or gotten formal
names, so there are no longer any unnamed branch-nodes along the path
I'm working on?
Is there a search engine where I could feed in a small set of species
names and get back just the minimal descendency tree that links all my
species together, where I get the latest official or well respected
descendency data, nothing seven years out of date? For example, if I
put in Tyranosaurus and Neornithes it'd give me a V-shaped tree that
had Coelurosauria or somesuch at the top and all the names from that
LCA to Neornithes along one branch of the V, and all the names from the
LCA to Tyranosaurus along the other branch of the V?
.
Augray
>> What I think you mean is the clade composed of all animals closer to
>> Tyrannosaurus than to modern birds.
>
>Closer (less distance between them) per what metric?
Common descent.
>> All animals closer to modern birds than to Tyrannosaurus.
>
>Same question.
>
>> Actually, it might instead be Maniraptoriformes (see below).
>
>I'm not interested in the details so much. I just want one clade that
>contains two sub-clades, one containing all the modern birds whatsoever
>(don't care whether Archy is in it or not),
Wasn't including Archaeopteryx the entire point of this exercise?
>and one sub-clade
>containing all the true dinsaurs ("giant lizards") within the whole
>clade, which must be at least one such true dinosaur. Is that possible?
>Well of course it's possible. Just start with the LCA of all modern
>birds, then progress upward in the hierarchy adding one side-branch at
>a time until at least one "giant lizard" is within the overall clade.
>So that would be the ideal place to take the "road less traveled"
>decision toward the modern birds or towards the "giant lizards" most
>closely related to the modern birds. What would be the largest dinosaur
>within the non-bird half-clade there? If it's only medium big, then
>maybe go up another node until some really large dinosaur joins the
>clade. Or maybe just go up to the LCA of Tyrannosaurus and modern birds
>like I originally planned.
Coelurosauria would be the place to start.
[snip]
>> >part of the "bird-specific branch". But since Tyranosaurus is a
>> >well-known dinosaur, not bird, it's unreasonable to include it within
>> >the bird clade, so that's why I would make it an outgroup.
>> Why is it unreasonable?
>
>Because I want to show the difference between the big dinosaurs and the
>modern birds, so one clade must not have any big dinosaurs, and the
>other clade must not have any modern birds. It's ok if either clade has
>some archaic bird-like critters and/or some small dinosaurs. I only
>care about the modern birds and the big dinosaurs for my fork in the
>road.
>
>> But then, a creationist could exclude the characters that you consider
>> important, on the basis that they're "noise".
>
>If a Creationist wants to make up a list of characters, such that they
>distinguish modern birds from large dinosaurs, and such that for some
>characters Archy is more like modern birds and for other characers
>Archy is more unlike modern birds, so that by appropriate weighting I
>can accomplish my goal of placing Archy at the half-and-half point in
>the weighted scale, that's fine with me.
But that's not what I meant. A creationist would exclude as "noise",
any trait that pushed Archaeopteryx out of Neornithes.
>> >> The problem is that Archaeopteryx isn't a modern bird.
>> >
>> >That sentence is meaningless in the modern context of evolution. It
>> >makes sense only in the old Platonic/Lamarckian philosophy of fixed
>> >ideal/metaphysical types which are only approximated by physical
>> >objects. (Or it's trivially true simply because Archaeopteryx didn't
>> >survive into modern times.)
>>
>> Actually, "Archaeopteryx isn't a modern bird" makes a *lot* of sense,
>> in that it lacked many traits possessed by the birds alive today.
>
>We're in agreement.
Then why did you state that my sentence was meaningless?
>If you list some such traits, where Archy is on the
>non-bird side of the 50% point, and some other traits, where Archy is
>bird side of the 50% point, that will be fine.
Since you're going to arbitrarily exclude traits that you don't like,
and arbitrarily give more or less weight to others, I don't see the
point.
>> Unless you want to debate how one recognizes a bird.
>
>Let us agree that in the context of ancient fossils, the word "bird" by
>itself is ambiguous and useless.
Fair enough.
>> >> Creationist "missing link" arguments seem to be devoted to showing
>> >> that Archaeopteryx is the same as extant birds.
>> >
>> >There's not a single species of modern bird wherein Archaeopteryx would
>> >be identical to a member of that species.
>>
>> So? You can say that about virtually any species of bird alive today
>> when compared to the rest. Imagine saying that "there's not a single
>> species of primate wherein humans would be identical to a member of
>> that species". This would demonstrate that humans aren't primates.
>
>No, invalid argument. The smallest clade containing all modern birds
>does *not* contain Archy.
Creationists don't recognize the existence of clades.
>The smallest clade containing all modern
>non-human apes *does* include humans.
And surprisingly, creationists don't recognize that humans are apes
either.
>So we can say that from a
>cladistic point of view, Archy shoudn't be considered a modern bird,
>but humans *should* be considered a modern ape.
We're not talking about cladists, we're talking about creationists.
>> And you yourself refer above to "bird and not-bird
>> clades" separating at the base of Coelurosauria.
>
>Any place I wrote such words, please understand that I meant to say
>"modern bird" and "not modern bird" clades, as the two sub-clades of
>the LCA of modern birds and some outgroup terrible lizard such as
>Tyranosaurus.
Then please write what you mean. And please use terms that have
commonly accepted meanings.
>> Because "Aves" has different meanings to different people. For some,
>> Aves=Avialae, and for others, Aves=Neornithes.
>
>I was assuming the cladograms I found on the Web were commonly agreed
>upon,
Keep in mind that this *is* the web we're talking about. Information
found there is under no obligation to be current, or even accurate.
>so that when I spliced them together into one large cladogram,
>and traced the descendency chain from Coelurosauria to Neornithes,
>what I got (below) would be agreed-upon.
That's quite a wild assumption.
No. Metornithes is probably synonymous with Maniraptoriformes.
>or is one an ancestor of the other,
Yes.
>or is one
>of them *not* on the direct line as shown?
Metornithes has no place in the path above.
>In the second dispute, does Euornithes sit ancestral to Ornithurae, or
>synonym to Ornithurae, or between Ornithurae and Carinatae, or synonym
>to Carinatae, or descendent of Carinatae ?
That depends on your definition of Ornithurae. Take a look at
http://www.taxonsearch.org/dev/taxon_edit.php?Action=View&tax_id=233
and pick one.
>> >If so, it's not clear to me whether AVES refers to *all* members of
>> >Maniraptora except not any of Dromaeosauridae and not the common
>> >parents of Dromaeosauridae and AVES, i.e. each of Dromaeosauridae and
>> >AVES are basically half-clades within Maniraptora,
>>
>> Which would make Aves "paraphyletic", which is not a desirable state.
>
>No, by half-clade I mean first-level sub-clade, assuming the node has
>three branches (one up and two down).
> --Ornithothoraces
> |--Enantiornithes
> `--Euornithes
>each of Enantiornithes and Euornithes is a half-clade with respect to
>Ornithothoraces. The entire clade of Ornithothoraces, minus the single
>toplevel node, is now two separate clades, each "half" of the total.
My comment was based upon a misunderstanding of what you wrote.
>> Aves has been used to label clades synonymous with both Avialae and
>> Neornithes, depending on the researcher, so yes, I'd suggest
>> abandoning it, at least for this thread.
>
>Done, up near the start of this particular followup article.
>
><snip a full screen I copied from a Web site>
>> >Before I go on, do you agree that looks correct by best evidence today?
>> Heck no! This page is almost seven years out of date.
>
>No way for me to know without your help, so thanks!!
You could always check the primary literature.
>Anyway, all I need for the purpose of this discussion is the single
>descendency chain from Coelurosauria to Neornithes. Should I merge what
>I found with what you found, with your help, i.e. fix the 1/1 and the
>1/2 discrepancies I asked about 40-45 lines above here? Or should I
>just discard the path I constructed myself and use the one you gave me?
You can do whatever you like.
>> I'd suggest:
>> Coelurosauria -> Maniraptoriformes -> Maniraptora -> Eumaniraptora ->
>> Avialae -> Pygostylia -> Ornithothoraces -> Euornithes -> Neornithes
>
>OK, that's basically taking what I had, except for those two sub-chains
>where there was a conflict, discarding my one named inner node from the
>first conflict and my two from the second conflict, and putting in your
>one named node into each instead, and omitting all my unnamed nodes
>completely. In cases where we have a clear branching, with known
>species along side branches from there, but no name for the branch
>point itself, I sort of like the idea of keeping the unnamed
>branch-nodes in the tree, hence also in the single descendency chain.
>With the latest-info you are working from, have all the unnamed nodes
>between Coelurosauria and Neornithes either gone away or gotten formal
>names, so there are no longer any unnamed branch-nodes along the path
>I'm working on?
At this point, not all nodes have names.
>Is there a search engine where I could feed in a small set of species
>names and get back just the minimal descendency tree that links all my
>species together, where I get the latest official or well respected
>descendency data, nothing seven years out of date? For example, if I
>put in Tyranosaurus and Neornithes it'd give me a V-shaped tree that
>had Coelurosauria or somesuch at the top and all the names from that
>LCA to Neornithes along one branch of the V, and all the names from the
>LCA to Tyranosaurus along the other branch of the V?
None that I know of.
John Harshman
> On Thu, 15 Dec 2005 23:07:42 -0800, an...@sci.sci wrote:
>
>
>>>What I think you mean is the clade composed of all animals closer to
>>>Tyrannosaurus than to modern birds.
>>
>>Closer (less distance between them) per what metric?
>
>
> Common descent.
That reply was a bit opaque. You mean relative recency of common
ancestry, i.e the clade composed of all animals that share a more recent
common ancestor with Tyrannoaurus than with modern birds. These are all
variants of a stem-based clade definition.
[snip]
Stanley Friesen
>
>Correct. I was using the term "theropod dinosaurs" to mean any animal
>which was traditionally classified as such before the bird link was
>established. Now I should perhaps say either "non-avian theropod
>dinosaurs" or "early-Triassic theropod dinosaurs" or something else
>like that to specifically include only the ancestors before they
>started evolving along the bird-specific branch. Do you have a
>suggestion as to the best group of ancient theropod dinosaurs to use as
>the starting point for such lineage-to-birds analysis, and the jargon
>to use to specify that particular group?
The general consensus is Maniraptora contain the closest relatives of
birds. The two most famous groups of maniraptorans are the troodontids
and the dromaeosaurids - the two groups with large sickle claws on one
toe. Also more or less closely related to birds are _Protarchaeopteryx_
and the alvarezsaurids (which some believe to actually *be* birds).
There is an early undescribed maniraptoran from the Late Jurassic
Morrison Formation, so the origin of birds could easily be some where
around the Mid Jurassic.
> We don't yet know exactly
>which Triassic species was the last-common-ancestor of modern birds and
>non-avian Jurassic/Cretaceous theropods such as Tyrannosaurus, and even
>if we could pinpoint that LCA, we might not have complete fossils, so
>we might not be able to rate that single species on all characters.
That LCA may well have been as recent as the mid Jurassic.
--
The peace of God be with you.
Stanley Friesen
Augray
<jharshman....@pacbell.net> wrote:
>Augray wrote:
>
>> On Thu, 15 Dec 2005 23:07:42 -0800, an...@sci.sci wrote:
>>
>>
>>>>What I think you mean is the clade composed of all animals closer to
>>>>Tyrannosaurus than to modern birds.
>>>
>>>Closer (less distance between them) per what metric?
>>
>>
>> Common descent.
>
>That reply was a bit opaque.
True.
>You mean relative recency of common
>ancestry, i.e the clade composed of all animals that share a more recent
>common ancestor with Tyrannoaurus than with modern birds. These are all
>variants of a stem-based clade definition.
At the moment I wrote the above, clarity abandoned me. Thanks for
clarifying it.
>[snip]
an...@sci.sci
> >> Tyrannosaurus than to modern birds.
> >Is there a difference? I found it entirely clear. It had the form of
> >some stem-based definitions in common use, except they substitute "most
> >inclusive" for "largest".
> I wasn't sure if he (?) was making reference to some paraphyletic
> clade or not.
*oxymoron*alert* Duck and cover!
(By definition, any clade is monophyletic!)
> >Better question: What is the point of defining a "bird-specific branch"?
> So far, it seems to be a need to arrive at a predetermined outcome.
> I'm just trying to peel back the layers.
Yes. The idea is to define a branch that leads from the last common
ancestor of the modern birds and some everyone-agrees non-bird dinosaur
(such as Tyranosaurus, the corresponding common ancestor being either
Coelurosauria or Maniraptoriformes), through the various transition
forms of ancient bird-like dinosaurs (common ancestors of modern birds
and Archaeopteryx) and dinosaur-like birds (cousins to Archaeopteryx)
and birds-but-not-modern-birds (such as Euornithes) all the way to the
LCA of all modern (currently living) birds (Neornithes). Then I have
one additional constraint that the LCA of modern birds and
Archaeopteryx, which is very close to Archaeopteryx in the evolutionary
tree, should score exactly halfway between Coelurosauria and
Neornithes, per some weighting of the various characters that
distinguish those two points in the tree.
Note: I'm going by the following cladogram for the non-bird clade just
under the bird+Tyranosaurus LCA:
<http://www.dinoruss.org/dml/clado/tyrannosauroidea.html>
+ Tyrannosauroidea
o Futabasaurus [i. s.]
o Iliosuchus incognitus
o Itemirus medullaris
o Siamotyrannus isanensis
o Stokesosaurus clevelandi
o Tonouchisaurus mongoliensis
o Tyrannosauridae
# Alectrosaurus olseni
# Shanshanosaurus huoyanshanensis
# Stygivenator molnari
# Tyrannosaurinae
@ Alioramus remotus
@ ---
- Albertosaurus sarcophagus
- Gorgosaurus libratus
- Tyrannosaurini
= Daspletosaurus sp.
= Daspletosaurus torosus*
= Dinotyrannus megagracilis
= Tyrannosaurus bataar
= Tyrannosaurus efremovi
= Tyrannosaurus lancensis
= Tyrannosaurus novojilovi
= Tyrannosaurus rex*
Is that cladogram current/uptodate/correctsofarasweknowtoday?
Revised: January 25, 1999; New: January 1, 1996
Oops, that's pretty old! Now I'm really worried, in a Hitchcock way!
Bad joke: Tyrannosaurini are mis-named, since they weren't placental
mammals, so they didn't have *any* kind of belly button, not an outi
nor an ini.
By the way, when I was a kid I always assumed that Pterosaurs were the
missing link between dinosaurs and birds, but now I realize they are
out a ways along just a side branch from quite a bit before the LCA of
Tyrannosaurus and birds, essentially convergent evolution with bats
rather than birds anyway.
Hey, here's a rather complete cladogram of all types of dinosaurs:
<http://town.morrison.co.us/dino-colo/taxonomy/dinoclad.php>
> >It looks as if Aves may be on the way out because of inability to agree
> >on its definition.
> That's too bad. I have to admit that I prefer Aves when referring to
> the crown clade, as it matches Linnaeus's original intent.
Linnaeus defined the terminology for currently-living organisms only.
Unfortunately when different people tried to exted the term 'Aves' to
fossils, they IMO assigned the term prematurely, before they knew
enough about the fossils to know where the right place should be. Maybe
there is in fact no one "right" place to be consistent with traditional
usage yet neverthelss include some fossils of extinct species.
Accordingly I am not throwing my support to people who want to abandon
it entirely in cladistic analysis, while keeping it (for the moment)
only as a classification of the currently/recently living species of
birds when using the old Linnaean system.
.
an...@sci.sci
> definitions are not all that useful when the tree is in flux.
I'm leaning stronger than you are in that same direction. If and when
there's a clear definition of a genus, which everyone can agree is a
single genus, not convergent evolution of distant relatives (such as
bats and pterosaurs which have a common reptile ancestor, or shark and
killer-whale which have a common chrodate ancestor), then it's good to
have a name assigned to that genus. But then when various genera are
put into a tentative family tree, it's premature to name most or all of
the inner nodes of the tree. Better leave them all unnamed until and
unless cladistic study shows without daubt that certain genera fit
within a particular clade and all other genera are definitely *not*
within that clade. Then maybe it's time to name that clade with a
toplevel name, such as Neornithes. But for all the branch nodes which
are unresolved, I'd rather see unnamed nodes in the cladogram, than
names that are going to be redefined to be different clades every
couple years from then on, such as Maniraptoriformes which may or may
not contain Tyrannosauroidea according to different experts:
<http://groups.google.com/group/talk.origins/msg/24995773eafe0b3a>
= Message-ID: <1svvp15vs2pnk2a60...@4ax.com>
The Maniraptoriformes branching-node would have been better left
un-named so-far, IMO. Even moreso for Aves which now has to be
abandoned as a node-label in cladistics, IMO.
.
an...@sci.sci
> quite different than you intended, with synonyms colliding en masse.
I think the problem may arise perhaps when somebody has some vague idea
which taxa he wants to include in a higher-level taxon, but doesn't
know the evolutionary relationships, and tries to guess which nice
clean definition would achieve the result, but makes a mistake. (As I
pointed out in another article, not yet appearing in sci.bio.evolution
because moderator didn't get my first attempt while he was on vacation
and apparently went to bed too early to see my second submission yet:
The kingdom of Fungi has such a half-baked formal definition, namely
all eukaryotes that *never* have any undulipodia at any point in their
life cycle, which simply means they include all Protictista which have
degenerated that characteristic, and therfore Fungi is ullikely to be
monophyletic.)
> Things work better when you have at least some stable nodes, or a
> backbone of stable taxa. That's why crown groups are so attractive.
I don't know what crown groups are. I tried to look it up on Google,
and got this totally sucky definition:
<http://www.peripatus.gen.nz/Biology/defStemCrown.html>
Clade
A monophyletic taxon; a group of organisms which includes the most
recent common ancestor of all of its members and all of the
descendants of that most recent common ancestor.
That's correct, but:
Crown Group
Those organisms descended from the last common ancestor of all living
members of a clade (Runnegar 1992).
The last common ancestor of a living clade plus all of its descendants
(Budd 2001, p. 333).
But that's exactly the same as the clade itself: One individual
(species, in case of eukaryotes that reproduce via meiosis), and all
descendents of that individual (species).
Do you know of any definition whereby the crown group isn't the whole clade?
.
Eric Rowley
<snip>
> > Things work better when you have at least some stable nodes,
> > or a backbone of stable taxa. That's why crown groups are so
> > attractive.
> I don't know what crown groups are. I tried to look it up on
> Google, and got this totally sucky definition:
> <http://www.peripatus.gen.nz/Biology/defStemCrown.html>
> Clade
> A monophyletic taxon; a group of organisms which includes the
> most recent common ancestor of all of its members and all of the
> descendants of that most recent common ancestor.
> That's correct, but:
> Crown Group
> Those organisms descended from the last common ancestor of all
> living members of a clade (Runnegar 1992).
> The last common ancestor of a living clade plus all of its
> descendants
> (Budd 2001, p. 333).
> But that's exactly the same as the clade itself:
Yes and no, it depends on your defination of "the". ;-)
As I read it, and it took a while to twist my mind around
the defination, a crown group is a clade but not all clades
are crown groups.
So depending on which clade you start with you may have to
prune it a bit to get a crown group.
> One individual
> (species, in case of eukaryotes that reproduce via meiosis), and
> all descendents of that individual (species).
But only if the next branching point leads to living decendents
on both branches, otherwise that individual is not the LCA of its
_living_ descendents.
> Do you know of any definition whereby the crown group isn't the
> whole clade? .
Sure, the clade of "humans" would include Neanderthals but the crown
group is just us (and our imediate ancestor?) since they are (I
presume) not a descendent of the LCA of all _living_ human species.
If you want to include Neanderthals in a crown group then you
also have to include the Chimps since they are our closest living
relatives.
Eric
an...@sci.sci
> >> Tyrannosaurus than to modern birds.
> >Closer (less distance between them) per what metric?
> Common descent.
Common descent is a theory, not a metric, except in the sense of a
boolean metric where two species either share a common ancestor
(distance zero) or don't (distance 100%).
You need to specify what metric you really mean when you say "closer".
For example, here's a hypothetical cladogram:
3my
+---Foo
---+
| 4my+--Bar
+----+2my
|
+----------------------------------------------------Blortch
52my
Bar is closer to:
Foo in terms of total evolutionary distance along both branches from LCA
(9 from Foo compared to 54 my from Blortch)
Blortch in terms of absolute recency of LCA
(LCA with Blortch is 4my more recent than LCA with Foo)
Foo in terms of relative recency of LCA compared to latest taxon of the two
(LCA with Foo is 6 my before latest of Foo and Bar,
LCA with Blortch is 52 my before latest of Bar and Blortch)
Foo in terms of relative recency of LCA compared to average taxon of the two
(LCA with Foo is 4.5 my before average of Foo and Bar,
LCA with Blortch is 27 my before average of Bar and Blortch)
So if "Blortch-like" is defined as "closer to Blortch than to Foo", is
Bar "Blortch-like" or not? And with a more complicated cladogram with
many taxa, is "Blorth-like" necessarily monophyletic under each of the
four definitions of "closer"?
I think a cladistic definition, either the smallest clade containing
two taxa, or the largest clade containing one taxon but not another, is
less ambiguous than any "closer than" definition.
> >I'm not interested in the details so much. I just want one clade that
> >contains two sub-clades, one containing all the modern birds whatsoever
> >(don't care whether Archy is in it or not),
> Wasn't including Archaeopteryx the entire point of this exercise?
Well, originally I wanted Archaeopteryx to be outside the "modern
birds" group but inside the overall clade that includes both "modern
birds" and at least one member of "ancient true dinosaurs, terrible
lizards". But I got distracted by the nitpicking about various things,
and for a moment I didn't care at all where Archy fit in the system.
At this point I'm pretty well decided that the scale should run from
Coelurosauria to Neornithes, where Coelurosauria is the LCA of
Tyranosaurus and Neornithes. Maniraptoriformes is not an acceptable
taxon because it's not well defined, and anything later such as
Maniraptora is too far down the path to modern birds. Anything before
Coelurosauria is unnecessarily inclusive.
> Coelurosauria would be the place to start.
Excellent. We came to the same conclusion, independently, based on
looking at the same evidence. (Try getting the YECs and the IDiots to
do the same, haha!) Hey, you want to play Darwin while I play Wallace,
or vice versa? I saw a guy named Tom Hendricks who should play Bush!
(I think he believes in the Solar thermal cycle of mass construction.)
> >> But then, a creationist could exclude the characters that you consider
> >> important, on the basis that they're "noise".
> >If a Creationist wants to make up a list of characters, such that they
> >distinguish modern birds from large dinosaurs, and such that for some
> >characters Archy is more like modern birds and for other characers
> >Archy is more unlike modern birds, so that by appropriate weighting I
> >can accomplish my goal of placing Archy at the half-and-half point in
> >the weighted scale, that's fine with me.
> But that's not what I meant. A creationist would exclude as "noise",
> any trait that pushed Archaeopteryx out of Neornithes.
Well I guess you can turn the creationist over to the NSA to be
exported to a foreign country where it's OK to torture prisoners,
and withhold food until he comes up with at least one character that
places Archaeopteryx closer to dinosaurs than to living birds,
regardless of how he feels Neornithes should be defined. Don't mention
Neornithes to him at all. Just mention two very different kinds of
currently-living birds that are in opposite sub-branches immediately
under Neornithes, for example
`--Neornithes
|--Paleognathae
| `--Ratites
| `--Rheiformes ***
`--GALLOANSERAE
`--Phasianidae ***
So starve him until he can find some character shared by Rheiformes and
Phasianidae but which neither Archaeopteryx nor Tyranosaurus has.
(Presumably he's already found it easy to find some character shared by
Rheiformes and Phasianidae and Archaeopteryx but not Tyranosaurus.)
I know what you must all be saying to yourselves. If that's all there
is to interrogation, just starve the terrorist symphathizer until he
gives up his secret, then why hasn't Bush found Osama in four years of
such torture-based interrogations? Well you see, Bush works within the
so-called "intelligence community", which in his case is a misnomer.
> >> >> The problem is that Archaeopteryx isn't a modern bird.
> >> >That sentence is meaningless in the modern context of evolution. It
> >> >makes sense only in the old Platonic/Lamarckian philosophy of fixed
> >> >ideal/metaphysical types which are only approximated by physical
> >> >objects. (Or it's trivially true simply because Archaeopteryx didn't
> >> >survive into modern times.)
> >> Actually, "Archaeopteryx isn't a modern bird" makes a *lot* of sense,
> >> in that it lacked many traits possessed by the birds alive today.
> >We're in agreement.
> Then why did you state that my sentence was meaningless?
We're in agreement that Archaeopteryx lacked many traits possessed by
the birds alive today. But modern birds are by definition those birds
which are alive today, not ones that went extinct more than fifty
million years ago, regardless of their characteristics. So the
statement that Archaeopteryx is not a modern bird is a worthless
statement. Anything that went extinct as recently as three million
years ago is not a modern bird in the sense of not ever co-existing
with modern humans, even if it is within the clade of modern birds.
Crocodiles are the nearest living relatives of modern birds (in the
sense of having a most recent common ancestor). Crocodiles are living
today, so there's an actual question whether to consider them modern
birds or not. The answer is no they aren't considered modern birds. But
Archaeopteryx died out long long ago, so there isn't even a question
whether it *is* a modern bird. It isn't modern, it hasn't been alive
any time that would be considered "modern", so it isn't a modern
anything.
Was Behe a 20th century scientist? Good question. Answer: No, he's a sham.
Was Galileo a 20th century scientist? Stupid question not worth answering.
Do you understand the difference between the questions whether
Archaeopteryx was within the smallest clade containing modern birds
(good question, no, not in that clade), vs. whether Archaeopteryx *is*
a modern bird (stupid question)?
It was just the wording of your remark that was meaningless. Slightly
different wording, using cladistic terminology, and I wouldn't have
nitpicked it.
> >If you list some such traits, where Archy is on the
> >non-bird side of the 50% point, and some other traits, where Archy is
> >bird side of the 50% point, that will be fine.
> Since you're going to arbitrarily exclude traits that you don't like,
> and arbitrarily give more or less weight to others, I don't see the
> point.
You're misunderstanding what I wrote if you think I'd exlude any
valid trait. Valid for this purpose means:
- It can be and has been measured accurately for all the taxa we consider.
- It clearly distinguishes between following two sub-clades of Coelurosauria:
-- the maximal sub-clade containing Tyranosaurus but not containing Neornithes
-- the very small sub-clade Neornithes itself.
- Each trait is quantitative, i.e. gives a numerical value instead of
just a natural-language description.
Do you know of a good Web site that lists/describes such quantitative
traits, and gives the values for each trait applied to at least the
following taxa: Tyranosaurus, Archaeopteryx, and a few modern-bird
taxa?
> Creationists don't recognize the existence of clades.
Well they're stupid enough that you might be able to convince them that
Archaeopteryx and a sister species were on the Ark, and all the of the
modern birds came from one or both of those via micro-evolution, and
then assign the task of glorifying God by studying the post-Flood
micro-evolution in detail using fossils whose time stamps are scaled to
the post-Flood timescale. (Archy was 1998 BC, Pygostylia founder was
1875 BC, Neornithes founder was 1500 BC, etc.) If they figure out,
based on fossil evidence, that Archy didn't leave any modern
descendents, only the sister species did, that would be a start toward
converting the creationist toward scientific methodology.
> >The smallest clade containing all modern
> >non-human apes *does* include humans.
> And surprisingly, creationists don't recognize that humans are apes
> either.
Well actually humans aren't apes any more than eukaryotes are bacteria.
Humans are an endosymbiosis of ape ancestor and immortal soul. :-)
(Likewise, according to the Hundus anyway, Cows are an endosymbiosis of ...)
> We're not talking about cladists, we're talking about creationists.
There's no such thing as a cladist! Them's fighting words! :-)
I suggest taking a Creationist to a photocopy (actually Xerox or Kodak
copy, it doesn't *really* use the photography process) room, and ask
the Creationist to draw a picture on a previously blank sheet of paper.
Then make two copies, and ask the Creationist to mark up each copy in
identical ways, as best he can do. Then make three copies of each of
those marked-copies. Then ask the Creationist to mark up each
second-generation copy in yet another identical way. Then shuffle all
the 1+3+9 = 13 sheets, toss three at random in the trash, and ask him
to figure out the pedigree of the ten remaining pieces. (If one of the
three tossed-out is the *original*, even more fun!) Don't give him any
food until he accepts that such reconstruction is possible and he
actually performs it successfully.
Now after he successfully reconstructs the correct cladogram for those
ten pieces of paper, ask him if he believes in descent (by such
copying) with modification. Now tell him you don't believe any of those
ten pieces of paper were copied. You believe they were all originals he
fabricated by hand, each individually. Also, complain about the three
"missing links", and say that proves they weren't copied, that copying
with mark-up is just a joke. See how he feels on the other end of the
debate.
> >Any place I wrote such words, please understand that I meant to say
> >"modern bird" and "not modern bird" clades, as the two sub-clades of
> >the LCA of modern birds and some outgroup terrible lizard such as
> >Tyranosaurus.
> Then please write what you mean. And please use terms that have
> commonly accepted meanings.
I'm just a layman in this field. I try to use correct jargon, but I
don't know the correct jargon when I first start talking about some
topic, so I make up terms that seem reasonable, and then when somebody
suggests the correct term I look it up online to see whether that's the
term I should have used or not. If not, then I explain the difference
between the proposed term and what I meant to say, and ask for another
suggestion what term to use.
So right now, given ancestral taxon C, descendent taxon T, and
descendent taxon N, I don't know the correct term for the largest clade
descended from C and containing T but not N. Please tell me the correct
jargon for that. (There's a reason for picking letters C T and N, which
I presume you have figured out almost immediately.)
> >I was assuming the cladograms I found on the Web were commonly agreed
> >upon,
> Keep in mind that this *is* the web we're talking about. Information
> found there is under no obligation to be current, or even accurate.
Well when somebody puts up a personal Web page or posts to a newsgroup,
yes I understand it's all half crap. But when there's an online service
that gives cladograms in nice format, I really assumed those were based
on some science, not gross guesswork.
> Metornithes is probably synonymous with Maniraptoriformes.
Metornithes seems to be sister taxa to Archaeopteryx.
Maniraptoriformes seems to be immediately under Coelurosauria, and
sister to Tyrannosauroidea, or parent of Tyrannosauroidea, according to
different experts.
There's no way Metornithes and Maniraptoriformes could be synomyms.
> You could always check the primary literature.
You mean like those *Research*Reports* in _Science_ which are 99% inscrutable?
> At this point, not all nodes have names.
That's very good. I wish even more nodes never had names, namely the
nodes that weren't well known at the time the names were assigned, and
now everyone wishes the name hadn't been mis-assigned as it was.
> >Is there a search engine where I could feed in a small set of species
> >names and get back just the minimal descendency tree that links all my
> >species together, where I get the latest official or well respected
> >descendency data, nothing seven years out of date? For example, if I
> >put in Tyranosaurus and Neornithes it'd give me a V-shaped tree that
> >had Coelurosauria or somesuch at the top and all the names from that
> >LCA to Neornithes along one branch of the V, and all the names from the
> >LCA to Tyranosaurus along the other branch of the V?
> None that I know of.
Then I guess maybe I should hire somebody to write that sofware.
All the needed data is on the Web. It's just a matter of collecting it,
comparing different sources to omit any parts of the tree that are in
dispute, and then putting all that collected data into a database and
computing ternary-string addresses for each node (0 for stop here, 1
for first branch, 2 for second branch), and then when anybody requests
two taxa the program looks up the ternary address for each, compares
them to determine LCA, then performs lookups on each ternary prefix of
the two specified taxa to generate the names of the nodes along the V.
I bet a high-school student who took a Java/DB class could do the main
part of the program in a week or two. And I bet one of the spiders used
to collect data for Google or Yahoo could be modified to collect all
the needed data to populate the database. (The Google API might do it
rather easily using data Google already collected.) All we'd need then
is another high school student to write the Web interface.
Given input Tyranosaurus and Rheiformes, it'd output something like this:
Coelurosauria
+-> Maniraptora -> Avialae -> Neornithes -> Paleognathae ->
| Ratites -> Rheiformes
`-> Tyrannosauroidea -> Tyrannosauridae -> Tyrannosaurinae ->
Tyrannosaurini -> Tyrannosaurus
(Every time I see that Tyrannosaur with ini it gets me silly.)
(Of course seeing Tyrannosaur with us is weird now that I think about it.)
(I think I need to go to bed. It's after 4 AM already, and I'm getting weird.)
Note: On wide screen it wouldn't break the long lines like I did here.
.
John Wilkins
....
>>Things work better when you have at least some stable nodes, or a
>>backbone of stable taxa. That's why crown groups are so attractive.
>
>
> I don't know what crown groups are. I tried to look it up on Google,
> and got this totally sucky definition:
> <http://www.peripatus.gen.nz/Biology/defStemCrown.html>
> Clade
> A monophyletic taxon; a group of organisms which includes the most
> recent common ancestor of all of its members and all of the
> descendants of that most recent common ancestor.
> That's correct, but:
>
> Crown Group
> Those organisms descended from the last common ancestor of all living
> members of a clade (Runnegar 1992).
> The last common ancestor of a living clade plus all of its descendants
> (Budd 2001, p. 333).
> But that's exactly the same as the clade itself: One individual
> (species, in case of eukaryotes that reproduce via meiosis), and all
> descendents of that individual (species).
>
> Do you know of any definition whereby the crown group isn't the whole clade?
> ..
>
Aren't crown groups *extant* groups? IOW, CGs are a subset of clades.
--
John S. Wilkins, Postdoctoral Research Fellow, Biohumanities Project
University of Queensland - Blog: evolvethought.blogspot.com
Nihil tam absurdum quod non quidam Philosophi dixerit - adapted from Cicero
John Harshman
Those are good definitions, and I have trouble seeing what you have
missed. Note "all *living* members" of a clade. To put it more simply,
it's a special form of node-based definition, in which all the reference
taxa are extant. Mammalia, defined as a crown group, includes the common
ancestor of you, a kangaroo, and a platypus, and all its descendants.
Thus Morganucodon is excluded from crown-group Mammalia. Mammalia,
defined this way, is the crown group of Synapsida (or Therapsida, or
Cynodontia, whatever you like). Neornithes is a crown group. Avialae is
not. And so on. Since we have a better chance of knowing the
relationships among extant taxa than among fossils (much more data to
work with for each taxon), crown groups are expected to be more stable
than other groups; at least their extant backbones are.
John Harshman
>>>>What I think you mean is the clade composed of all animals closer to
>>>>Tyrannosaurus than to modern birds.
>>>
>>>Closer (less distance between them) per what metric?
>>
>>Common descent.
>
>
> Common descent is a theory, not a metric, except in the sense of a
> boolean metric where two species either share a common ancestor
> (distance zero) or don't (distance 100%).
>
> You need to specify what metric you really mean when you say "closer".
> For example, here's a hypothetical cladogram:
>
> 3my
> +---Foo
> ---+
> | 4my+--Bar
> +----+2my
> |
> +----------------------------------------------------Blortch
> 52my
>
> Bar is closer to:
[snip]
> Blortch in terms of absolute recency of LCA
> (LCA with Blortch is 4my more recent than LCA with Foo)
That was the intended meaning.
[snip]
> So if "Blortch-like" is defined as "closer to Blortch than to Foo", is
> Bar "Blortch-like" or not? And with a more complicated cladogram with
> many taxa, is "Blorth-like" necessarily monophyletic under each of the
> four definitions of "closer"?
"Closer to" and "-like" are two different concepts. It's not clear to me
when or where you woudl want to use the latter.
> I think a cladistic definition, either the smallest clade containing
> two taxa, or the largest clade containing one taxon but not another, is
> less ambiguous than any "closer than" definition.
I don't know why you guys are arguing here, since you are actually both
using the same definition. I think you must be trying hard to
misunderstand each other.
[snip]
>>>>Actually, "Archaeopteryx isn't a modern bird" makes a *lot* of sense,
>>>>in that it lacked many traits possessed by the birds alive today.
>>>
>>>We're in agreement.
>>
>>Then why did you state that my sentence was meaningless?
>
> We're in agreement that Archaeopteryx lacked many traits possessed by
> the birds alive today. But modern birds are by definition those birds
> which are alive today, not ones that went extinct more than fifty
> million years ago, regardless of their characteristics.
I bet Augray was using "modern birds" as a synonym for "Neornithes".
That's how I read it. Thus there were modern birds alive 50 million
years ago, most of which are of course extinct today.
> So the
> statement that Archaeopteryx is not a modern bird is a worthless
> statement. Anything that went extinct as recently as three million
> years ago is not a modern bird in the sense of not ever co-existing
> with modern humans, even if it is within the clade of modern birds.
This is a silly argument. You seem to be arguing for the sake of
arguing. (No I'm not. Yes you are. No I'm not.)
> Crocodiles are the nearest living relatives of modern birds (in the
> sense of having a most recent common ancestor). Crocodiles are living
> today, so there's an actual question whether to consider them modern
> birds or not.
This too is silly. You might as well, in that case, ask whether turtles
are modern birds, or whether lizards are, or foxes, or frogs... Where
could you stop?
> The answer is no they aren't considered modern birds. But
> Archaeopteryx died out long long ago, so there isn't even a question
> whether it *is* a modern bird. It isn't modern, it hasn't been alive
> any time that would be considered "modern", so it isn't a modern
> anything.
>
> Was Behe a 20th century scientist? Good question. Answer: No, he's a sham.
> Was Galileo a 20th century scientist? Stupid question not worth answering.
This all hinges on a perverse (and apparently willfull) misunderstanding
of what Augray means by "modern". If you guys would just try to
understand what the other one is saying, rather than look for things to
pounce on, everything would go better.
[snip]
>>Since you're going to arbitrarily exclude traits that you don't like,
>>and arbitrarily give more or less weight to others, I don't see the
>>point.
>
>
> You're misunderstanding what I wrote if you think I'd exlude any
> valid trait. Valid for this purpose means:
> - It can be and has been measured accurately for all the taxa we consider.
> - It clearly distinguishes between following two sub-clades of Coelurosauria:
> -- the maximal sub-clade containing Tyranosaurus but not containing Neornithes
> -- the very small sub-clade Neornithes itself.
> - Each trait is quantitative, i.e. gives a numerical value instead of
> just a natural-language description.
Why would you want to do this?
> Do you know of a good Web site that lists/describes such quantitative
> traits, and gives the values for each trait applied to at least the
> following taxa: Tyranosaurus, Archaeopteryx, and a few modern-bird
> taxa?
No. Systematists tend to like qualitative characters that can be scored
as discrete states. Easier to analyze. If you wanted quantitative traits
you would end up mostly with descriptions of Tyrannosaurus' large size.
[snip]
Stanley Friesen
>
>Those are good definitions, and I have trouble seeing what you have
>missed. Note "all *living* members" of a clade. To put it more simply,
>it's a special form of node-based definition, in which all the reference
>taxa are extant. Mammalia, defined as a crown group, includes the common
>ancestor of you, a kangaroo, and a platypus, and all its descendants.
However, when the platypus and echidna become extinct, that will no
longer be a crown group.
>Thus Morganucodon is excluded from crown-group Mammalia. Mammalia,
>defined this way, is the crown group of Synapsida (or Therapsida, or
>Cynodontia, whatever you like). Neornithes is a crown group. Avialae is
>not. And so on. Since we have a better chance of knowing the
>relationships among extant taxa than among fossils (much more data to
>work with for each taxon), crown groups are expected to be more stable
>than other groups; at least their extant backbones are.
Well, except for extinction.
Augray
>> >> What I think you mean is the clade composed of all animals closer to
>> >> Tyrannosaurus than to modern birds.
>> >Is there a difference? I found it entirely clear. It had the form of
>> >some stem-based definitions in common use, except they substitute "most
>> >inclusive" for "largest".
>> I wasn't sure if he (?) was making reference to some paraphyletic
>> clade or not.
>
>*oxymoron*alert* Duck and cover!
>(By definition, any clade is monophyletic!)
Let me rephrase that to "paraphyletic *group*"
I'd suggest http://tolweb.org/tree?group=Tyrannosauroidea and
http://tolweb.org/tree?group=Tyrannosauridae&contgroup=Tyrannosauroidea
which are more recent.
>Bad joke: Tyrannosaurini are mis-named, since they weren't placental
>mammals, so they didn't have *any* kind of belly button, not an outi
>nor an ini.
>
>By the way, when I was a kid I always assumed that Pterosaurs were the
>missing link between dinosaurs and birds, but now I realize they are
>out a ways along just a side branch from quite a bit before the LCA of
>Tyrannosaurus and birds, essentially convergent evolution with bats
>rather than birds anyway.
I'm not sure that I'd necessarily call it convergent. Aside from the
patagium (and even that's debatable), they don't seem to have had a
lot in common.
>Hey, here's a rather complete cladogram of all types of dinosaurs:
><http://town.morrison.co.us/dino-colo/taxonomy/dinoclad.php>
>
>> >It looks as if Aves may be on the way out because of inability to agree
>> >on its definition.
>> That's too bad. I have to admit that I prefer Aves when referring to
>> the crown clade, as it matches Linnaeus's original intent.
>
>Linnaeus defined the terminology for currently-living organisms only.
>Unfortunately when different people tried to exted the term 'Aves' to
>fossils, they IMO assigned the term prematurely, before they knew
>enough about the fossils to know where the right place should be.
They felt that they knew what the right place should be.
Augray
>> >> What I think you mean is the clade composed of all animals closer to
>> >> Tyrannosaurus than to modern birds.
>> >Closer (less distance between them) per what metric?
>>
>> Common descent.
>
>Common descent is a theory, not a metric, except in the sense of a
>boolean metric where two species either share a common ancestor
>(distance zero) or don't (distance 100%).
Are you saying that common descent isn't a fact?
>You need to specify what metric you really mean when you say "closer".
>For example, here's a hypothetical cladogram:
>
> 3my
> +---Foo
> ---+
> | 4my+--Bar
> +----+2my
> |
> +----------------------------------------------------Blortch
> 52my
>
>Bar is closer to:
> Foo in terms of total evolutionary distance along both branches from LCA
> (9 from Foo compared to 54 my from Blortch)
> Blortch in terms of absolute recency of LCA
> (LCA with Blortch is 4my more recent than LCA with Foo)
> Foo in terms of relative recency of LCA compared to latest taxon of the two
> (LCA with Foo is 6 my before latest of Foo and Bar,
> LCA with Blortch is 52 my before latest of Bar and Blortch)
> Foo in terms of relative recency of LCA compared to average taxon of the two
> (LCA with Foo is 4.5 my before average of Foo and Bar,
> LCA with Blortch is 27 my before average of Bar and Blortch)
What makes you think that time is a useful metric for evolutionary
distance? What if you don't know the date of divergence?
>So if "Blortch-like" is defined as "closer to Blortch than to Foo", is
>Bar "Blortch-like" or not? And with a more complicated cladogram with
>many taxa, is "Blorth-like" necessarily monophyletic under each of the
>four definitions of "closer"?
>
>I think a cladistic definition, either the smallest clade containing
>two taxa, or the largest clade containing one taxon but not another, is
>less ambiguous than any "closer than" definition.
What you've just described are respectively called node definitions
and stem definitions.
>> >I'm not interested in the details so much. I just want one clade that
>> >contains two sub-clades, one containing all the modern birds whatsoever
>> >(don't care whether Archy is in it or not),
>> Wasn't including Archaeopteryx the entire point of this exercise?
>
>Well, originally I wanted Archaeopteryx to be outside the "modern
>birds" group but inside the overall clade that includes both "modern
>birds" and at least one member of "ancient true dinosaurs, terrible
>lizards". But I got distracted by the nitpicking about various things,
>and for a moment I didn't care at all where Archy fit in the system.
>
>At this point I'm pretty well decided that the scale should run from
>Coelurosauria to Neornithes, where Coelurosauria is the LCA of
>Tyranosaurus and Neornithes. Maniraptoriformes is not an acceptable
>taxon because it's not well defined,
What makes you think that? It's very well defined. It's the last
common ancestor of _Ornithomimus_ and Neornithes and all it's
descendents.
[snip]
In favour of a scientific inquisition, are we?
>> >> >> The problem is that Archaeopteryx isn't a modern bird.
>> >> >That sentence is meaningless in the modern context of evolution. It
>> >> >makes sense only in the old Platonic/Lamarckian philosophy of fixed
>> >> >ideal/metaphysical types which are only approximated by physical
>> >> >objects. (Or it's trivially true simply because Archaeopteryx didn't
>> >> >survive into modern times.)
>> >> Actually, "Archaeopteryx isn't a modern bird" makes a *lot* of sense,
>> >> in that it lacked many traits possessed by the birds alive today.
>> >We're in agreement.
>> Then why did you state that my sentence was meaningless?
>
>We're in agreement that Archaeopteryx lacked many traits possessed by
>the birds alive today. But modern birds are by definition those birds
>which are alive today, not ones that went extinct more than fifty
>million years ago, regardless of their characteristics.
I'd debate that. Was the Passenger Pigeon a modern bird?
>So the
>statement that Archaeopteryx is not a modern bird is a worthless
>statement. Anything that went extinct as recently as three million
>years ago is not a modern bird in the sense of not ever co-existing
>with modern humans, even if it is within the clade of modern birds.
What makes you think that I was using it in the "in the sense of not
ever co-existing with modern humans"?
>Crocodiles are the nearest living relatives of modern birds (in the
>sense of having a most recent common ancestor). Crocodiles are living
>today, so there's an actual question whether to consider them modern
>birds or not. The answer is no they aren't considered modern birds. But
>Archaeopteryx died out long long ago, so there isn't even a question
>whether it *is* a modern bird.
But "living" isn't the criteria as to whether an animal should be
included in Neornithes.
>It isn't modern, it hasn't been alive
>any time that would be considered "modern", so it isn't a modern
>anything.
>
>Was Behe a 20th century scientist? Good question. Answer: No, he's a sham.
>Was Galileo a 20th century scientist? Stupid question not worth answering.
>
>Do you understand the difference between the questions whether
>Archaeopteryx was within the smallest clade containing modern birds
>(good question, no, not in that clade), vs. whether Archaeopteryx *is*
>a modern bird (stupid question)?
Let me play Creationist's Advocate for a moment, and say that I don't.
Then what?
>It was just the wording of your remark that was meaningless. Slightly
>different wording, using cladistic terminology, and I wouldn't have
>nitpicked it.
>
>> >If you list some such traits, where Archy is on the
>> >non-bird side of the 50% point, and some other traits, where Archy is
>> >bird side of the 50% point, that will be fine.
>> Since you're going to arbitrarily exclude traits that you don't like,
>> and arbitrarily give more or less weight to others, I don't see the
>> point.
>
>You're misunderstanding what I wrote if you think I'd exlude any
>valid trait. Valid for this purpose means:
>- It can be and has been measured accurately for all the taxa we consider.
>- It clearly distinguishes between following two sub-clades of Coelurosauria:
>-- the maximal sub-clade containing Tyranosaurus but not containing Neornithes
>-- the very small sub-clade Neornithes itself.
>- Each trait is quantitative, i.e. gives a numerical value instead of
>just a natural-language description.
My apologies for claiming that you were going to arbitrarily exclude
traits.
>Do you know of a good Web site that lists/describes such quantitative
>traits, and gives the values for each trait applied to at least the
>following taxa: Tyranosaurus, Archaeopteryx, and a few modern-bird
>taxa?
Not really.
>> Creationists don't recognize the existence of clades.
>
>Well they're stupid enough that you might be able to convince them that
>Archaeopteryx and a sister species were on the Ark, and all the of the
>modern birds came from one or both of those via micro-evolution,
That's doubtful, considering that a pair of ravens and at least one
pair of doves were on board.
>and
>then assign the task of glorifying God by studying the post-Flood
>micro-evolution in detail using fossils whose time stamps are scaled to
>the post-Flood timescale. (Archy was 1998 BC, Pygostylia founder was
>1875 BC, Neornithes founder was 1500 BC, etc.) If they figure out,
>based on fossil evidence, that Archy didn't leave any modern
>descendents, only the sister species did, that would be a start toward
>converting the creationist toward scientific methodology.
>
>> >The smallest clade containing all modern
>> >non-human apes *does* include humans.
>> And surprisingly, creationists don't recognize that humans are apes
>> either.
>
>Well actually humans aren't apes any more than eukaryotes are bacteria.
>Humans are an endosymbiosis of ape ancestor and immortal soul. :-)
>(Likewise, according to the Hundus anyway, Cows are an endosymbiosis of ...)
I see...
>> We're not talking about cladists, we're talking about creationists.
>
>There's no such thing as a cladist! Them's fighting words! :-)
>
>I suggest taking a Creationist to a photocopy (actually Xerox or Kodak
>copy, it doesn't *really* use the photography process) room, and ask
>the Creationist to draw a picture on a previously blank sheet of paper.
>Then make two copies, and ask the Creationist to mark up each copy in
>identical ways, as best he can do. Then make three copies of each of
>those marked-copies. Then ask the Creationist to mark up each
>second-generation copy in yet another identical way. Then shuffle all
>the 1+3+9 = 13 sheets, toss three at random in the trash, and ask him
>to figure out the pedigree of the ten remaining pieces. (If one of the
>three tossed-out is the *original*, even more fun!) Don't give him any
>food until he accepts that such reconstruction is possible and he
>actually performs it successfully.
>
>Now after he successfully reconstructs the correct cladogram for those
>ten pieces of paper, ask him if he believes in descent (by such
>copying) with modification. Now tell him you don't believe any of those
>ten pieces of paper were copied. You believe they were all originals he
>fabricated by hand, each individually. Also, complain about the three
>"missing links", and say that proves they weren't copied, that copying
>with mark-up is just a joke. See how he feels on the other end of the
>debate.
A creationist would say that since there were no witnesses in the
past, coming to a conclusion of common ancestry would be premature.
>> >Any place I wrote such words, please understand that I meant to say
>> >"modern bird" and "not modern bird" clades, as the two sub-clades of
>> >the LCA of modern birds and some outgroup terrible lizard such as
>> >Tyranosaurus.
>> Then please write what you mean. And please use terms that have
>> commonly accepted meanings.
>
>I'm just a layman in this field. I try to use correct jargon, but I
>don't know the correct jargon when I first start talking about some
>topic, so I make up terms that seem reasonable, and then when somebody
>suggests the correct term I look it up online to see whether that's the
>term I should have used or not. If not, then I explain the difference
>between the proposed term and what I meant to say, and ask for another
>suggestion what term to use.
>
>So right now, given ancestral taxon C, descendent taxon T, and
>descendent taxon N, I don't know the correct term for the largest clade
>descended from C and containing T but not N. Please tell me the correct
>jargon for that.
That's a stem clade. Do you remember figure 1 in
http://palaeo.gly.bris.ac.uk/Essays/phylocode/biolrev.html ?
> (There's a reason for picking letters C T and N, which
>I presume you have figured out almost immediately.)
>
>> >I was assuming the cladograms I found on the Web were commonly agreed
>> >upon,
>> Keep in mind that this *is* the web we're talking about. Information
>> found there is under no obligation to be current, or even accurate.
>
>Well when somebody puts up a personal Web page or posts to a newsgroup,
>yes I understand it's all half crap. But when there's an online service
>that gives cladograms in nice format, I really assumed those were based
>on some science, not gross guesswork.
Creationist websites have nice formats too. Why not use the info
there?
>> Metornithes is probably synonymous with Maniraptoriformes.
>
>Metornithes seems to be sister taxa to Archaeopteryx.
Based on what?
>Maniraptoriformes seems to be immediately under Coelurosauria, and
>sister to Tyrannosauroidea, or parent of Tyrannosauroidea, according to
>different experts.
>There's no way Metornithes and Maniraptoriformes could be synomyms.
Why not?
>> You could always check the primary literature.
>
>You mean like those *Research*Reports* in _Science_ which are 99% inscrutable?
Yes. Learning things can be fun.
>> At this point, not all nodes have names.
>
>That's very good. I wish even more nodes never had names, namely the
>nodes that weren't well known at the time the names were assigned, and
>now everyone wishes the name hadn't been mis-assigned as it was.
>
>> >Is there a search engine where I could feed in a small set of species
>> >names and get back just the minimal descendency tree that links all my
>> >species together, where I get the latest official or well respected
>> >descendency data, nothing seven years out of date? For example, if I
>> >put in Tyranosaurus and Neornithes it'd give me a V-shaped tree that
>> >had Coelurosauria or somesuch at the top and all the names from that
>> >LCA to Neornithes along one branch of the V, and all the names from the
>> >LCA to Tyranosaurus along the other branch of the V?
>> None that I know of.
>
>Then I guess maybe I should hire somebody to write that sofware.
Go right ahead.
John Harshman
> John Harshman <jharshman....@pacbell.net> wrote:
>
>>Those are good definitions, and I have trouble seeing what you have
>>missed. Note "all *living* members" of a clade. To put it more simply,
>>it's a special form of node-based definition, in which all the reference
>>taxa are extant. Mammalia, defined as a crown group, includes the common
>>ancestor of you, a kangaroo, and a platypus, and all its descendants.
>
>
> However, when the platypus and echidna become extinct, that will no
> longer be a crown group.
There's a statute of limitations. Don't worry.
>>Thus Morganucodon is excluded from crown-group Mammalia. Mammalia,
>>defined this way, is the crown group of Synapsida (or Therapsida, or
>>Cynodontia, whatever you like). Neornithes is a crown group. Avialae is
>>not. And so on. Since we have a better chance of knowing the
>>relationships among extant taxa than among fossils (much more data to
>>work with for each taxon), crown groups are expected to be more stable
>>than other groups; at least their extant backbones are.
>
> Well, except for extinction.
The usual way around this is to propose a limiting date: extant as of
such and such a date, or as of the date the definition was proposed, or
whatever.
John Harshman
> On Sat, 17 Dec 2005 18:19:10 -0800, an...@sci.sci wrote:
>
>
>>>>>What I think you mean is the clade composed of all animals closer to
>>>>>Tyrannosaurus than to modern birds.
>>>>
>>>>Is there a difference? I found it entirely clear. It had the form of
>>>>some stem-based definitions in common use, except they substitute "most
>>>>inclusive" for "largest".
>>>
>>>I wasn't sure if he (?) was making reference to some paraphyletic
>>>clade or not.
>>
>>*oxymoron*alert* Duck and cover!
>>(By definition, any clade is monophyletic!)
>
>
> Let me rephrase that to "paraphyletic *group*"
Thanks. I've always found "monophyletic clade" annoying, and
"paraphyletic clade" is worse. S. J. Gould was guilty of both of these,
so you're in good company.
>>>>Better question: What is the point of defining a "bird-specific branch"?
>>>
>>>So far, it seems to be a need to arrive at a predetermined outcome.
>>>I'm just trying to peel back the layers.
>>
>>Yes. The idea is to define a branch that leads from the last common
>>ancestor of the modern birds and some everyone-agrees non-bird dinosaur
>>(such as Tyranosaurus, the corresponding common ancestor being either
>>Coelurosauria or Maniraptoriformes), through the various transition
>>forms of ancient bird-like dinosaurs (common ancestors of modern birds
>>and Archaeopteryx) and dinosaur-like birds (cousins to Archaeopteryx)
>>and birds-but-not-modern-birds (such as Euornithes) all the way to the
>>LCA of all modern (currently living) birds (Neornithes). Then I have
>>one additional constraint that the LCA of modern birds and
>>Archaeopteryx, which is very close to Archaeopteryx in the evolutionary
>>tree, should score exactly halfway between Coelurosauria and
>>Neornithes, per some weighting of the various characters that
>>distinguish those two points in the tree.
I repeat: why?
[snip]
>>By the way, when I was a kid I always assumed that Pterosaurs were the
>>missing link between dinosaurs and birds, but now I realize they are
>>out a ways along just a side branch from quite a bit before the LCA of
>>Tyrannosaurus and birds, essentially convergent evolution with bats
>>rather than birds anyway.
>
>
> I'm not sure that I'd necessarily call it convergent. Aside from the
> patagium (and even that's debatable), they don't seem to have had a
> lot in common.
Oh, come on. Wing made of skin stretched over a bone framework?
Quadrupedal when on the ground? It's not that bad.
[snip]
>>>That's too bad. I have to admit that I prefer Aves when referring to
>>>the crown clade, as it matches Linnaeus's original intent.
>>
>>Linnaeus defined the terminology for currently-living organisms only.
>>Unfortunately when different people tried to exted the term 'Aves' to
>>fossils, they IMO assigned the term prematurely, before they knew
>>enough about the fossils to know where the right place should be.
>
>
> They felt that they knew what the right place should be.
The better question is whether there is any right place, objectively
speaking.
[snip]
John Harshman
> On Sun, 18 Dec 2005 04:06:00 -0800, an...@sci.sci wrote:
>
>
>>>>>What I think you mean is the clade composed of all animals closer to
>>>>>Tyrannosaurus than to modern birds.
>>>>
>>>>Closer (less distance between them) per what metric?
>>>
>>>Common descent.
>>
>>Common descent is a theory, not a metric, except in the sense of a
>>boolean metric where two species either share a common ancestor
>>(distance zero) or don't (distance 100%).
>
>
> Are you saying that common descent isn't a fact?
I'm not sure how this started, but you guys are in a descending spiral
of weird misunderstandings and accusations. What's going on here?
[snip]
an...@sci.sci
> birds.
Given that the term "birds" by itself is pretty much meaningless (might
include only currently-living birds, and others within the same LCA, or
might include ancient bird-like dinosaurs all the way back to before
Archy, your sentence there can't possibly be true.
> Also more or less closely related to birds are _Protarchaeopteryx_
> and the alvarezsaurids (which some believe to actually *be* birds).
That parenthetical remark of yours is meaningless, except in some
Platonic/religious system where not only are there ideal categories in
metaphysics-never-land but God assigned names to each of them.
> There is an early undescribed maniraptoran from the Late Jurassic
> Morrison Formation, so the origin of birds could easily be some where
> around the Mid Jurassic.
More meaningless gibberish in that part after the comma.
> > We don't yet know exactly
> >which Triassic species was the last-common-ancestor of modern birds and
> >non-avian Jurassic/Cretaceous theropods such as Tyrannosaurus, and even
> >if we could pinpoint that LCA, we might not have complete fossils, so
> >we might not be able to rate that single species on all characters.
> That LCA may well have been as recent as the mid Jurassic.
Finally, something you said that has any meaning.
There seems to be concensus that the LCA happened sometime before the
end of the Jurassic, which is good enough for me now.
<http://www.taxonsearch.org/dev/taxon_edit.php?Action=View&tax_id=97>
Taxon Coelurosauria
...
Active Definition The most inclusive clade containing Passer
domesticus (Linnaeus 1758) but not Allosaurus fragilis Marsh 1877,
Sinraptor dongi Currie and Zhao 1993, Carcharodontosaurus saharicus
(Depiret and Savornin 1927).
I looked up those genera:
Passer = the sparrow
<http://www.miketaylor.org.uk/dino/faq/s-class/whatis/>
(I like the explanations there!)
<http://www.dinosauria.com/dml/clado/carnosauria.html>
I see Allosaurus, Sinraptor and Carcharodontosaurus are representatives
of the three major sub-clades of Allosauroidea, which is the one major
sub-clade of Carnosauria, which is top-of-tree on that Web page.
To go up the tree before Carnosauria:
<http://palaeos.com/Vertebrates/Units/340Theropoda/340.200.html>
Theropoda
`--Tetanurae
|--Torvosauroidea
`--Avetheropoda
|--Carnosauria
`--Coelurosauria
I see Carnosauria is sister-clade to Coelurosauria.
So in summary, clade defined by naming one Genus (Passer = the sparrow)
within modern birds deep down inside the Coelurosauria clade, and three
major branches of the sister clade Carnosauria named as excluded,
thereby in effect redundantly excluding that sister clade. Unless some
of that taxonomy is totally wrong, it looks like a good include/exclude
definition of the clade, forcing it to be directly under the LCA of all
four genera even if some new genera are discovered that branch off very
close to the top of Coelurosauria.
So I see two purposes of such cladistic definitions.
(1) If you have some taxa you want to encompass, in the narrowest way,
you pick two of them as far apart as possible and cite the clade
started by the LCA of them. The problem would be if some intended
member had branched off slightly earlier than that defined LCA hence
excluded by mistake. Neornithes (LCA of all modern birds), and
Coelurosauria (LCA of modern birds and Tyrannosaurus) are examples of
such. If Neornithes is defined as LCA of {Rheiformes + Phasianidae} and
all descendents thereof, you might someday discover some modern bird
whose ancestors branched off just before that point, oops.
(2) If you want to include *everything* from some really high common
ancestor down to the clade you know about, excluding the LCA itself but
including everything that started down that branch toward your clade of
interest, even if it diverged off it quite early, then you define your
clade as *largest* clade including one representative genus of the
branch of interest and excluding one or more representative genera of
the other branch you don't want. Coelurosauria (see above) is defined
that way, so it's guaranteed to forever be a sister clade to
Carnosauria (assuming the latter is defined oppositely to
Coelurosauria). Or is it? What if someday we find a third branch from
Avetheropoda not within either Carnosauria or Coelurosauria,
and upon resolution we discover something like this:
+--Carnosauria
--+
| +--NewBranch
+--+
+--Coelurosauria
Then suddenly they are no longer sister taxa? Wrong, by definition if
by "Coelurosauria" in that diagram we mean all old members of
Coelurosauria, then the definition of Coelurosauria now refers to the
common branch point above NewBranch and old-Coelurosauria, so it's
really like this:
+--Carnosauria
--+
| +--NewBranch
+-CoelurosauriaPerDefinition
+--old-Coelurosauria
So upon full (binary) resolution of the node which is LCA of
Carnosauria and Coelurosauria, we are guaranteed that the two are
forever sister taxa. Is my analysis now correct?
So combining the in-out clade definitions for two sister taxa, and the
in+in definition for their parent taxon, we're guaranteed that one
branching is fully named (parent and two children), with no way the
naming will ever turn out illogical, right?
.
Augray
Actually, the pterosaur patagium had fibers embedded in it to give it
stiffness in the front-back direction, and muscle fibers at right
angles to directly control the tension, allowing them to change the
patagium's shape "on the fly", so to speak. It also might have been
partially pneumaticized.
>[snip]
>
>>>>That's too bad. I have to admit that I prefer Aves when referring to
>>>>the crown clade, as it matches Linnaeus's original intent.
>>>
>>>Linnaeus defined the terminology for currently-living organisms only.
>>>Unfortunately when different people tried to exted the term 'Aves' to
>>>fossils, they IMO assigned the term prematurely, before they knew
>>>enough about the fossils to know where the right place should be.
>>
>>
>> They felt that they knew what the right place should be.
>
>The better question is whether there is any right place, objectively
>speaking.
Clades exist, so it doesn't seem unreasonable to name them.
Determining where one clade is in relation to another is the dicey
part.
>[snip]
John Harshman
[snip]
> What if someday we find a third branch from
> Avetheropoda not within either Carnosauria or Coelurosauria,
> and upon resolution we discover something like this:
>
> +--Carnosauria
> --+
> | +--NewBranch
> +--+
> +--Coelurosauria
>
> Then suddenly they are no longer sister taxa? Wrong, by definition if
> by "Coelurosauria" in that diagram we mean all old members of
> Coelurosauria, then the definition of Coelurosauria now refers to the
> common branch point above NewBranch and old-Coelurosauria, so it's
> really like this:
>
> +--Carnosauria
> --+
> | +--NewBranch
> +-CoelurosauriaPerDefinition
> +--old-Coelurosauria
>
> So upon full (binary) resolution of the node which is LCA of
> Carnosauria and Coelurosauria, we are guaranteed that the two are
> forever sister taxa. Is my analysis now correct?
Yes, though it was such a strange way to say it that I have to rephrase
for clarity. Any new species we find to be closer (are we clear now on
what "closer" means here?) to Passer than to Carnosauria belongs to
Coelurosauria by definition, even if it's basal to all previously known
coelurosaurs. That's the entire point of a stem-based clade definition.
> So combining the in-out clade definitions for two sister taxa,
The term is "stem-based definition".
> and the
> in+in definition for their parent taxon,
The term is "node-based definition".
> we're guaranteed that one
> branching is fully named (parent and two children), with no way the
> naming will ever turn out illogical, right?
The term is "node-stem triplet". Don't know what you mean by
"illogical". If you mean that a node-stem triplet using the same two
reference taxa will always define a node-stem triplet, then yes. This is
not guaranteed if we use different reference taxa, or in fact more than
two reference taxa, as the Coelurosauria definition does.
an...@sci.sci
> > Those organisms descended from the last common ancestor of all
> > living members of a clade (Runnegar 1992).
> > The last common ancestor of a living clade plus all of its
> > descendants
> > (Budd 2001, p. 333).
> > But that's exactly the same as the clade itself:
> Yes and no, it depends on your defination of "the". ;-)
> As I read it, and it took a while to twist my mind around
> the defination, a crown group is a clade but not all clades
> are crown groups.
> So depending on which clade you start with you may have to
> prune it a bit to get a crown group.
> > One individual
> > (species, in case of eukaryotes that reproduce via meiosis), and
> > all descendents of that individual (species).
> But only if the next branching point leads to living decendents
> on both branches, otherwise that individual is not the LCA of its
> _living_ descendents.
Aha, so the definition "crown group" is dependent on time. So long as
at least one individual is still alive along both branches from a
particular LCA, the clade from that LCA is a "crown group", but as soon
as the last invidual dies without leaving even one descendent, that
clade ceases to be a "crown group" (of itself), and instead a smaller
clade from some more recent LCA along the still-surviving branch
becomes the "crown group" of that same original clade. Given the high
rate of extinctions lately (100 times background rate), this seems a
somewhat unstable definition. Currently there are surviving gorillas,
although Ebola wiped out 80% of one tribe of them a few years ago. The
clade of the LCA of gorillas and chimp+human currently forms a "crown
group", but if Ebola or anything wipes out the last of the gorillas,
suddenly that "crown group" shrinks to the clade of the LCA of chimps
and humans. I can see the advantage to defining "crown groups" when a
population is stable, or at least both branches from the LCA are strong
enough to survive in the forseeable future, such as living birds with
crown group Neornithes which branches to Paleognathae and Galloanserae,
neither of which is likely to go extinct any time soon. But in cases
where one of the two branches is just barely hanging on, such as
monotremes as one branch of mammals, or gorillas as one branch of
African apes, I have misgivings about use of the idea.
But thanks for correcting my mis-reading of the definitions.
> If you want to include Neanderthals in a crown group then you
> also have to include the Chimps since they are our closest living
> relatives.
And indeed it's very unlikely that the chimp+human LCA is the "right"
place to start a hominid clade in some study or discussion, so
regardless of whether you *ever* use crown groups, you definitely don't
want to restrict yourself to crown groups as the only kinds of clades
you talk about or use as the focus of study or discussion. Even more
drastic, if you want to discuss the evolution from semi-birds similar
to Archaeopteryx to modern birds, the smallest crown group covering
your topic would also include crocodiles, which is very much overkill.
(And if global warming kills off all the crocodiles...)
One big advantage I can think of is that for any crown group we have
living descendents along both branches so we might be able to
reconstruct the genome by merging the genomes via both branches. But
for LCAs which lived very long ago and suffered massive extinctions
along most branches, such as the crocodile+bird LCA (clade Archosauria
which splits into branches Crurotarsi and Ornithodira), although it
might be possible to reconstruct the genome of the LCA of two tiny
crown groups (Neornithes and Crocodylia), it would be much more
difficult to work all the way back up from those tiny groups' genomes
to their LCA's genome. Here are the cladograms I've found, with some
annotations I've collected from various online sources:
REPTILOMORPHA (These higher clades here merely for context)
|--LEPIDOSAUROMORPHA (lizards + snakes)
`--Archosauromorpha (unable to find any description)
|--Rhynchosauria (Triassic only)
`--+--Prolacertiformes (Late Permian - Jurassic)
`--Archosauriformes (unable to find any clear description/definition)
|--Euparkeriidae (unable to find any description)
`--Archosauria **crown*group** (LCA of birds and crocodiles)
|--Ornithodira
| |--Scleromochlus
| `--+--PTEROSAURIA
| `--DINOSAUROMORPHA
`--Crurotarsi (stem: Crocodylus, not Passer, 245 Mya - current)
|--Phytosauridae (node: Angistorhinus + Mystriosuchus, 228 Mya - 204 Mya))
`--Rauisuchiformes (no definition, no such taxon any more)
|--Rauisuchia (no definition, no such taxon any more)
| |--Ornithosuchidae (stem: Ornithosuchus, not Rutiodon)
| `--+--Prestosuchidae
| `--Rauisuchidae
`--Suchia (node def: Aetosaurus + Rauisuchus 245 Mya - current)
|--Aetosauridae (no definition, may include: Aetosaurus + Stegomus)
`--CROCODYLOMORPHA
Crocodylomorpha (228 Mya - current)
|--Sphenosuchia (stem contains: Terrestrisuchus + Sphenosuchus, extinct 197 Mya)
`--Crocodyliformes (node def: Protosuchus + Crocodylus, 228 Ma - current)
|--Protosuchia (stem contains: Protosuchus richardsoni, extinct 136 Mya)
`--Mesoeucrocodylia (stem def, 151 Mya - current)
|--Thalattosuchia (stem contains: Geosaurus giganteus, extinct 136 Mya)
`--Metasuchia (node def: Notosuchus + Crocodylus, 156 Mya - current)
|--Notosuchidae (stem contains: Nototsuchus terrestris, extinct 71 Mya)
`--Neosuchia (stem def, 151 Ma - current)
|--[Dyrosauridae] (no cladistic definition, extinct unknown time)
`--Crocodylia **crown*group** (node def: Gavialis + Alligator, 84 Ma - current)
|--Gavialidae (one surviving species: Gavialis gangeticus)
`--+--Alligatoridae
`--Crocodylidae
|--Crocodylinae
`--Tomistominae
<http://www.dinoruss.org/de_4/5c5d1f6.htm>
both Protosuchia and Neosuchia are now recognised as paraphyletic.
which apparently implies the above cladogram is garbage?
<http://www.taxonsearch.org/dev/taxon_edit.php?Action=View&tax_id=370>
Taxonomic Content (*) stem (entirely extinct)
( ) crown (extant-bounded)
( ) mixed (extant/extinct-bounded)
incorrectly claims that Crocodylomorpha is entirely extinct!!
IMO "mixed" should have been selected instead.
Earliest Record Carnian (228 Ma)
Latest Record Recent
claims it's recent, contradicting the mis-info earlier
Same problem here too:
<http://www.taxonsearch.org/dev/taxon_edit.php?Action=View&tax_id=369>
<http://www.taxonsearch.org/dev/taxon_edit.php?Action=View&tax_id=373>
<http://www.taxonsearch.org/dev/taxon_edit.php?Action=View&tax_id=374>
Is/was there a consistent bug in their software at the time these Web
pages were built, or am I mis-reading the phrase "entirely extinct"??
<http://academic.emporia.edu/mooredwi/nathist/chap1.htm>
nice explanation of purpose and limits of cladograms
.
an...@sci.sci
> it's a special form of node-based definition, in which all the reference
> taxa are extant.
You were the second to explain my mis-reading. Thanks.
> Since we have a better chance of knowing the
> relationships among extant taxa than among fossils (much more data to
> work with for each taxon), crown groups are expected to be more stable
> than other groups; at least their extant backbones are.
Except for one problem: Extinction during our modern age!
Consider any endangered species E, and its nearest relatives N (as
determined by most recent common ancestor).
How the clade defined by {E + N} is, at the moment, a crown group.
Now suppose E goes extinct. Now that clade is no longer a grown group.
One species goes extinct and the definition is pulled like a rug out
from under you. Not very stable IMO. The original clade hasn't changed,
but the crown group of it has suddenly shrunk.
The only solution seems to be an "epoch" for the definition, similar to
the epoch such as 1950 or 2000 used to define coordinates (right
ascension and declination) for astronomical objects. At this time, I
think regressing back to 2000 would be the best choice, since that's
roughly the time when genome sequencing got going strong, so any
species that went extinct since 2000.01.01:00:00 GMT is likely to have
some preserved specimens we could later sequence, so pretending like it
hasn't yet gone extinct isn't a method-breaker. For any crown group
which ceased to exist prior to that epoch, but where we have
well-preserved specimens, we'd make an explicit exception wherby we
note the extinction date in the node-based clade definition, saying for
example it was a pre-1974 crown group if one branch went extinct during
1974.
.
John Harshman
>>Note "all *living* members" of a clade. To put it more simply,
>>it's a special form of node-based definition, in which all the reference
>>taxa are extant.
>
>
> You were the second to explain my mis-reading. Thanks.
>
>
>>Since we have a better chance of knowing the
>>relationships among extant taxa than among fossils (much more data to
>>work with for each taxon), crown groups are expected to be more stable
>>than other groups; at least their extant backbones are.
>
>
> Except for one problem: Extinction during our modern age!
That would be the case if we were blindly following the definition and
were indeed forced to alter it with every extinction. But nobody
actually proposes that. Credit scientists with a little common sense here.
[snip]
an...@sci.sci
> >>one additional constraint that the LCA of modern birds and
> >>Archaeopteryx, which is very close to Archaeopteryx in the evolutionary
> >>tree, should score exactly halfway between Coelurosauria and
> >>Neornithes, per some weighting of the various characters that
> >>distinguish those two points in the tree.
> I repeat: why?
Why not? If there are 50 different characters that distinguish between
ancient dinosaurs and modern birds, thereby giving 49 degrees of
freedom in weighting those various characters to achieve a numeric
scale from ancient-dinosaur-like to modern-bird-like, why won't you let
me place the most well-known transitional fossil along that sequence at
the midpoint, therby reducing the degrees of freedom to 48, and then
see where the rest of the transitional fossils lie per various allowed
weightings? If you were to do it differently, where would Archaeopteryx
sit along the numeric scale from ancient-dinosaur-like to
modern-bird-like, and what other well-known fossil would sit near the
midpoint?
Here's a simple metric I'd like to propose for evaluating how well
we've filled the "gap" with transitional fossils: We compute the
position of each transitional fossil along that scale, sort into
ascending sequence, and subtract adjacent values to get a list of gap
lengths. Then take square of each gap length and add them all together
to get a measure of how gappy our sequence remains.
For example, prior to the discovery of Archaeopteryx, we had:
Coelurosauria...............(1).............................Neornithes
Badness = 1**2 = 1 (100% bad)
After the discovery of Archaeopteryx, we had:
Coelurosauria......(0.5)... Archaeopteryx......(0.5)......Neornithes
Badness = 0.5**2 + 0.5**2 = 0.25 + 0.25 = 0.5 (50% bad)
Now suppose we get additional transitional fossils A B C D like this:
Coelurosauria(0.1)A(0.2)B(0.2)Archaeopteryx(0.2)C(0.1)D(0.2)Neornithes
Badness = 0.1**2 + 0.2**2 + 0.2**2 + 0.2**2 + 0.1**2 + 0.2**2
= 0.01 + 0.04 + 0.04 + 0.04 + 0.01 + 0.04
= 0.18 (18% bad)
So the badness of the gaps gets less and less as we find new fossils
that nicely split large gaps, but nothing hardly changes when we find
new fossils filling already-tiny gaps or when we find yet another
specimen of an already-known transitional species.
Note that if we find species that are far from the main branch, they
don't help the situation at all, since their distances from previous
and next main-branch species may be as large as the old gap they are
trying to split, so it's best to omit them unless we can deduce the
characters of the place where their branch split from the main line and
use that set of characters instead of the actual fossil characters. But
that would probably be cheating, and rightly contested by
anti-evolutionists as circular reasoning, so IMO we should just omit
any way-off-main-branch species we find from our list of alleged
aDinosaur-mBird transitional fossils. Reserve such species for *other*
arguments in favor of evolution by natural means, to show how often the
road forks and one fork turns out to be a dead end but there was no
Intelligent Designer who foresaw that fate and avoided making the
dead-end branch in the first place.
Note that if Gould's theory of Punctuated Eqlibrium is correct,
including expected lack of any fossils of founders of new species
during the population bottleneck when most of the drift and major
adaption occurred, then even with near-100% recovery of fossils the
"badness" figure of the fossil record won't approach zero, it'll stay
above whatever number indicates the actual P.E. history.
(Re convergent evolution of bats and pterodactyls:)
> > I'm not sure that I'd necessarily call it convergent. Aside from the
> > patagium (and even that's debatable), they don't seem to have had a
> > lot in common.
> Oh, come on. Wing made of skin stretched over a bone framework?
> Quadrupedal when on the ground? It's not that bad.
Ah, I was thinking of skin-between-bones only, thanks for the second
thing they have in common. But actually the latter is an ancestral
character, not an evolved character, so it doesn't really count as
strongly. In fact I would guess that quadrupedal travel wasn't ever
lost along the path from amphibians to bats or to pterodactyls, so it
isn't even convergent reversion. But skin-over-bones seems to be not
only convergent evolution but convergent novel invention. Do we have
any well-preserved pterodactyls that show precisely how the skin was
wrapped over the bones? Did both bats and pterodactyls wrap skin
between finger bones? Did they both use the same precise method,
basically web-feet as applied to fingers instead? Or did either invent
something different as the basic wrapping/stretching method? Are any of
the actual genes homologous to web-feet (in amphibians such as frogs)
genes, perhaps adapted by moving those genes from a foot-hox section to
a hand-hox section of the genome? Or did any amphibian already have
webbed-hands for swimming and both bats and pterodactyls reverted to
that and then re-adapted them for flying instead of swimming? Did any
close ancestors to either group actually use them for swimming and then
adapt them directly for flying instead? Since apoptosis is the cause of
gaps between fingers, could a simple *loss* of one case of apoptosis
cause webbed fingers, which would be likely to recur in many many
branches of tetrapods, of which most die out but just those two groups
have survived because the loss happened in the right place at the right
time for the loss to be selected for?
> >>Linnaeus defined the terminology for currently-living organisms only.
> >>Unfortunately when different people tried to exted the term 'Aves' to
> >>fossils, they IMO assigned the term prematurely, before they knew
> >>enough about the fossils to know where the right place should be.
> > They felt that they knew what the right place should be.
> The better question is whether there is any right place, objectively
> speaking.
Indeed. If there were an Intelligent Designer, who assigned brand names
to His products (such as Taurus and Fairlane for models of Ford Motor
Company's passenger vehicles), then the "right" assignment of the name
would be however the Designer assigned the product names. If the
Designer called Archaeopteryx a bird then so be it, and if not then so
be it not. (I forbid anybody to call me the Shakespeare, or the Hamlet,
of cladistics!!) But in fact, there isn't (ID), so there isn't (RA), IMO.
IMO "Aves" and "birds" has no place in cladistics.
I'm content with node-based Neornithes (LCA of living birds) and
stem-based Maniraptora = {Neornithes, not Tyrannosaurus}, with neither
of them nor any other clade equated to the traditional (pre-Darwin)
taxon "Aves" or "birds".
.
John Harshman
>>>>Then I have
>>>>one additional constraint that the LCA of modern birds and
>>>>Archaeopteryx, which is very close to Archaeopteryx in the evolutionary
>>>>tree, should score exactly halfway between Coelurosauria and
>>>>Neornithes, per some weighting of the various characters that
>>>>distinguish those two points in the tree.
>>
>>I repeat: why?
>
>
> Why not? If there are 50 different characters that distinguish between
> ancient dinosaurs and modern birds, thereby giving 49 degrees of
> freedom in weighting those various characters to achieve a numeric
> scale from ancient-dinosaur-like to modern-bird-like, why won't you let
> me place the most well-known transitional fossil along that sequence at
> the midpoint, therby reducing the degrees of freedom to 48, and then
> see where the rest of the transitional fossils lie per various allowed
> weightings? If you were to do it differently, where would Archaeopteryx
> sit along the numeric scale from ancient-dinosaur-like to
> modern-bird-like, and what other well-known fossil would sit near the
> midpoint?
Again, why? Do you learn anything from this exercise? It seems wholly
arbitrary.
This doesn't do anything to tell me why you would want to do it.
[snip]
> Do we have
> any well-preserved pterodactyls that show precisely how the skin was
> wrapped over the bones?
If you mean what the shape of the wing was, yes.
> Did both bats and pterodactyls wrap skin
> between finger bones? Did they both use the same precise method,
> basically web-feet as applied to fingers instead? Or did either invent
> something different as the basic wrapping/stretching method?
If you mean how the wing was supported, yes. Pterodactyls supported the
wing entirely with a single finger, which in consequence was enormous.
> Are any of
> the actual genes homologous to web-feet (in amphibians such as frogs)
> genes, perhaps adapted by moving those genes from a foot-hox section to
> a hand-hox section of the genome?
How would anyone know this, pterodactyl genomes being entirely
unavailable. At any rate, I doubt anyone knows what the genetic basis of
foot-webs (or hand-webs) is in any of the various taxa that have them.
We probably know the developmental basis of some, though obviously not
pterodactyls. And bat development is almost certainly little studied.
Your idea of moving genes is very odd, though.
> Or did any amphibian already have
> webbed-hands for swimming and both bats and pterodactyls reverted to
> that and then re-adapted them for flying instead of swimming? Did any
> close ancestors to either group actually use them for swimming and then
> adapt them directly for flying instead?
No.
> Since apoptosis is the cause of
> gaps between fingers, could a simple *loss* of one case of apoptosis
> cause webbed fingers, which would be likely to recur in many many
> branches of tetrapods, of which most die out but just those two groups
> have survived because the loss happened in the right place at the right
> time for the loss to be selected for?
I doubt it. This apoptosis occurs during the development of the fingers
themselves, and any prevention of that would result in a big mass of
tissue, not just webs between discrete digits.
>>>>Linnaeus defined the terminology for currently-living organisms only.
>>>>Unfortunately when different people tried to exted the term 'Aves' to
>>>>fossils, they IMO assigned the term prematurely, before they knew
>>>>enough about the fossils to know where the right place should be.
>>>
>>>They felt that they knew what the right place should be.
>>
>>The better question is whether there is any right place, objectively
>>speaking.
>
> Indeed. If there were an Intelligent Designer, who assigned brand names
> to His products (such as Taurus and Fairlane for models of Ford Motor
> Company's passenger vehicles), then the "right" assignment of the name
> would be however the Designer assigned the product names. If the
> Designer called Archaeopteryx a bird then so be it, and if not then so
> be it not. (I forbid anybody to call me the Shakespeare, or the Hamlet,
> of cladistics!!)
Little danger of that.
> But in fact, there isn't (ID), so there isn't (RA), IMO.
> IMO "Aves" and "birds" has no place in cladistics.
Why? They're just names. You can give any name to any node you like. The
only problem with those names is that the first has been given to too
many mutually contradictory nodes, and the second isn't a formal
scientific name at all.
> I'm content with node-based Neornithes (LCA of living birds) and
> stem-based Maniraptora = {Neornithes, not Tyrannosaurus}, with neither
> of them nor any other clade equated to the traditional (pre-Darwin)
> taxon "Aves" or "birds".
> .
Why limit yourself? There are many more named nodes, including Avialae,
Maniraptoriformes, Tetanurae, Coelurosauria, et multa cetera.
an...@sci.sci
> > boolean metric where two species either share a common ancestor
> > (distance zero) or don't (distance 100%).
> > You need to specify what metric you really mean when you say "closer".
> > For example, here's a hypothetical cladogram:
> >
> > 3my
> > +---Foo
> > ---+
> > | 4my+--Bar
> > +----+2my
> > |
> > +----------------------------------------------------Blortch
> > 52my
> > Bar is closer to:
> [snip]
> > Blortch in terms of absolute recency of LCA
> > (LCA with Blortch is 4my more recent than LCA with Foo)
> That was the intended meaning.
Thanks for letting me know. Note: That isn't the most natural
definition. Evolutionary distance is defined between ancestor and
descendent as total time (or DNA base changes) between those two
points, and evolutionary distance between two descendents is naturally
defined as total of distance along paths from LCA to each. "Closer" has
the natural meaning "less evolutionary distance" per that definition of
evolutionary distance, thus:
> > Bar is closer to:
> > Foo in terms of total evolutionary distance along both branches from LCA
> > (9 from Foo compared to 54 my from Blortch)
Because that's a more natural definition of "closer", IMO the word
"closer" shouldn't be used at all in (stem-based) cladistic
definitions. Instead the LCA definition should be spelled out as part
of the stem-based definition:
- All taxa with more recent common ancestor with Rheiformes than with Tyrannosaurus.
or better the "largest clade" text should be stated explicitly as I've
indeed seen in some online pages:
- The largest clade containing Rheiformes but not Tyrannosaurus.
> > I think a cladistic definition, either the smallest clade containing
> > two taxa, or the largest clade containing one taxon but not another, is
> > less ambiguous than any "closer than" definition.
> I don't know why you guys are arguing here, since you are actually both
> using the same definition. I think you must be trying hard to
> misunderstand each other.
But the way he expressed the definition sounded different from the
standard stem-based definition. It sounded like it used the natural
definition of "closer" (total evolutionary distance down two branches
from LCA), when that's not what he intended. I was merely noting that
the natural definition, which seemed pretty much worthless, as well as
several unnatural definitions, one of which was the correct one, were
all possible interpretations of his ambiguous usage. I think I did a
good job of listing all semi-reasonble definitions for "closer", so
that you could pick out the correct one and I could also cite the
natural one, so we quickly resolved the problem in nomenclature.
Try getting a Creationist or IDiot to list possible meanings of a word,
one of which is correct, to allow quick resolution of debate.
(We can't get them to even agree what ID means!! See my various ideas
what it *could* mean, including Omphalos, nested committees of angels
or space aliens with strict-sequential or non-sequential production,
natural evolution with intelligent tinkering or deliberate breeding,
Deism to create natural evolution, etc.)
I won't say "count your blessings" regarding debate with me, since
there's no such thing as a blessing, but I might say "count anon1's
intelligence and reasonableness and eagerness to resolve disputes".
(I know what you must be thinking: "Gag me with a spoon")
> I bet Augray was using "modern birds" as a synonym for "Neornithes".
I'd rather not place wagers on what somebody might have meant.
It's too easy for con+shill to fix the game.
> > So the
> > statement that Archaeopteryx is not a modern bird is a worthless
> > statement. Anything that went extinct as recently as three million
> > years ago is not a modern bird in the sense of not ever co-existing
> > with modern humans, even if it is within the clade of modern birds.
> This is a silly argument. You seem to be arguing for the sake of
> arguing.
I interpret "modern" to mean "within the past 50 years" in sociology,
or "within the past half million years" in evolution, using the latter
in this thread. Suppose we replace "modern" with "extant", and somebody
asks whether Archaeopteryx is an extant bird today. Would you consider
that a deep question, or a triviality not worth even asking?
> You might as well, in that case, ask whether turtles
> are modern birds, or whether lizards are, or foxes, or frogs.
Each is a meaningful question, since all of them are modern species, so
it's a real question whether they ar considered birds. For example,
some people don't accept that ostriches are birds, or that whales are
mammals. Some experts place turtles within the dinosaur+bird clade,
although majority concensus says they are quite separate. Some experts
expand the bird clade to include all dinosaurs, not just theropods.
Showing that turtles and foxes are disjoint clades, that their LCA is
way back before the first of either evolved, would be a good exercise
for beginning students.
> If you guys would just try to understand what the other one is
> saying, rather than look for things to pounce on, everything would go
> better.
Sometimes indeed when I see any ambiguous usage I write a one-line
reply asking what the person means by that usage, and refuse to respond
to the substance of the article until and unless that usage is defined.
I could do that *all* the time if you wish.
> > - Each trait is quantitative, i.e. gives a numerical value instead of
> > just a natural-language description.
> Why would you want to do this?
So that you can produce numerical evidence that the gaps that formerly
existed are now being well-filled with transitional fossils. If all you
have are words, no numbers, then every pair of different descriptions
are infinitely far from each other, and no gaps are *ever* filled even
partly.
> Systematists tend to like qualitative characters that can be scored
> as discrete states. Easier to analyze.
Easier to map from specimen to description, but impossible to map from
set of descriptions to cladogram. If species A has traits A1 A2 and A3,
and species B has traits B1 B2 and B3, and species C has traits C1 C2
and C3, and species D has traits D1 D2 and D3, which unrooted tree is
correct? Is A+B/C+D correct? Or is A+C/B+D correct? Or is A+D/B+C correct?
You tell me.
> If you wanted quantitative traits you would end up mostly with
> descriptions of Tyrannosaurus' large size.
Nope, that's a blatantly dishonest strawman. You should be ashamed of
what you said there. In fact, size would be just one character, which
would probably have low weight compared to density of feathers,
adaption of feathers for flying, hollowness of bones for respiration,
fusion of vertibrates, etc. etc. etc.
.
an...@sci.sci
> >boolean metric where two species either share a common ancestor
> >(distance zero) or don't (distance 100%).
> Are you saying that common descent isn't a fact?
No, please try to read more carefully. Two species are either related
by common descent, or not. If two are related, then all that the fact
of common descent per se tells you is that they are both related, not
that either is closer than another.
> What makes you think that time is a useful metric for evolutionary
> distance?
For two species, one of which is the ancestor of the other, it gives
you a rough estimate how dissimilar they are. If the time span is 5
years, they are not much different at all, whereas if the time span is
500 million years then they are grossly different.
> What if you don't know the date of divergence?
If you haven't the foggiest idea whether two species diverged last year
or 500 million years ago, you have more trouble than I can help you
with. If you know they diverged somewhere between 380 and 410 million
years ago, that's a pretty accurate date if you're comparing that date
with something that occurred between 64 and 66 million years ago.
There's no such thing as knowing "the date" exactly to the day for
something that happened more than 5000 years ago, except for
predictable astronomical events such as eclipses. So your question
isn't meaningful until you explain what you mean by "know the date".
> >I think a cladistic definition, either the smallest clade containing
> >two taxa, or the largest clade containing one taxon but not another, is
> >less ambiguous than any "closer than" definition.
> What you've just described are respectively called node definitions
> and stem definitions.
Yes, that's correct. Tell me something I don't already know.
> >At this point I'm pretty well decided that the scale should run from
> >Coelurosauria to Neornithes, where Coelurosauria is the LCA of
> >Tyranosaurus and Neornithes. Maniraptoriformes is not an acceptable
> >taxon because it's not well defined,
> What makes you think that? It's very well defined. It's the last
> common ancestor of _Ornithomimus_ and Neornithes and all it's
> descendents.
You misspelled the word "its". Please correct and re-submit.
> Was the Passenger Pigeon a modern bird?
The reason for glorfying "crown groups" was that we have the living
species available to study in great detail, including studying
development of the embryo, and sequencing the DNA or any affordable
useful part thereof, as well as conducting new tests we didn't even
think of last year, such as measuring brain waves by scanning NMR, or
submitting blood at different times or under different stress to
micro-asseys of expression for millions of genes. Can we do all those
such things with a Passenger Pigeon? Please answer your own question.
(Regarding your use of the word "modern":)
> What makes you think that I was using it in the "in the sense of not
> ever co-existing with modern humans"?
That was just a guess, based on common usage in paleontology and evolution.
You didn't define the word before using it, so I could only guess its meaning.
> But "living" isn't the criteria as to whether an animal should be
> included in Neornithes.
But it *is* the criteria as to whether a given species should be forced
into that clade as part of its definition. If the last of the rattites
had gone extinct 200 years ago, we'd be justified in defining a clade
that excluded them, and using that smaller clade instead of our current
Neornithes as the crown group of living birds.
> >Do you understand the difference between the questions whether
> >Archaeopteryx was within the smallest clade containing modern birds
> >(good question, no, not in that clade), vs. whether Archaeopteryx *is*
> >a modern bird (stupid question)?
> Let me play Creationist's Advocate for a moment, and say that I don't.
> Then what?
The *is* question is a matter of religion, of God-given names as
revealed by sacred writings or recent revelation. The *clade* question
is a matter of natural history, as revealed by fossils. If you find a
Biblical passage stating clearly that Archaeopteryx *is* (or "was",
since it's now extinct and might have already been extinct before our
ancestors received that Word of God) a "bird" (as translated from
Hebrew or Greek etc. in some non-ambiguous way), I'd be glad to accept
that. But if you tell me that *you* personally got such a revelation,
I'd probably choose to call you a liar unless I happened to get that
same revelation myself.
(Regarding my metaphor of a xerox/kodak copy shop, descent with modification:)
> A creationist would say that since there were no witnesses in the
> past, coming to a conclusion of common ancestry would be premature.
There are no witnesses even from yesterday. All we have today are
imprints in the minds of people alive today, which we believe are
memories of what they experienced yesterday, but there's no way to
prove that. Fossils are older imprints, and natural rather than mental,
but it's the same basic idea, and in fact fossils are more reliable
than eyewitness testimony.
> >Well when somebody puts up a personal Web page or posts to a newsgroup,
> >yes I understand it's all half crap. But when there's an online service
> >that gives cladograms in nice format, I really assumed those were based
> >on some science, not gross guesswork.
> Creationist websites have nice formats too. Why not use the info
> there?
I believe a cladogram on a Creationist Web site would be an oxymoron.
> >> Metornithes is probably synonymous with Maniraptoriformes.
> >Metornithes seems to be sister taxa to Archaeopteryx.
> Based on what?
I don't remember. I'll try to look it up now ... found:
<http://www.dinosauria.com/dml/clado/avialae.html>
+ Avialae
o Rahonavis ostromi
o Unenlagia comahuensis
o Aves
# Archaeopteryx bavarica
# Archaeopteryx lithographica*
# Metornithes
@ Confuciusornis chuonzhous
@ Confuciusornis sanctus*
@ Confuciusornis suniae
@ Jibeinia luanhera [i. s.]
@ Alvarezsauria
@ ---
- ?Noguerornis
- Ornithothoraces
I'll check another source ... found:
<http://www.taxonsearch.org/dev/taxon_edit.php?Action=View&tax_id=203>
The initial definition of Metornithes in Perle et al. (1993) as "a
group within Aviale not including Archaeopteryx" is best interpreted
as a stem-based definition, although not of the most typical form. It
clearly referes to all members of Avialae except the basal member
Archaeopteryx. ...
Per that definition, it seems to me that Archaeopteryx and Metornithes
are sister clades with respect to each other.
(snipped commentary about several attempts to re-define the same name)
Metornithes is here considered inactive.
That sounds like a prudent decision, given the conflicting definitions.
Definition 2 The common ancestor of Mononykus and Neornithes plus all
its descendants.
Do you have a cladogram to show me where Metornithes satisfies that
second definition, so I can compare it with the cladogram I cited above
which is consistent with the original definition?
<http://www.taxonsearch.org/dev/taxon_edit.php?Action=View&tax_id=196>
Maniraptoriformes = LCA {Ornithomimus edmontonicus + Passer domesticus}
Like you said there, but given the difference between this definition
and the various conflicting definitions of Metornithes, it's rather
questionable whether Metornithes could reasonably be considered
synonymous with Maniraptoriformes.
I wish there were some way to browse the cladistic tree, like start
with some term and browse to next-ancestor or to each next-descendent.
Does anybody know how to write such a program? The browser itself
should be easy. It's collecting all the data from various Web sites and
organizing it into a database that would be the hard part.
Even better would be if I could enter more than one taxon and it'd
first tell me which aren't active taxa and then from the two or more
taxa which are active it'd connect them all and draw a skeleton
cladogram with indication how many between nodes are along each segment
shown. For example, if I put in Rheiformes and Tyrannosaurus, it might
show me:
Coelurosauria
+(5 intermediate nodes) Rheiformes
`(4 intermediate nodes) Tyrannosaurus
and then I'd have the option of showing all the intermediate nodes:
Coelurosauria
+Maniraptora--Avialae--Neornithes--Paleognathae--Ratites--Rheiformes
`Tyrannosauroidea--Tyrannosauridae--Tyrannosaurinae--Tyrannosaurini--Tyrannosaurus
or in more standard cladogram format:
Coelurosauria
+Maniraptora
| `Avialae
| `Neornithes
| `Paleognathae
| `Ratites
| `Rheiformes
`Tyrannosauroidea
`Tyrannosauridae
`Tyrannosaurinae
`Tyrannosaurini
`Tyrannosaurus
or showing intermediate nodes and sister branches too (exercise for reader).
> >> You could always check the primary literature.
> >You mean like those *Research*Reports* in _Science_ which are 99% inscrutable?
> Yes. Learning things can be fun.
Not having the Internet available at the same time I'm trying to read
one of those *Reports* makes it very difficult to understand most of
the jargon. By comparision, reading newsgroups is easy because I do
that when I'm online and can go to Google or an online dictionary or
thesaurus to look up terms whenever I don't understand some term
somebody used. Being frustrated by not understanding twenty jargon
terms per paragraph is no fun at all.
> >Then I guess maybe I should hire somebody to write that sofware.
> Go right ahead.
If I found somebody to write it and install it online, would you use it
too, or would I be the only person using it?
By the way, I see a problem with trying to make lots of fully-defined
triples while avoiding duplicate definitions along a non-branching
stem. A triple requires the parent taxon to be node-based and each
child-taxa to be stem-based. But if you want to link taxa like this:
Grampa
+Daddy
| +Billy
| `Sue
`Mommy
it's impossible for both Grampa = (Daddy + Mommy) and Daddy = (Billy + Sue)
to both be triples, because the first imples Daddy is stem-based while
the second implies Daddy is node-based. The only way to make both of
the desired triples is to duplicated Daddy as both stem-based and
node-based, like this:
Grampa
+DaddyStem -- defined as largest clade containing {Billy, but not Mommy}
| +DaddyNode -- defined as smallest clade containing {Billy + Sue}
| +Billy
| `Sue
`Mommy
If later a new fossil is found that branches off between DaddyStem and
DaddyNode, it fits in nicely (but breaks the adjacent triples), but in
the mean time having both Daddy taxa is rather ugly, especially if this
happens along *every* segment in the entire tree! I have a possible
solution: Use only stem-based definitions for every non-leaf node in
the tree, with one exception: crown groups. There would never be a
triple except at crown groups, but that's OK with me. But how to define
the very topmost node, the UCA? Answer (hack): UCA = largest clade
containing {Sue, but not inanimateobject}.
.
John Harshman
I don't think it matters.
[snip]
>>I bet Augray was using "modern birds" as a synonym for "Neornithes".
>
>
> I'd rather not place wagers on what somebody might have meant.
> It's too easy for con+shill to fix the game.
That's paranoia. It was entirely obvious what he meant, and worrying
about it is perverse.
>>>So the
>>>statement that Archaeopteryx is not a modern bird is a worthless
>>>statement. Anything that went extinct as recently as three million
>>>years ago is not a modern bird in the sense of not ever co-existing
>>>with modern humans, even if it is within the clade of modern birds.
>>
>>This is a silly argument. You seem to be arguing for the sake of
>>arguing.
>
> I interpret "modern" to mean "within the past 50 years" in sociology,
> or "within the past half million years" in evolution, using the latter
> in this thread. Suppose we replace "modern" with "extant", and somebody
> asks whether Archaeopteryx is an extant bird today. Would you consider
> that a deep question, or a triviality not worth even asking?
You are wasting bandwidth, and I really don't know why I'm even
responding to this.
>>You might as well, in that case, ask whether turtles
>>are modern birds, or whether lizards are, or foxes, or frogs.
>
> Each is a meaningful question, since all of them are modern species, so
> it's a real question whether they ar considered birds. For example,
> some people don't accept that ostriches are birds, or that whales are
> mammals.
Name one.
> Some experts place turtles within the dinosaur+bird clade,
> although majority concensus says they are quite separate.
Name one. (Some experts think turtles may be archosaurs, but that's not
quite a dinosaur.)
> Some experts
> expand the bird clade to include all dinosaurs, not just theropods.
Name one.
> Showing that turtles and foxes are disjoint clades, that their LCA is
> way back before the first of either evolved, would be a good exercise
> for beginning students.
>
>
>>If you guys would just try to understand what the other one is
>>saying, rather than look for things to pounce on, everything would go
>>better.
>
>
> Sometimes indeed when I see any ambiguous usage I write a one-line
> reply asking what the person means by that usage, and refuse to respond
> to the substance of the article until and unless that usage is defined.
> I could do that *all* the time if you wish.
I would prefer that you try to understand what the person is saying.
>>>- Each trait is quantitative, i.e. gives a numerical value instead of
>>>just a natural-language description.
>>
>>Why would you want to do this?
>
> So that you can produce numerical evidence that the gaps that formerly
> existed are now being well-filled with transitional fossils. If all you
> have are words, no numbers, then every pair of different descriptions
> are infinitely far from each other, and no gaps are *ever* filled even
> partly.
But this quantification that you are proposing is arbitrary. Why should
any creationist accept it as evidence superior to the simple presence of
intermediate states?
>>Systematists tend to like qualitative characters that can be scored
>>as discrete states. Easier to analyze.
>
>
> Easier to map from specimen to description, but impossible to map from
> set of descriptions to cladogram. If species A has traits A1 A2 and A3,
> and species B has traits B1 B2 and B3, and species C has traits C1 C2
> and C3, and species D has traits D1 D2 and D3, which unrooted tree is
> correct? Is A+B/C+D correct? Or is A+C/B+D correct? Or is A+D/B+C correct?
> You tell me.
If they are unordered characters, they provide no information about the
cladogram at all. If they are ordered characters, what they say would
depend on the ordering. But why do you present this strawman?
>>If you wanted quantitative traits you would end up mostly with
>>descriptions of Tyrannosaurus' large size.
>
> Nope, that's a blatantly dishonest strawman. You should be ashamed of
> what you said there. In fact, size would be just one character, which
> would probably have low weight compared to density of feathers,
> adaption of feathers for flying, hollowness of bones for respiration,
> fusion of vertibrates, etc. etc. etc.
Those all sound like qualitative characters to me, except for "density
of feathers", and I have no idea how you would measure that one. I am
not being dishonest. You just have no idea how this sort of thing works,
and I am trying to educate you just a little bit.
John Harshman
>>>Common descent is a theory, not a metric, except in the sense of a
>>>boolean metric where two species either share a common ancestor
>>>(distance zero) or don't (distance 100%).
>>
>>Are you saying that common descent isn't a fact?
>
>
> No, please try to read more carefully. Two species are either related
> by common descent, or not. If two are related, then all that the fact
> of common descent per se tells you is that they are both related, not
> that either is closer than another.
>
>
>>What makes you think that time is a useful metric for evolutionary
>>distance?
>
>
> For two species, one of which is the ancestor of the other, it gives
> you a rough estimate how dissimilar they are. If the time span is 5
> years, they are not much different at all, whereas if the time span is
> 500 million years then they are grossly different.
That would be true if there were an evolutionary clock. But there
doesn't seem to be, hence the large numbers of "living fossils" and the
large numbers of fast transitions. This idea works somewhat better in
molecular evolution than in morphological evolution. But of course we
can't use that with fossils.
[snip]
>>>I think a cladistic definition, either the smallest clade containing
>>>two taxa, or the largest clade containing one taxon but not another, is
>>>less ambiguous than any "closer than" definition.
>>
>>What you've just described are respectively called node definitions
>>and stem definitions.
>
>
> Yes, that's correct. Tell me something I don't already know.
That it's better to try to understand people than to try to
misunderstand them. You guys spent a whole series of post arguing over
two slightly different ways to say exactly the same thing.
>>>At this point I'm pretty well decided that the scale should run from
>>>Coelurosauria to Neornithes, where Coelurosauria is the LCA of
>>>Tyranosaurus and Neornithes. Maniraptoriformes is not an acceptable
>>>taxon because it's not well defined,
>>
>>What makes you think that? It's very well defined. It's the last
>>common ancestor of _Ornithomimus_ and Neornithes and all it's
>>descendents.
>
> You misspelled the word "its". Please correct and re-submit.
Pointless spelling flames also impede communication, if that is indeed
your purpose.
>>Was the Passenger Pigeon a modern bird?
>
>
> The reason for glorfying "crown groups" was that we have the living
> species available to study in great detail, including studying
> development of the embryo, and sequencing the DNA or any affordable
> useful part thereof, as well as conducting new tests we didn't even
> think of last year, such as measuring brain waves by scanning NMR, or
> submitting blood at different times or under different stress to
> micro-asseys of expression for millions of genes. Can we do all those
> such things with a Passenger Pigeon? Please answer your own question.
We can do some of them, including a bit of DNA sequencing.
> (Regarding your use of the word "modern":)
>
>>What makes you think that I was using it in the "in the sense of not
>>ever co-existing with modern humans"?
>
> That was just a guess, based on common usage in paleontology and evolution.
> You didn't define the word before using it, so I could only guess its meaning.
I knew what he meant. It was clear from context. Again, you should try
harder to understand what people are trying to say rather than seizing
on any possible ambiguity as a chance to argue, or whatever it is you're
trying to do here.
>>But "living" isn't the criteria as to whether an animal should be
>>included in Neornithes.
> But it *is* the criteria as to whether a given species should be forced
> into that clade as part of its definition.
Not true. Neornithes includes more extinct species (most of them
unknown) than extant ones. Only if one entire side of the basal
branching goes extinct would we have to think about changing the
definition of a crown group.
> If the last of the rattites
> had gone extinct 200 years ago, we'd be justified in defining a clade
> that excluded them, and using that smaller clade instead of our current
> Neornithes as the crown group of living birds.
Fortunately, this hasn't happened.
[snip]
> I wish there were some way to browse the cladistic tree, like start
> with some term and browse to next-ancestor or to each next-descendent.
This would require defining ancestors and descendants, which can't be
done. If instead you mean from more inclusive to less inclusive clades
(which is what I think you probably mean), there are web sites that do
that, for example the Tree of Life:
http://www.tolweb.org
[snip]
Augray
>> >Common descent is a theory, not a metric, except in the sense of a
>> >boolean metric where two species either share a common ancestor
>> >(distance zero) or don't (distance 100%).
>> Are you saying that common descent isn't a fact?
>
>No, please try to read more carefully. Two species are either related
>by common descent, or not.
Can you name two species that you believe are not related by common
descent?
>If two are related, then all that the fact
>of common descent per se tells you is that they are both related, not
>that either is closer than another.
But then, I wasn't talking about two species and/or clades, but at
least three; the species we're trying to categorize, the "closer to"
species/clade, and the "than to" species/clade. For example, the
definition of Paraves is all animals more closely related to
Neornithes than to Oviraptor. 1 = "all animals", 2 = Neornithes, and 3
= Oviraptor.
>> What makes you think that time is a useful metric for evolutionary
>> distance?
>
>For two species, one of which is the ancestor of the other, it gives
>you a rough estimate how dissimilar they are. If the time span is 5
>years, they are not much different at all, whereas if the time span is
>500 million years then they are grossly different.
>
>> What if you don't know the date of divergence?
>
>If you haven't the foggiest idea whether two species diverged last year
>or 500 million years ago, you have more trouble than I can help you
>with. If you know they diverged somewhere between 380 and 410 million
>years ago, that's a pretty accurate date if you're comparing that date
>with something that occurred between 64 and 66 million years ago.
How does one recognize divergence? For example, Archaeopteryx occurs
in the Late Jurassic, Caudipteryx is from the Early Cretaceous, and
the Pigeon is alive today. Of these three, which two are the closest
in the sense of evolutionary distance?
>There's no such thing as knowing "the date" exactly to the day for
>something that happened more than 5000 years ago, except for
>predictable astronomical events such as eclipses. So your question
>isn't meaningful until you explain what you mean by "know the date".
Yet you yourself state "that's a pretty accurate date if you're
comparing that date with something that occurred between 64 and 66
million years ago".
>> >I think a cladistic definition, either the smallest clade containing
>> >two taxa, or the largest clade containing one taxon but not another, is
>> >less ambiguous than any "closer than" definition.
>> What you've just described are respectively called node definitions
>> and stem definitions.
>
>Yes, that's correct. Tell me something I don't already know.
Then why didn't you use those terms? You said you preferred a
cladistic definition, but didn't use one that you were aware of.
>> >At this point I'm pretty well decided that the scale should run from
>> >Coelurosauria to Neornithes, where Coelurosauria is the LCA of
>> >Tyranosaurus and Neornithes. Maniraptoriformes is not an acceptable
>> >taxon because it's not well defined,
>> What makes you think that? It's very well defined. It's the last
>> common ancestor of _Ornithomimus_ and Neornithes and all it's
>> descendents.
>
>You misspelled the word "its". Please correct and re-submit.
>
>> >> >> >> >But with terms defined as I intended (near-LCA-theropods vs. modern
>> >> >> >> >birds), surely there is some character unique to modern birds?
>> >> >> >>
>> >> >> >> The problem is that Archaeopteryx isn't a modern bird.
>> >> >> >That sentence is meaningless in the modern context of evolution. It
>> >> >> >makes sense only in the old Platonic/Lamarckian philosophy of fixed
>> >> >> >ideal/metaphysical types which are only approximated by physical
>> >> >> >objects. (Or it's trivially true simply because Archaeopteryx didn't
>> >> >> >survive into modern times.)
>> >> >> Actually, "Archaeopteryx isn't a modern bird" makes a *lot* of sense,
>> >> >> in that it lacked many traits possessed by the birds alive today.
>> >> >We're in agreement.
>> >> Then why did you state that my sentence was meaningless?
>> >
>> >We're in agreement that Archaeopteryx lacked many traits possessed by
>> >the birds alive today. But modern birds are by definition those birds
>> >which are alive today, not ones that went extinct more than fifty
>> >million years ago, regardless of their characteristics.
>>
>> I'd debate that. Was the Passenger Pigeon a modern bird?
>
>The reason for glorfying "crown groups" was that we have the living
>species available to study in great detail, including studying
>development of the embryo, and sequencing the DNA or any affordable
>useful part thereof, as well as conducting new tests we didn't even
>think of last year, such as measuring brain waves by scanning NMR, or
>submitting blood at different times or under different stress to
>micro-asseys of expression for millions of genes.
Do you have a reference for that claim?
>Can we do all those
>such things with a Passenger Pigeon?
Why is it necessary from a cladistic point of view?
>Please answer your own question.
>
>(Regarding your use of the word "modern":)
Actually, you're the one who used it first in
news:cd205$4390e10d$c690c02a$41...@TSOFT.COM
>> What makes you think that I was using it in the "in the sense of not
>> ever co-existing with modern humans"?
>
>That was just a guess, based on common usage in paleontology and evolution.
>You didn't define the word before using it, so I could only guess its meaning.
You're the one who used it first. In
news:9c906$4399c964$c690c02a$30...@TSOFT.COM you asked "But with terms
defined as I intended (near-LCA-theropods vs. modern birds), surely
there is some character unique to modern birds?"
Just what meaning did you have in mind when you used it? What reason
do you have to believe that there's some trait present in all birds
alive today that wasn't possessed by the Passenger Pigeon?
>> But "living" isn't the criteria as to whether an animal should be
>> included in Neornithes.
>
>But it *is* the criteria as to whether a given species should be forced
>into that clade as part of its definition.
So?
>If the last of the rattites
>had gone extinct 200 years ago, we'd be justified in defining a clade
>that excluded them, and using that smaller clade instead of our current
>Neornithes as the crown group of living birds.
Presbyornis hasn't been around for roughly 40 million years, yet is
still considered to be part of Neornithes. Even if rattites had gone
extinct 200 years ago, it would still be part of your smaller crown
clade.
>> >Do you understand the difference between the questions whether
>> >Archaeopteryx was within the smallest clade containing modern birds
>> >(good question, no, not in that clade), vs. whether Archaeopteryx *is*
>> >a modern bird (stupid question)?
>> Let me play Creationist's Advocate for a moment, and say that I don't.
>> Then what?
>
>The *is* question is a matter of religion, of God-given names as
>revealed by sacred writings or recent revelation. The *clade* question
>is a matter of natural history, as revealed by fossils. If you find a
>Biblical passage stating clearly that Archaeopteryx *is* (or "was",
>since it's now extinct and might have already been extinct before our
>ancestors received that Word of God) a "bird" (as translated from
>Hebrew or Greek etc. in some non-ambiguous way), I'd be glad to accept
>that.
I don't believe that this is a criterion that a creationist would use,
and I'm unaware of any that do. After all, hummingbirds aren't
mentioned in the Bible, yet no creationist would deny that they're
birds.
>But if you tell me that *you* personally got such a revelation,
>I'd probably choose to call you a liar unless I happened to get that
>same revelation myself.
Not many creationists claim personal revelation, yet all would say
that Archaeopteryx was a "bird".
>(Regarding my metaphor of a xerox/kodak copy shop, descent with modification:)
>> A creationist would say that since there were no witnesses in the
>> past, coming to a conclusion of common ancestry would be premature.
>
>There are no witnesses even from yesterday. All we have today are
>imprints in the minds of people alive today, which we believe are
>memories of what they experienced yesterday, but there's no way to
>prove that.
I see.
>Fossils are older imprints, and natural rather than mental,
>but it's the same basic idea, and in fact fossils are more reliable
>than eyewitness testimony.
But taking your argument to its logical conclusion, it would seem that
we can never be sure of anything in this matter since we'll never have
all the fossils in front of us for a comparison. Anything less would
force us to rely on our fallible memories of five minutes ago. Do you
use such a worldview in day-to-day life?
>> >Well when somebody puts up a personal Web page or posts to a newsgroup,
>> >yes I understand it's all half crap. But when there's an online service
>> >that gives cladograms in nice format, I really assumed those were based
>> >on some science, not gross guesswork.
>> Creationist websites have nice formats too. Why not use the info
>> there?
>
>I believe a cladogram on a Creationist Web site would be an oxymoron.
Because?
Attempts which happen to prove my point.
> Metornithes is here considered inactive.
>That sounds like a prudent decision, given the conflicting definitions.
> Definition 2 The common ancestor of Mononykus and Neornithes plus all
> its descendants.
>Do you have a cladogram to show me where Metornithes satisfies that
>second definition, so I can compare it with the cladogram I cited above
>which is consistent with the original definition?
--Coelurosauria
|--Ornitholestes
|--Compsognathidae
`--Maniraptoriformes
|--Ornithomimosauria
| |--Ornithomimoidea
| | |--Alvarezsauridae
| | `--Ornithomimidae
| `--Therizinosauridae
`--Tyrannoraptora
|--Tyrannosauroidea
`->Maniraptora
|--Oviraptorosauria
`--Paraves
|--Deinonychosauria
| |--Dromaeosauridae
| `--Troodontidae
`->Avialae ("Aves" in Sereno)
Note that Mononykus is a member of Alvarezsauridae. The above
cladogram is from:
Sereno, P. C., 1999. The Evolution of Dinosaurs. Science
284:2137-2147.
Available on-line at:
http://cas.bellarmine.edu/tietjen/Ecology/evolution_of_dinosaurs.htm
Get a photocopy, and read it in front of your computer. That works for
me.
>By comparision, reading newsgroups is easy because I do
>that when I'm online and can go to Google or an online dictionary or
>thesaurus to look up terms whenever I don't understand some term
>somebody used. Being frustrated by not understanding twenty jargon
>terms per paragraph is no fun at all.
>
>> >Then I guess maybe I should hire somebody to write that sofware.
>> Go right ahead.
>
>If I found somebody to write it and install it online, would you use it
>too, or would I be the only person using it?
I can't speak for others, but I doubt I'd use it.
>By the way, I see a problem with trying to make lots of fully-defined
>triples while avoiding duplicate definitions along a non-branching
>stem. A triple requires the parent taxon to be node-based and each
>child-taxa to be stem-based. But if you want to link taxa like this:
> Grampa
> +Daddy
> | +Billy
> | `Sue
> `Mommy
>it's impossible for both Grampa = (Daddy + Mommy) and Daddy = (Billy + Sue)
>to both be triples, because the first imples Daddy is stem-based while
>the second implies Daddy is node-based.
That's why a clade is either stem-based, node-based, or apomorphy
based. The situation you're outlining would never happen.
an...@sci.sci
restricting the degrees of freedom to see what scale results:)
> Again, why? Do you learn anything from this exercise? It seems wholly
> arbitrary.
Yes, any particular restriction of degrees of freedom is arbitrary.
For each such arbitrary assignments of weights to the various
characters, we get a different scale, where the various fossils sit at
different places compared to where they'd sit on other scales.
By choosing the most obvious assignment of weights first, then varying
them within the limits that they must all be possible, and looking at
the results for each, we get an idea whether certain fossils really are
transitional along a sequence or perhaps they are outliers that are not
well enough constrained to be useful in demonstrating evolution from
early dinosaurs to modern birds.
Ideally the whole process should be automated. You feed in the various
data about fossil characters, and the computer performs a complete
degrees-of-freedom exploration and prints out upper and lower bounds
for each fossil on the scale from early dinosaurs to modern birds, and
also prints out all orderings between fossils that can be deduced
despite overlapping bounds. (For example, fossil #25 might be
restricted to range of 0.5 to 0.7, while fossil #92 might be restricted
to range of 0.6 to 0.75, which indicates either could come first, but
in fact fossil #25 might be provably before fossil #92, that is
whenever #25 is near the high end of its range #92 is also near the
high end of its range, sufficiently to keep #92 after #25 always.)
Combining the various cladogram programs that provide trees, which
don't all agree with each other, and the weighted-scale information, we
can tease the data into giving us a good idea how evolution progressed
from early dinosaurs to modern birds.
Also for presentation as "sight bites" on news programs, a sequence of
transitional forms, neatly spaced along a numeric scale based on actual
characters of fossils, would be better (more honest and enlightening)
than some concoction whereby a hunched-over chimp evolves via
artistic-morphing to a modern upright human, or the equivalent where an
obvious dinosaur artistically-morphs to modern bird, either of which
does not at all correspond to the sequence of character changes as they
actually occurred.
.
an...@sci.sci
> >>descriptions of Tyrannosaurus' large size.
> > Nope, that's a blatantly dishonest strawman. You should be ashamed of
> > what you said there. In fact, size would be just one character, which
> > would probably have low weight compared to density of feathers,
> > adaption of feathers for flying, hollowness of bones for respiration,
> > fusion of vertibrates, etc. etc. etc.
> Those all sound like qualitative characters to me, except for "density
> of feathers", and I have no idea how you would measure that one.
Every one of those is inherently quantitative. For example, you measure
what fraction of the bones are hollowed out and adapted for
respiration. If some unit of space is ambiguous, you assign it an
intermediate value based on your confidence. If some unit of space is
clearly hollow or not hollow, you assign it 1 or 0. Then you integrate
that over the total volume, and out comes a number. You check what
number you get for ancient dinosaurs (probably zero) and what number
you get for modern fully-flying birds, and there's one factor in your
scale.
Converting such measurements to words of a qualitative nature is a
crock. Best to leave them as numerical measurements. For example, in
horse evolution, what were originally claws, like on a modern housecat,
evolved to be hoofs, by getting shorter, and by thickening on the
underside. Clearly you can measure how much shorter and thicker they
got. Likewise toes were lost, but first the toes shrunk to near
uselessness. You can measure the shrinkage as a number.
.
an...@sci.sci
> >>included in Neornithes.
> > But it *is* the criteria as to whether a given species should be forced
> > into that clade as part of its definition.
> Not true. Neornithes includes more extinct species (most of them
> unknown) than extant ones.
You should stop deliberately trying to misunderstand me. I clearly said
"definition". The *definition* of Neornithes is the LCA of all *living*
birds and all descendents of that LCA, or in formal defintion the clade
from the LCA of just two living species
{Struthio camelus + Passer domesticus}
<http://www.taxonsearch.org/dev/taxon_edit.php?Action=View&tax_id=210>
Per the original definition, all living birds are forced into the
definition, and per the formal definition just those two particular
species (one from each major branch) are forced into the definition. In
either case, all other birds are in or out of that clade as a
consequence of the definition. No other birds are *forced* in as part
of the *definition*.
Species listed in the definition of a clade, and all species in total
contents of a clade, are not the same. When I talk about one, I don't
mean the other, and it's un-civil of you to deliberately mis-understand
what I said.
> > I wish there were some way to browse the cladistic tree, like start
> > with some term and browse to next-ancestor or to each next-descendent.
> This would require defining ancestors and descendants, which can't be
> done. If instead you mean from more inclusive to less inclusive clades
> (which is what I think you probably mean) ...
The two are equivalent, assuming common descent with no endosymbiosis
events in the part of the tree I'm browsing. For any clade, there is a
presumed founder species of that clade. For any species of the past,
there is a clade forward from that point. (For any presently-living
species that hasn't yet spawned another species, the clade is currently
just that single species. For any extinct species that didn't spawn any
other species before it went extinct, the clade is forever just that
one species.) For any founder species, there is a last-known-ancestor
species, and exactly two first-known-descendent species. Equivalently
for each clade there is a next-larger-known clade and two
next-smaller-known clades. If and when we discover new branch points in
the middle, the next-larger shrinks (last-ancestor gets later) and one
of the next-smaller grows (earliest-descendent gets earlier).
I suppose in some sense, in cladistic thinking and jargon, we *never*
name an actual species, only a clade. In the case of terminal nodes, we
use the name of a genus or species, but actually we are talking about
the clade containing that single genus or species.
Anyway, whether I'm thinking of the entire clade or just the
founder-species of the clade, yes that's what I meant by "browse the
cladistic tree".
I've tried that several times, including just now, but never have been
able to figure out how to get started with it. There's a search form,
but it doesn't place me in the tree, instead it executes a Google
search. For example, just now I tried "Neornithes" in the search form
and got 87 different matches, which is totally useless in my opinion.
"Neornithes" is a well-defined taxon, and so when I search for it in
tolweb I should get only a single result, the node in the tree for that
particular taxon.
I tried the first of the 87 matches, and found an up-stack where Aves
is supposed to be the immediate parent. Given that Aves is not well
defined, and per most of the definitions I've seen it is not
immediately the parent of Neornithes, that seems to be rather flaky. In
fact Aves is the *only* node above Neornithes and below Coelurosauria
according to that Web page.
Also on the same Web page, a down-list where there are three instead of
two sub-clades:
* Paleognathae
* Galloanserae
* Neoaves
I recognize the first two, but that third one looks wrong, and indeed a
Google search for that term either with taxon search or with cladogram
didn't turn up any of the usual reputable sources. The cladogram I get
for Neornithes shows five nodes instead of just one node between
Coelurosauria and Neornithes, and no mention of Neoaves whatsoever:
<http://www.palaeos.com/Vertebrates/Lists/Cladograms/320Theropoda.html>
--Coelurosauria DC, MH
|--Tyrannosauroidea X Th, ToL, MH
`--Maniraptora Th
|--Dromaeosauridae X Th
`--AVES MH, ToL
|--Archaeornithes X
`--Metornithes
|--Alvarezsauridae X
`--Ornithothoraces
|--Enantiornithes X MH
`--Ornithurae
|--Hesperornithiformes X MH
`--Neornithes MH
|--Paleognathae MH
| |--Lithornithiformes X MH
| `--Ratites Tax
`--+--GALLOANSERAE MH
| |--Galliformes
| | ...
| |--Anseriformes
| | ...
| `--CHARADRIOMORPHA MH
| ...
`--GRUIMORPHA MH
|--+--Gruiformes
| `--Podicipediformes
`--+--+--Strigiformes
| ...
`--+--Musophagidae
...
I tried just ignoring the fact that tolweb has only one instead of five
branch nodes between Coelurosauria and Neornithes. I clicked my way up
to Coelurosauria, saw subroups:
* Tyrannosauroidea
* Aves
clicked on Tyrannosauroidea and saw only one subroup:
* Tyrannosauridae
clicked on that and reached a dead-end, no subgroups in there.
By comparison, a cladogram elsewhere shows:
<http://town.morrison.co.us/dino-colo/taxonomy/dinoclad.php>
--Coelurosauria
`--Maniraptoriformes
|?-Ornitholestes
|?-Coeluridae
| `-->Coelurus
|?-Compsognathidae
|--Arctometatarsalia
| ...
|--Maniraptora
| |--Oviraptorosauria
| `--Eumaniraptora
| |--Deinonychosauria
| | ...
| `--Avialae
| ...
`--Tyrannosauroidea
|?->Stokesosaurus
`--Tyrannosauridae
`--Tyrannosaurinae
`--+-->Albertosaurus
`--Tyrannosaurini
`-->Tyrannosaurus
i.e. the whole path from Tyrannosauridae to Tyrannosaurus.
There are two possibilities: tolweb is very messed up (lots of missing
nodes, and some wrong nodes), or I haven't yet figured out how to make
any use of it to give me valid browsing links.
Oh by the way, I found in "advanced search" a way to directly find
exactly the Neornithes node I want instead of all 87 results that
Google turns up, but that's just the node I was already frustrated with
there.
.
an...@sci.sci
> descent?
No, all species have at least one common ancestor. Any two eukaryotes
have at least one, any two prokaryotes might have at least one, and one
prokaryote and one eukaryote have at least two common ancestors
(different common ancestors via the nucleus and mitochondria), and if
the eukaryote is algae or plant you get a third (the photosynthetic
part, the plastid or chloroplast etc.). But two prokaryotes might have
more than one, maybe as many as ten. We don't know for sure.
(All counts there are only locally-most-recent common ancestor. Any
ancestor of a common ancestor is also a common ancestor, but I'm not
counting those, only common ancestors which aren't ancestors of any
other common ancestor.)
In some cases, two species may be only partly related by common
ancestry, in that some of their genome comes from one or more common
ancestors, while the rest of their genomes comes from unrelated
ancestors of each.
> the definition of Paraves is all animals more closely related to
> Neornithes than to Oviraptor.
"more closely related" is ambiguous. I prefer:
<http://www.taxonsearch.org/dev/taxon_edit.php?Action=View&tax_id=248>
Active Definition The most inclusive clade containing Passer
domesticus (Linnaeus 1758) but not Oviraptor philoceratops Osborn
1924.
> >> What if you don't know the date of divergence?
> >There's no such thing as knowing "the date" exactly to the day for
> >something that happened more than 5000 years ago, except for
> >predictable astronomical events such as eclipses. So your question
> >isn't meaningful until you explain what you mean by "know the date".
> Yet you yourself state "that's a pretty accurate date if you're
> comparing that date with something that occurred between 64 and 66
> million years ago".
Indeed, if the best you know is that one event happened sometime
between 240 and 220 million years ago, and the other event happened
between 64 and 66 million years ago, that's sufficient accuracy to
determine for sure which event happened before the other. But in either
case is it correct or incorrect to say that you "know the date" as you
used that phrase? I don't know what you mean by "know the date" because
you still haven't defined that term. I know what "know the date" means
for a modern event (of the past 100 years for example) which happened
in an instant or during a very brief time during a single calendar day,
such as Kennedy's assassination. But to "know the date" in that sense
doesn't apply to any event that happened gradually over more than one
calendar day, such as the 2004-2005 new-years celebrations around the
world, and even moreso with pre-historic evolutionary events that
spanned thousands of years at the minimum and are known only to
hundreds of thousands of years accuracy at the very best.
> Not many creationists claim personal revelation, yet all would say
> that Archaeopteryx was a "bird".
On what basis would they make such a statement? And why would anybody care?
And what if any meaning, if any, would that statement have???
It seems to me to be a vacuous statemtment.
> >I believe a cladogram on a Creationist Web site would be an oxymoron.
> Because?
Because a Creationist doesn't believe there's any clade that includes
more than one species. Like begats like, same species, never a
different species, per Creationist belief.
.
John Harshman
I'm sorry, but this method of yours seems both arbitrary and
meaningless, as does any result you might come up with. There are actual
methods used for estimating the course of evolution, given a tree, and
they have much more basis in reality than anything you have proposed.
The simplest method is called parsimony optimization, and there are more
complicated maximum likelihood models too.
You seem to be using "degrees of freedom" as another non-standard
buzzword, too.
John Harshman
>>>>But "living" isn't the criteria as to whether an animal should be
>>>>included in Neornithes.
>>>
>>>But it *is* the criteria as to whether a given species should be forced
>>>into that clade as part of its definition.
>>
>>Not true. Neornithes includes more extinct species (most of them
>>unknown) than extant ones.
>
>
> You should stop deliberately trying to misunderstand me. I clearly said
> "definition". The *definition* of Neornithes is the LCA of all *living*
> birds and all descendents of that LCA, or in formal defintion the clade
> from the LCA of just two living species
> {Struthio camelus + Passer domesticus}
> <http://www.taxonsearch.org/dev/taxon_edit.php?Action=View&tax_id=210>
> Per the original definition, all living birds are forced into the
> definition, and per the formal definition just those two particular
> species (one from each major branch) are forced into the definition. In
> either case, all other birds are in or out of that clade as a
> consequence of the definition. No other birds are *forced* in as part
> of the *definition*.
Sorry. It was the term "forced in" that was confusing. I don't know what
you meant by that.
> Species listed in the definition of a clade, and all species in total
> contents of a clade, are not the same. When I talk about one, I don't
> mean the other, and it's un-civil of you to deliberately mis-understand
> what I said.
It's naive of you to think you were being at all clear about this. Now
that we understand each other, I have no idea what point you were trying
to make.
>>>I wish there were some way to browse the cladistic tree, like start
>>>with some term and browse to next-ancestor or to each next-descendent.
>>
>>This would require defining ancestors and descendants, which can't be
>>done. If instead you mean from more inclusive to less inclusive clades
>>(which is what I think you probably mean) ...
>
> The two are equivalent, assuming common descent with no endosymbiosis
> events in the part of the tree I'm browsing.
Not true. Clades are not ancestors. Primates is not the ancestor of
Homo. Primates is a clade that includes Homo. Two different things. Now
presumably there was an ancestral primate species whose descendants
include all other primates. But that species is most certainly not
identical to Primates.
> For any clade, there is a
> presumed founder species of that clade. For any species of the past,
> there is a clade forward from that point.
True but irrelevant.
> (For any presently-living
> species that hasn't yet spawned another species, the clade is currently
> just that single species. For any extinct species that didn't spawn any
> other species before it went extinct, the clade is forever just that
> one species.) For any founder species, there is a last-known-ancestor
> species, and exactly two first-known-descendent species. Equivalently
> for each clade there is a next-larger-known clade and two
> next-smaller-known clades. If and when we discover new branch points in
> the middle, the next-larger shrinks (last-ancestor gets later) and one
> of the next-smaller grows (earliest-descendent gets earlier).
>
> I suppose in some sense, in cladistic thinking and jargon, we *never*
> name an actual species, only a clade. In the case of terminal nodes, we
> use the name of a genus or species, but actually we are talking about
> the clade containing that single genus or species.
In fact we are not. Genera are supposed to be clades, but whether
species are clades depends on what species concept you are choosing to
use. Ancestral species, of course, can't be clades.
> Anyway, whether I'm thinking of the entire clade or just the
> founder-species of the clade, yes that's what I meant by "browse the
> cladistic tree".
I thought so. But if you're thinking of the founder-species of a clade,
like I said there can be no such program, since we don't know of any
such species.
>>http://www.tolweb.org
>
> I've tried that several times, including just now, but never have been
> able to figure out how to get started with it. There's a search form,
> but it doesn't place me in the tree, instead it executes a Google
> search. For example, just now I tried "Neornithes" in the search form
> and got 87 different matches, which is totally useless in my opinion.
> "Neornithes" is a well-defined taxon, and so when I search for it in
> tolweb I should get only a single result, the node in the tree for that
> particular taxon.
You are seriously search-impaired.
> I tried the first of the 87 matches, and found an up-stack where Aves
> is supposed to be the immediate parent. Given that Aves is not well
> defined, and per most of the definitions I've seen it is not
> immediately the parent of Neornithes, that seems to be rather flaky. In
> fact Aves is the *only* node above Neornithes and below Coelurosauria
> according to that Web page.
The reason for this is that TOL concentrates on extant taxa. But did you
notice that you found a tree, and that clicking on nodes gets you to
other trees, just like you wanted?
> Also on the same Web page, a down-list where there are three instead of
> two sub-clades:
> * Paleognathae
> * Galloanserae
> * Neoaves
> I recognize the first two, but that third one looks wrong, and indeed a
> Google search for that term either with taxon search or with cladogram
> didn't turn up any of the usual reputable sources.
This is an index of both your ignorance of modern terminology and of the
web's unreliability. Neoaves is the generally accepted name (within the
last 10 years or so, granted) for the clade that includes all extant
birds except for paleognaths and galloanserines.
> The cladogram I get
> for Neornithes shows five nodes instead of just one node between
> Coelurosauria and Neornithes, and no mention of Neoaves whatsoever:
That's because this is a fantasy cladogram based mostly on whim, and
unsupported by data. I forget where they dug up that tree, but it
doesn't conform to what we know.
There is a third possibility: TOL doesn't go into as much detail as you
would like (especially for extinct taxa), but where the two trees
conflict, TOL is correct.
> Oh by the way, I found in "advanced search" a way to directly find
> exactly the Neornithes node I want instead of all 87 results that
> Google turns up, but that's just the node I was already frustrated with
> there.
Some people are never happy.
an...@sci.sci
> Primates is a clade that includes Homo. Two different things. Now
> presumably there was an ancestral primate species whose descendants
> include all other primates. But that species is most certainly not
Who cares? There's an isomorphism between the heirarchy of clades per
set-inclusion, and the hierarchy of founders of clades per ancestry.
For every clade, there's a founder. For every founder, there's a clade.
The clade of a founder of a clade is the original clade.
The founder of a clade of a founder is the original founder.
Let F be the founder-of operator.
For any two clades C1 C2:
F(C1) = F(C2) if and only if C1 = C2.
F(C1) ancestor of F(C2) if and only if C1 contains C2 as proper subset.
(If we allow individual to be his own ancestor, then you can just use subset
instead of proper subset.)
The tree structure you get by drawing all the clades and all the
subset-set relations between them is exactly the same tree you get by
drawing all the founders-of-clades and all the ancestor-descendent
relations between them.
The structures, the trees, the hierarchies in the structural sense,
are identical whether you are talking about clades or ancestors of clades.
42 is not a number. It's a digital representation of a number.
More specificlly "42" is a string which is decimal notation for an integer.
But who cares? The set of numbers, under the usual operations that
mathematicians speak of, is exactly the same arithmetic system as the
set of decimal representations of numbers under the decimal arithmetic
operations that we learn in school. There's a single arithmetic system
we call "integers under addition and multiplication", it's just that
there are multiple ways we can express that system.
Likewise it's the same hierarchy whether we explicitly talk about the
nodes (founders of clades), or the sub-trees (the clades themselves),
and regardless of whether we draw the (rooted) tree horizontally across
the page or vertically down the page, and regardless of whether we draw
the tree in symmetric form (parent at left, children diagonally up-ight
and down-right or broken-line up-then-right and down-then-right) or in
outline form (parent listed first, children indented underneath). To
say it's a different hierarchy depending on whether we look at the
nodes or the sub-trees, or depending on how we draw or label the graph,
is perverse. The *hierarchy*, the structure of all those related
things, is the same regardless. It's just the form that is different.
> Genera are supposed to be clades,
If you have your cladistics correct, then every genus has a founder
species. In many cases that founder hasn't been sampled by fossils, but
if you have enough members of the clade you might be able to deduce the
genome hence the phenotype of the founder (the LCA).
So there's a mapping between genus and founder-of-genus.
> but whether species are clades depends on what species concept you
> are choosing to use. Ancestral species, of course, can't be clades.
Actually no species is supposed to be a clade for all time, except if
it goes extinct before spawning any new species. But for every species
there's a clade founded by that species.
So there's a mapping between clade-founded-by-species and species.
And it's the same mapping in both cases, except the latter applies to
larger clades than just genera. Every genus is a clade founded by a
species, but not every clade founded by a species is a genus. Every
founder of genus is a species, but not every species is a founder of
species. Anyway, it's just a convention of convenience whether we talk
about the founding-species of a clade or the whole clade as a group
when assigning names to nodes in an evolutionary tree. Some nodes
(usually branch nodes) are labeled by names of clades, while some other
nodes (usual terminal nodes) are labeled sometimes by names of species
and sometimes by names of genera and sometimes by names of larger
clades that we don't wish to sub-divide further at a particular time
because we want the tree to fit on one page. When I see a mix of clade
names and individual names in an evolutionary tree, it doesn't bother
me, I understand that each individual name can be mapped to a clade
that it founded and each clade can be mapped to its founder. When I
look at such a tree, I generally think of nodes as founders of clades
which are the entire sub-tree from there, even if the label on the
graph is usually the name of the whole clade. For example, when I see:
+--Chordata
--+
| +--Hemichordata
+--+
+--Echinodermata
I think of a common ancestor to all three pyhla, which begat the common
ancestor of all Chordates and the common ancestor of Hemichordata and
Echinodermata, and that latter begat the common ancestor of
Hemichordata and the common ancestor of Echinodermata. In addition to
the three individual phyla, there are two higher-up (larger) clades
(sub-trees) shown here, the H+E sub-tree, and the whole C+H+E tree
which is a sub-tree of some larger tree of life. Does that diagram show
a nested hierarchy of clades, or a non-nested hierarchy of founders?
Yes, it shows both. It's all the same tree, the same hierarchy.
> > Anyway, whether I'm thinking of the entire clade or just the
> > founder-species of the clade, yes that's what I meant by "browse the
> > cladistic tree".
> I thought so. But if you're thinking of the founder-species of a clade,
> like I said there can be no such program, since we don't know of any
> such species.
In most cases I agree. The founder is implied, not sampled.
So the name on the node is usually the clade name, not the implied species name.
It doesn't matter to me. I'd like to be able to browse in some way that
doesn't skip a majority of the branch points the way tolweb does, and
which actually goes all the way down to known genera likewise unlike
tolweb. Lacking any such browing capability, I suppose I could use
tolweb to browse just the major nodes and then use Google search for
cladograms to find the rest of the nodes in the vicinity of the major
nodes, a pain but maybe serviceable, haven't tried that method
extensively yet. Still it'd be nice to have a single Web service for
browsing *all* the known branch points and leaves. If manually
integrating the very sparse tree from tolweb with the detailed parts of
tree from Google cladogram search is generally successful, then it
might be easy to automate the process, so it wouldn't cost me a bundle
to hire somebody to hack a service that merges the two sources of data.
(in tolweb:)
> > Aves is the *only* node above Neornithes and below Coelurosauria
> > according to that Web page.
> The reason for this is that TOL concentrates on extant taxa.
But that's inconsistent, because Aves is the LCA of Neornithes and an
extinct species, so there's no reason for Aves to be a node in their
tree. Apparently they concentrate on extant taxa, but throw in a few
extinct taxa too, not the taxa themselves, only the branch to the
extinct taxa. Is that what you meant to say? But in the case of
dinosaurs, they have a whole slew of non-extant (extinct) taxa forced
in to make branch points that wouldn't exist in a pure-extant (crown
group) tree.
> But did you notice that you found a tree, and that clicking on nodes
> gets you to other trees, just like you wanted?
But the tree is so very very sparse, showing only a single node between
ancient dinosaurs and modern birds, and that single node "Aves" happens
to be the one such transitional node that is defined a different way by
each different researcher to where it's useless as a taxon. That's of
no use whatsoever for my purposes of interpolating weighted mixes of
characters from ancient dinosaurs to modern birds.
In fact I don't even know which of the several definitions of "Aves"
tolweb is using, if any. They might be deliberately having that be the
only branch-node along that entire path so it doesn't matter what the
definition is it still is in the "right" place between ancient
dinosaurs and modern birds. (That's per the founder view of the tree.
Per the clade-subset view of the tree, Aves is smaller than clade of
theropods and birds, but larger than clade of modern birds only.)
Let me go back to tolweb and see if the descriptive English gives any
clue which definition of "Aves" they are using, if any ...
<http://tolweb.org/tree?group=Aves&contgroup=Coelurosauria>
Not a single word of description of this clade, only a large list of
references in alphabetical order. So I have no idea how inclusive Aves
is supposed to be according to tolweb. One line says:
Classification follows Sereno 1999.
which is useless to me in guessing how Aves is defined, what it is the
LCA of. I can't even tell whether Archaeopteryx is within Aves or not.
There is mention of Archaeopteryx on that Web page, but it isn't clear
whether it's considered a member or a sister group. It's definitely not
included in the sub-clade buttons, not even as plain text without a
link. Neornithesis the only sub-clade listed.
Compare with:
<http://tolweb.org/tree?group=Neornithes>
Neornithes includes all extant birds. The earliest divergence within
Neornithes is between Paleognathae (ratites and tinamous) and
Neognathae which includes the two primary taxa Galloanserae and
Neoaves ...
See the descriptive text there telling how the clade is defined?
Later on that page there are additional sections giving much more
detail about modern birds. No question how this clade is defined.
> > Also on the same Web page, a down-list where there are three instead of
> > two sub-clades:
> > * Paleognathae
> > * Galloanserae
> > * Neoaves
> > I recognize the first two, but that third one looks wrong, and indeed a
> > Google search for that term either with taxon search or with cladogram
> > didn't turn up any of the usual reputable sources.
> This is an index of both your ignorance of modern terminology and of the
> web's unreliability. Neoaves is the generally accepted name (within the
> last 10 years or so, granted) for the clade that includes all extant
> birds except for paleognaths and galloanserines.
I agree. There is no source of information, at least none that I know
of, none that you have told me about, whereby I can get recent correct
data. So I'm pretty much limited to scrounging up whatever crap has
been indexed by Google, showing it to you, and having you say it's
crap, and giving me only the slightest clues what the latest best
cladogram might really be. tolweb with not a single well-defined node
between Coelurosauria and Neornithes is definitely not a good source of
recent complete data. I thought taxonsearch.org was pretty good, except
for showing me *only* the formal definition of the taxon, not a
cladogram around the main taxon, not even just the parent taxon and
child taxa which I could use for browsing, but still pretty nice
cladistic definitions. But it doesn't list Neoaves, at least it doesn't
turn up in Google search, so should I dismiss it as an authority? Let
me see if I can find a direct search page at taxonsearch.org, bypass
Google. Maybe the data is there but Google hasn't indexed it yet?
<http://www.taxonsearch.org/dev/taxon_search.php>
Oh, I tried that once before and didn't have any luck. Trying again,
first one I know it has, searching for Neornithes, one match:
# Taxon Nominal Author Taxon Status Definitional Author Defn Type
1 Neornithes Gadow 1883 active Sereno 2005 node
<http://www.taxonsearch.org/dev/taxon_edit.php?Action=View&tax_id=210>>
Active Definition The least inclusive clade containing Struthio
camelus (Linnaeus 1758) and Passer domesticus (Linnaeus 1758).
Now the real test, same kind of search, but for Neoaves:
0 Records Found
So do you know of *any* free Web-based taxon search that finds Neoaves?
And even if we can find that, how to browse without skipping all the
way from Coelurosauria to Neornithes in a single jump?
> > The cladogram I get
> > for Neornithes shows five nodes instead of just one node between
> > Coelurosauria and Neornithes, and no mention of Neoaves whatsoever:
> That's because this is a fantasy cladogram based mostly on whim, and
> unsupported by data. I forget where they dug up that tree, but it
> doesn't conform to what we know.
Good, I'll shun palaeos.com as a reliable source of cladograms on the
Web. You know any better source that is also freely accessible on the Web?
> > There are two possibilities: tolweb is very messed up (lots of missing
> > nodes, and some wrong nodes), or I haven't yet figured out how to make
> > any use of it to give me valid browsing links.
> There is a third possibility: TOL doesn't go into as much detail as you
> would like (especially for extinct taxa), but where the two trees
> conflict, TOL is correct.
That's covered under "I haven't figured out how to make effective use
of it yet". So how about I use tolweb only for the bare skeleton, a
framework very likely to be correct, but use other crufty sources to
fill in the details within the huge gaps that tolweb leaves empty, and
ask you to confirm each and every such crufty-detail set I find, until
we find any reliable source for the details?
> Some people are never happy.
I don't know about *never*, but I'm not happy yet about sources for
cladistic hierarchy data on the Web. Some (tolweb.org) are grossly
sparse, some (taxonsearch.org) are hopelessly incomplete or out of
date, and others (palaeos.com) are hopelessly
unreliable/fantasylandish.
.
John Harshman
[snip]
> In most cases I agree. The founder is implied, not sampled.
> So the name on the node is usually the clade name, not the implied species name.
> It doesn't matter to me. I'd like to be able to browse in some way that
> doesn't skip a majority of the branch points the way tolweb does, and
> which actually goes all the way down to known genera likewise unlike
> tolweb. Lacking any such browing capability, I suppose I could use
> tolweb to browse just the major nodes and then use Google search for
> cladograms to find the rest of the nodes in the vicinity of the major
> nodes, a pain but maybe serviceable, haven't tried that method
> extensively yet. Still it'd be nice to have a single Web service for
> browsing *all* the known branch points and leaves. If manually
> integrating the very sparse tree from tolweb with the detailed parts of
> tree from Google cladogram search is generally successful, then it
> might be easy to automate the process, so it wouldn't cost me a bundle
> to hire somebody to hack a service that merges the two sources of data.
I don't know if the web has what you want, even if you combine sources.
A lot is still done on paper, and there is still no substitute for a
university library. If you will note, each TOL tree is extensively
documented by citations from the scientific literature that you could
consult if you want more detail.
TOL also has the virtue that it seems to be updated regularly. The bird
part, at least, seems pretty much up to date. The tree that used to be
there has been replaced as of 2004 with one that's a bit better. I would
make a few changes, but not all that many; it does represent to a large
degree the consensus among workers in the field, and it's maintained by
someone who actually does work in that field. The trees you can find at
other points on the web are a bit more dicey. Paleos has a weird tree,
and I don't know where it came from. I think one of their problems may
be that they ignore the molecular literature.
> (in tolweb:)
>
>>>Aves is the *only* node above Neornithes and below Coelurosauria
>>>according to that Web page.
>>
>>The reason for this is that TOL concentrates on extant taxa.
>
> But that's inconsistent, because Aves is the LCA of Neornithes and an
> extinct species,
Archaeopteryx lithographica, to be precise.
> so there's no reason for Aves to be a node in their
> tree. Apparently they concentrate on extant taxa, but throw in a few
> extinct taxa too, not the taxa themselves, only the branch to the
> extinct taxa.
Actually, if you look you will see that Archaeopteryx and a number of
non-neornithines are listed, though they don't show a tree for some reason.
> Is that what you meant to say? But in the case of
> dinosaurs, they have a whole slew of non-extant (extinct) taxa forced
> in to make branch points that wouldn't exist in a pure-extant (crown
> group) tree.
Yes, they go into more detail in some areas than in others. The
different taxa are maintained by different people who may have different
ideas of what to include or leave out. And some pages are just
place-holders, under construction. I'm sure if you complained then
everyone would try harder.
>>But did you notice that you found a tree, and that clicking on nodes
>>gets you to other trees, just like you wanted?
>
>
> But the tree is so very very sparse, showing only a single node between
> ancient dinosaurs and modern birds, and that single node "Aves" happens
> to be the one such transitional node that is defined a different way by
> each different researcher to where it's useless as a taxon. That's of
> no use whatsoever for my purposes of interpolating weighted mixes of
> characters from ancient dinosaurs to modern birds.
The definition is in fact clear from context. And your purposes are just
bizarre.
> In fact I don't even know which of the several definitions of "Aves"
> tolweb is using, if any.
I think it's obvious, given the taxa that are listed under "Aves". Look
again.
> They might be deliberately having that be the
> only branch-node along that entire path so it doesn't matter what the
> definition is it still is in the "right" place between ancient
> dinosaurs and modern birds. (That's per the founder view of the tree.
> Per the clade-subset view of the tree, Aves is smaller than clade of
> theropods and birds, but larger than clade of modern birds only.)
> Let me go back to tolweb and see if the descriptive English gives any
> clue which definition of "Aves" they are using, if any ...
> <http://tolweb.org/tree?group=Aves&contgroup=Coelurosauria>
> Not a single word of description of this clade, only a large list of
> references in alphabetical order. So I have no idea how inclusive Aves
> is supposed to be according to tolweb.
This is silly. They list its contents. The choice among the several
existing definitions is obvious.
> One line says:
> Classification follows Sereno 1999.
> which is useless to me in guessing how Aves is defined, what it is the
> LCA of. I can't even tell whether Archaeopteryx is within Aves or not.
That's a perceptual problem on your part, since it's clearly on that
list of the contents of Aves. You could also go to a library and look up
Sereno 1999. Is there a reason not to?
> There is mention of Archaeopteryx on that Web page, but it isn't clear
> whether it's considered a member or a sister group.
It's clear to me. I don't know why it's not clear to you.
> It's definitely not
> included in the sub-clade buttons, not even as plain text without a
> link. Neornithesis the only sub-clade listed.
>
> Compare with:
> <http://tolweb.org/tree?group=Neornithes>
> Neornithes includes all extant birds. The earliest divergence within
> Neornithes is between Paleognathae (ratites and tinamous) and
> Neognathae which includes the two primary taxa Galloanserae and
> Neoaves ...
> See the descriptive text there telling how the clade is defined?
> Later on that page there are additional sections giving much more
> detail about modern birds. No question how this clade is defined.
Seldom have I heard such whining. If you don't like it, do your own. Or
realize that the web is not currently as good as the print literature,
and go to a library.
>>>Also on the same Web page, a down-list where there are three instead of
>>>two sub-clades:
>>> * Paleognathae
>>> * Galloanserae
>>> * Neoaves
>>>I recognize the first two, but that third one looks wrong, and indeed a
>>>Google search for that term either with taxon search or with cladogram
>>>didn't turn up any of the usual reputable sources.
>>
>>This is an index of both your ignorance of modern terminology and of the
>>web's unreliability. Neoaves is the generally accepted name (within the
>>last 10 years or so, granted) for the clade that includes all extant
>>birds except for paleognaths and galloanserines.
>
> I agree. There is no source of information, at least none that I know
> of, none that you have told me about, whereby I can get recent correct
> data.
I'm not sure what you are asking here. Data? That would be, mostly, on
GenBank. If you want trees, that will be in the paper literature. Note
that TOL gives references to this literature in case you want more
information. Ignoring those references and whining about the lack of
information is both dishonest and lazy. The most reliable web source for
up to date trees that I know of is TOL. Don't know what else to do for you.
> So I'm pretty much limited to scrounging up whatever crap has
> been indexed by Google, showing it to you, and having you say it's
> crap, and giving me only the slightest clues what the latest best
> cladogram might really be. tolweb with not a single well-defined node
> between Coelurosauria and Neornithes is definitely not a good source of
> recent complete data. I thought taxonsearch.org was pretty good, except
> for showing me *only* the formal definition of the taxon, not a
> cladogram around the main taxon, not even just the parent taxon and
> child taxa which I could use for browsing, but still pretty nice
> cladistic definitions. But it doesn't list Neoaves, at least it doesn't
> turn up in Google search, so should I dismiss it as an authority?
On the classification of modern birds, at least, yes. Actually, the
problem there is not being wrong, but just lack of coverage. Sereno
doesn't go within Neornithes at all. Not the absense of Neognathae too,
or or any other clade within Neornithes. Taxonsearch is a work in
progress and is at a very early stage of it -- so far, more or less just
those taxa that Sereno himself works on. And he pretty much ends where
Neornithes begins.
> So do you know of *any* free Web-based taxon search that finds Neoaves?
No, but I haven't looked either. OK, now I googled, and found several
references. Why exactly do you need a taxon search, specifically? But
you should try Systema Naturae 2000 and the Taxonomicon, which might fit
your needs, or might not. Both of these use Sibley & Ahlquist's
classification, which is bogus in many places. If you are looking for
another web source on Neoaves, Evowiki isn't bad either, though like
many sites it largely ignores the molecular literature.
The best current guide to the phylogeny of Neoaves is Cracraft, J.,
Barker, F. K., Braun, M. J., Harshman, J., Dyke, G., Feinstein, J.,
Stanley, S., Cibois, A., Schikler, P., Beresford, P., Garcia-Moreno, J.,
Sorenson, M. D., Yuri, T. and Mindell, D. P. 2004. Phylogenetic
relationships among modern birds (Neornithes): Toward an avian tree of
life. Pages 468-489 In J. Cracraft and M. J. Donoghue (eds), Assembling
the tree of life. Oxford University Press, New York.
You might even want to buy the book, since it's a great resource on all
sorts of phylogenetic questions.
> And even if we can find that, how to browse without skipping all the
> way from Coelurosauria to Neornithes in a single jump?
Paleos is probably the best you can get for that, on the web, right now.
Print is better.
>>>The cladogram I get
>>>for Neornithes shows five nodes instead of just one node between
>>>Coelurosauria and Neornithes, and no mention of Neoaves whatsoever:
>>
>>That's because this is a fantasy cladogram based mostly on whim, and
>>unsupported by data. I forget where they dug up that tree, but it
>>doesn't conform to what we know.
>
> Good, I'll shun palaeos.com as a reliable source of cladograms on the
> Web. You know any better source that is also freely accessible on the Web?
No.
>>>There are two possibilities: tolweb is very messed up (lots of missing
>>>nodes, and some wrong nodes), or I haven't yet figured out how to make
>>>any use of it to give me valid browsing links.
>>
>>There is a third possibility: TOL doesn't go into as much detail as you
>>would like (especially for extinct taxa), but where the two trees
>>conflict, TOL is correct.
>
> That's covered under "I haven't figured out how to make effective use
> of it yet". So how about I use tolweb only for the bare skeleton, a
> framework very likely to be correct, but use other crufty sources to
> fill in the details within the huge gaps that tolweb leaves empty, and
> ask you to confirm each and every such crufty-detail set I find, until
> we find any reliable source for the details?
You could. I'm not the world's expert on non-avian theropods, but I do
know the literature fairly well. A better idea would be to go to a
libarary, look up recent literature on theropod phylogeny, and see what
the points of agreement and disagreement are among workers. The May 2005
issue of Natural History has a nice tree, and the discussion around that
tree gives some idea of points of disagreement. I believe there's also a
short list of useful literature that should at least get you into it.
>>Some people are never happy.
>
> I don't know about *never*, but I'm not happy yet about sources for
> cladistic hierarchy data on the Web. Some (tolweb.org) are grossly
> sparse, some (taxonsearch.org) are hopelessly incomplete or out of
> date, and others (palaeos.com) are hopelessly
> unreliable/fantasylandish.
The last is not necessarily the case throughout Palaeos. Their
Neornithes tree is hopeless, but that doesn't mean they have similar
problems with other groups. Taxonsearch has just started; give it time.
And even TOL is a work in progress. Solution: go straight to paper. Read
actual books and journals if you want to know this stuff.
Augray
>> >No, please try to read more carefully. Two species are either related
>> >by common descent, or not.
>>
>> Can you name two species that you believe are not related by common
>> descent?
>
>No, all species have at least one common ancestor.
Then why did you write "Two species are either related by common
descent, or not"?
But how is that determined?
>But in either
>case is it correct or incorrect to say that you "know the date" as you
>used that phrase? I don't know what you mean by "know the date" because
>you still haven't defined that term.
I mean the same thing as you when you stated "that's a pretty accurate
date".
> I know what "know the date" means
>for a modern event (of the past 100 years for example) which happened
>in an instant or during a very brief time during a single calendar day,
>such as Kennedy's assassination. But to "know the date" in that sense
>doesn't apply to any event that happened gradually over more than one
>calendar day, such as the 2004-2005 new-years celebrations around the
>world, and even moreso with pre-historic evolutionary events that
>spanned thousands of years at the minimum and are known only to
>hundreds of thousands of years accuracy at the very best.
So what did you mean when you wrote "that's a pretty accurate date if
you're comparing that date with something that occurred between 64 and
66 million years ago".
>> Not many creationists claim personal revelation, yet all would say
>> that Archaeopteryx was a "bird".
>
>On what basis would they make such a statement?
The characteristics of the animal.
>And why would anybody care?
Well, you seem to.
>And what if any meaning, if any, would that statement have???
>It seems to me to be a vacuous statemtment.
So, taxonomic classification is vacuous?
>> >I believe a cladogram on a Creationist Web site would be an oxymoron.
>> Because?
>
>Because a Creationist doesn't believe there's any clade that includes
>more than one species. Like begats like, same species, never a
>different species, per Creationist belief.
This is incorrect. See
http://www.answersingenesis.org/creation/v22/i3/ligers_wolphins.asp
an...@sci.sci
> descent, or not"?
Because that's the only way to satisfy what you claimed earlier in the
thread, that the mere fact of having a common ancestor provides a
metric. My point is that such existance of common ancestor provides at
best a boolean metric, all or nothing, related or not related, zero
distance or infinite distance, which is completely useless in deciding
which tree to draw to span the various taxa.
For all the taxa of interest within a single domain, it's highly likely
that any two share a common ancestor, so per the all-or-nothing metric
they would all be zero distance from each other.
> >> Not many creationists claim personal revelation, yet all would say
> >> that Archaeopteryx was a "bird".
> >On what basis would they make such a statement?
> The characteristics of the animal.
That's only the "minor premise" in the logic. What is the major
premise, i.e. the definition of "bird" by which they judge some animals
as "bird" and some others as "not bird"?? I personally don't consider a
monster with ferocious looking teeth to be a "bird". At best I'd
consider it to be some kind of chimera between a bird and some
crocodile-like monster, like a gryphon in some sense.
> So, taxonomic classification is vacuous?
No, but treating Archaeopteryx as if it were identical to regular
(modern) birds would be an absurd way to classify animals. Treating
Archaeopteryx as transitional between some ancient group (such as
dinosaurs) and birds would make a lot more sense. Given that
Archaeopteryx indeed lived and went extinct appx. 150 million years
ago, trying to cram it into a fixed set of modern taxa seems a bad
policy.
In most cases the last common ancestor of two modern groups (and any
side group very similar to that LCA) should be considered a three-way
transitional, between the two modern groups and also an earlier
ancestal form. For example, the LCA between modern reptiles and modern
mammals might be considered a transitional between reptiles and mammals
and ancient amphibians.
> http://www.answersingenesis.org/creation/v22/i3/ligers_wolphins.asp
If two animals or two plants can hybridize (at least enough to produce
a truly fertilized egg), then they must belong to (i.e. have descended
from) the same original created kind.
So if A can hybridize with B, and B can hybridize with C, but A cannot
hybridize with C, does that mean A and B are of the same kind, and B
and C are of the same kind, but A and C are not of the same kind?
Whatever sense can that have?? None as far as I can tell.
In the case of three species, A, B and C, if A and B can each
hybridize with C, then it suggests that all three are of the same
created kind -- whether or not A and B can hybridize with each other.
Then by that logic, if there's a chain of pairwise hybridizable links
A-B-C-...-X-Y-Z then all of A thru Z are of the same kind? Suppose we
find that I can potentially mate with somebody five years older than
myself, and that person can potentially mate with somebody five years
older still, etc. by five-year links all the way from myself today to
some distant ancestor 400 million years ago. So by that logic, I and
that distant incestor are of the same kind? And likewise a parakeet
could mate with another parakeet a year or two older, and by 1-2 year
links that chain could also be carried back to 400 million years ago,
in fact to the common ancestor of all birds and all mammals. So all
birds and mammals are of the same kind? Perhaps all eukaryotes are of
the same kind via that transitive definition. Of what value is such a
definition?
it is
overwhelmingly likely that horses, asses and zebras (six species of
Equus) are the descendants of the one created kind which left the Ark.
But the Ark was just something in a story, not anything which really
existed. The whole Flood story was a fiction.
.
John Harshman
[snip boring parts]
>>>>Not many creationists claim personal revelation, yet all would say
>>>>that Archaeopteryx was a "bird".
>>>
>>>On what basis would they make such a statement?
>>
>>The characteristics of the animal.
>
> That's only the "minor premise" in the logic. What is the major
> premise, i.e. the definition of "bird" by which they judge some animals
> as "bird" and some others as "not bird"??
Mostly, it seems to be feathers. But all the feathered theropods confuse
them. There was, briefly, a fashion for considering Archaeopteryx to be
just a theropod whose feathers had been faked. And strangely, Fred Hoyle
was the author. But it does demonstrate the crucial feature: feathers,
bird; no feathers or faked feathers, not a bird. Conveniently, all
extraneous variables were eliminated from the test, because both groups
were talking about the same animal.
There are other points of contention among creationists too. Homo
erectus is variously described as "just an ape" or "just a modern
human", depending on who or when you read.
> I personally don't consider a
> monster with ferocious looking teeth to be a "bird". At best I'd
> consider it to be some kind of chimera between a bird and some
> crocodile-like monster, like a gryphon in some sense.
They aren't really all that ferocious. In fact, they're tiny, and
missing from the forward third of the mouth.
>>So, taxonomic classification is vacuous?
>
> No, but treating Archaeopteryx as if it were identical to regular
> (modern) birds would be an absurd way to classify animals. Treating
> Archaeopteryx as transitional between some ancient group (such as
> dinosaurs) and birds would make a lot more sense. Given that
> Archaeopteryx indeed lived and went extinct appx. 150 million years
> ago, trying to cram it into a fixed set of modern taxa seems a bad
> policy.
Not sure what you guys are arguing about here, if anything. There are
various clades. We give them various names. What's the problem?
> In most cases the last common ancestor of two modern groups (and any
> side group very similar to that LCA) should be considered a three-way
> transitional, between the two modern groups and also an earlier
> ancestal form. For example, the LCA between modern reptiles and modern
> mammals might be considered a transitional between reptiles and mammals
> and ancient amphibians.
What benefit do we gain by doing this? And what's the point of arguing
about something we have no actual way of identifying?
>
>>http://www.answersingenesis.org/creation/v22/i3/ligers_wolphins.asp
>
> If two animals or two plants can hybridize (at least enough to produce
> a truly fertilized egg), then they must belong to (i.e. have descended
> from) the same original created kind.
>
> So if A can hybridize with B, and B can hybridize with C, but A cannot
> hybridize with C, does that mean A and B are of the same kind, and B
> and C are of the same kind, but A and C are not of the same kind?
> Whatever sense can that have?? None as far as I can tell.
I believe it's commonly (by creationists) considered that in the case
you discuss, they are all the same kind. And in fact this works for me.
The question is not how you can show that two species belong to the same
kind. The question is how you can show that they don't. And the
existence of non-transitive relationships of this sort does indeed show
that reproductive isolation can evolve, so the interbreeding test is
clearly not able to show that two species don't belong to the same kind.
They could have been created that way (different kinds) or they could
have evolved that way (same kind) and the creationists do not suggest a
criterion to distinguish these cases.
> In the case of three species, A, B and C, if A and B can each
> hybridize with C, then it suggests that all three are of the same
> created kind -- whether or not A and B can hybridize with each other.
>
> Then by that logic, if there's a chain of pairwise hybridizable links
> A-B-C-...-X-Y-Z then all of A thru Z are of the same kind?
Yes. Makes sense to me. Why doesn't it make sense to you?
> Suppose we
> find that I can potentially mate with somebody five years older than
> myself, and that person can potentially mate with somebody five years
> older still, etc. by five-year links all the way from myself today to
> some distant ancestor 400 million years ago. So by that logic, I and
> that distant incestor are of the same kind?
Well of course they are, if there is any such ancestor. Do you know at
all what creationists mean by "kind"? It's just a term for groups that
share ancestry within the group but not outside the group. Of course
it's pretty clear that by this definition there is only one kind, but
creationists don't see that.
At any rate, your little thought experiment assumes common descent, so
it's no surprise that you find common descent. A creationist would claim
that around 6000 years ago, if you traced this all back, you would come
to a stopping point before which there are no ancestors at all.
> And likewise a parakeet
> could mate with another parakeet a year or two older, and by 1-2 year
> links that chain could also be carried back to 400 million years ago,
> in fact to the common ancestor of all birds and all mammals. So all
> birds and mammals are of the same kind? Perhaps all eukaryotes are of
> the same kind via that transitive definition. Of what value is such a
> definition?
I'm afraid that you can't assemble any actual data to support this
theory, since we can't identify any actual ancestors. If you put it
quite another way, and rely on that nested hierarchy, then there's
plenty of data. But not in the way you are saying here. What's your
evidence for the existence of a continuous chain of human ancestors
reaching back 400ma? It's there, but your simple, fiat solution
accomplishes nothing.
> it is
> overwhelmingly likely that horses, asses and zebras (six species of
> Equus) are the descendants of the one created kind which left the Ark.
>
> But the Ark was just something in a story, not anything which really
> existed. The whole Flood story was a fiction.
Not to creationists.
an...@sci.sci
> erectus is variously described as "just an ape" or "just a modern
> human", depending on who or when you read.
Has anybody collected such data and posted a Web page that lists how
each vocal anti-evolutionist stands on that issue?
> > I personally don't consider a
> > monster with ferocious looking teeth to be a "bird". At best I'd
> > consider it to be some kind of chimera between a bird and some
> > crocodile-like monster, like a gryphon in some sense.
> They aren't really all that ferocious. In fact, they're tiny, and
> missing from the forward third of the mouth.
Ah, now we know why they all died (they can't produce a fossil until
*after* they die; there's no such thing as a fossil of a living
animal): They were so ferocious that they lost all their front teeth in
attacks, after which they starved to death because that's the only way
they knew to catch any prey, or they desperately tried other tactics to
catch prey, none of which worked, and one of those attempts resulted in
a fatal accident. Whatever the proximate cause of death, ultimately it
was all because of lack of front teeth. :-)
> Do you know at all what creationists mean by "kind"? It's just a term
> for groups that share ancestry within the group but not outside the
> group. Of course it's pretty clear that by this definition there is
> only one kind, but creationists don't see that.
Yeah, they're pretty dumb in that way. I suppose we can tease them back
after they way they sometimes tease us. They pretend to know something
(about salvation) which we don't know. We can tease them that we know
something they don't know. After holding out on them for months to
build up the suspense, finally we reveal that according to the
transitive definition of "kind", all eukaryotes are of one "kind".
> A creationist would claim that around 6000 years ago, if you traced
> this all back, you would come to a stopping point before which there
> are no ancestors at all.
No, only a Young Earth Creationist would say that. An OEC would have a
different story to tell, which includes ancestors back 3.5 billion
years with evolution by some supernatural process such as successive
creation (evolution of design rather than physical evolution) or
supernatural tinkering (analagous either to breeding i.e. artificial
selection, or to genetic modification i.e. artificial mutation). I
think it would be fun to find one well-known anti-evolutionist who has
taken a stand in favor of one of those four theories (YEC, OE-suc-C,
OE-art-sel, OE-art-mut), and start a petition to force that one theory
to be taught in schools, and get that one well-known person to sign
that particular petition, and then present them at random to undecided
people such as Behe and have the ardent supporter begging Behe to sign
that particular petition and getting mad if he refuses, then after he
signs one have the other three get mad at him. Also, hold public
debates of two out of the four to decide which is the better theory,
and invite Behe and two other undecideds to judge the debate. Encourage
the controversy!!! Enjoy any fisticuffs that may ensue.
> > But the Ark was just something in a story, not anything which really
> > existed. The whole Flood story was a fiction.
> Not to creationists.
That's one more item for "I know something you don't know, nya nya nya".
But I suggest we all attack anti-evolutionists by the following conversation:
We: Why is it bright in the day but dark at night?
YEC: Because of the Sun.
We: What does the Sun have to do with it?
YEC: The light during the day comes from the Sun.
We: How can that be true? How can light come from a Sun that wasn't even there?
YEC: What do you mean? There can't be any daylight without the Sun.
We: According to Genesis, the light comes all by itself, without any Sun.
On the first day the Lord said "Let there be light",
and later the same day the Lord divided the light from the dark,
and called the light "day" and the dark "night",
but He didn't make the Sun until the fourth day.
For three full days there was a day/night cycle without any Sun.
On the fourth day, he never said "Let the light during the day
that formerly just happened without the Sun, stop just happening
and instead come from the Sun from now on", so obviously
the light during the day *still* doesn't come from the Sun,
just the same as it did before the Sun was ever created.
YEC: But the light during the day *does* come from the Sun.
We: Then Genesis is wrong about that. I wonder what else Genesis is wrong about.
Anybody want to give it a try and report back how the conversation went?
.
John Harshman
>>There are other points of contention among creationists too. Homo
>>erectus is variously described as "just an ape" or "just a modern
>>human", depending on who or when you read.
>
>
> Has anybody collected such data and posted a Web page that lists how
> each vocal anti-evolutionist stands on that issue?
I don't know.
[snip]
>>Do you know at all what creationists mean by "kind"? It's just a term
>>for groups that share ancestry within the group but not outside the
>>group. Of course it's pretty clear that by this definition there is
>>only one kind, but creationists don't see that.
>
> Yeah, they're pretty dumb in that way. I suppose we can tease them back
> after they way they sometimes tease us. They pretend to know something
> (about salvation) which we don't know. We can tease them that we know
> something they don't know. After holding out on them for months to
> build up the suspense, finally we reveal that according to the
> transitive definition of "kind", all eukaryotes are of one "kind".
Except that you couldn't do it with living species, and to do it with
extinct species you have to assume evolution. Good try.
>>A creationist would claim that around 6000 years ago, if you traced
>>this all back, you would come to a stopping point before which there
>>are no ancestors at all.
>
> No, only a Young Earth Creationist would say that. An OEC would have a
> different story to tell, which includes ancestors
Ancestors? I think not, unless you have some private definition for that
word too.
[snip]
There have been such conversations here. You really aren't going to trip
up any creationists with this sort of thing.
Jon Fleming
>> There are other points of contention among creationists too. Homo
>> erectus is variously described as "just an ape" or "just a modern
>> human", depending on who or when you read.
>
>Has anybody collected such data and posted a Web page that lists how
>each vocal anti-evolutionist stands on that issue?
<http://www.talkorigins.org/faqs/homs/compare.html>
<snip>
--
jrf
replace nospam with group to email
Joe Cooper
different story to tell, which includes ancestors back 3.5 billion
years with evolution by some supernatural process such as successive
creation (evolution of design rather than physical evolution) or
supernatural tinkering (analagous either to breeding i.e. artificial
selection, or to genetic modification i.e. artificial mutation). I
think it would be fun to find one well-known anti-evolutionist who has
taken a stand in favor of one of those four theories (YEC, OE-suc-C,
"OE-art-sel, OE-art-mut), and start a petition to force that one theory
to be taught in schools, and get that one well-known person to sign
that particular petition, and then present them at random to undecided
people such as Behe and have the ardent supporter begging Behe to sign
that particular petition and getting mad if he refuses, then after he
signs one have the other three get mad at him. Also, hold public
debates of two out of the four to decide which is the better theory,
and invite Behe and two other undecideds to judge the debate. Encourage
the controversy!!! Enjoy any fisticuffs that may ensue. "
Isnt there a law against cruelty against (insert your choice) ?
Bob D
an...@sci.sci wrote:
> But I suggest we all attack anti-evolutionists by the following conversation:
> We: Why is it bright in the day but dark at night?
> YEC: Because of the Sun.
> We: What does the Sun have to do with it?
> YEC: The light during the day comes from the Sun.
> We: How can that be true? How can light come from a Sun that wasn't even there?
> YEC: What do you mean? There can't be any daylight without the Sun.
> We: According to Genesis, the light comes all by itself, without any Sun.
> On the first day the Lord said "Let there be light",
> and later the same day the Lord divided the light from the dark,
> and called the light "day" and the dark "night",
> but He didn't make the Sun until the fourth day.
> For three full days there was a day/night cycle without any Sun.
> On the fourth day, he never said "Let the light during the day
> that formerly just happened without the Sun, stop just happening
> and instead come from the Sun from now on", so obviously
> the light during the day *still* doesn't come from the Sun,
> just the same as it did before the Sun was ever created.
> YEC: But the light during the day *does* come from the Sun.
> We: Then Genesis is wrong about that. I wonder what else Genesis is wrong about.
>
Nice one. this is so obvious but it hadn't occurred to me.
My Bible says "And there was evening and there was morning, one day
.... a second day .... a third day." It is quite clear.
Augray
>> Then why did you write "Two species are either related by common
>> descent, or not"?
>
>Because that's the only way to satisfy what you claimed earlier in the
>thread, that the mere fact of having a common ancestor provides a
>metric.
I never claimed that the mere fact of having a common ancestor
provides a metric. In news:7h58q1517qa2k0v89...@4ax.com
I agreed with John Harshman's clarification of an earlier statement of
mine when he wrote
You mean relative recency of common ancestry, i.e the clade
composed of all animals that share a more recent common ancestor
with Tyrannoaurus than with modern birds.
Note that there is more to this metric than mere common ancestry.
>My point is that such existance of common ancestor provides at
>best a boolean metric, all or nothing, related or not related, zero
>distance or infinite distance, which is completely useless in deciding
>which tree to draw to span the various taxa.
Surprisingly, cladistics doesn't assume common ancestry. Common
ancestry is a conclusion, not a premise.
>For all the taxa of interest within a single domain, it's highly likely
>that any two share a common ancestor, so per the all-or-nothing metric
>they would all be zero distance from each other.
What is a "domain", and how does one recognize one?
>> >> Not many creationists claim personal revelation, yet all would say
>> >> that Archaeopteryx was a "bird".
>> >On what basis would they make such a statement?
>> The characteristics of the animal.
>
>That's only the "minor premise" in the logic. What is the major
>premise, i.e. the definition of "bird" by which they judge some animals
>as "bird" and some others as "not bird"??
It's mostly the presence of feathers, although you'll see other
characteristics put forth as well. Creationist emphasize (or
misrepresent) the commonalities, and ignore the differences.
>I personally don't consider a
>monster with ferocious looking teeth to be a "bird". At best I'd
>consider it to be some kind of chimera between a bird and some
>crocodile-like monster, like a gryphon in some sense.
>
>> So, taxonomic classification is vacuous?
>
>No, but treating Archaeopteryx as if it were identical to regular
>(modern) birds would be an absurd way to classify animals.
Why? It all depends upon one's method of classification.
>Treating
>Archaeopteryx as transitional between some ancient group (such as
>dinosaurs) and birds would make a lot more sense.
Remember, we're talking about creationists here.
>Given that
>Archaeopteryx indeed lived and went extinct appx. 150 million years
>ago, trying to cram it into a fixed set of modern taxa seems a bad
>policy.
That all depends on how one recognizes "modern taxa".
>In most cases the last common ancestor of two modern groups (and any
>side group very similar to that LCA) should be considered a three-way
>transitional, between the two modern groups and also an earlier
>ancestal form. For example, the LCA between modern reptiles and modern
>mammals might be considered a transitional between reptiles and mammals
>and ancient amphibians.
>
>
>> >> >I believe a cladogram on a Creationist Web site would be an oxymoron.
>> >> Because?
>> >
>> >Because a Creationist doesn't believe there's any clade that includes
>> >more than one species. Like begats like, same species, never a
>> >different species, per Creationist belief.
>>
>> This is incorrect. See
>> http://www.answersingenesis.org/creation/v22/i3/ligers_wolphins.asp
>
> If two animals or two plants can hybridize (at least enough to produce
> a truly fertilized egg), then they must belong to (i.e. have descended
> from) the same original created kind.
>
>So if A can hybridize with B, and B can hybridize with C, but A cannot
>hybridize with C, does that mean A and B are of the same kind, and B
>and C are of the same kind, but A and C are not of the same kind?
>Whatever sense can that have?? None as far as I can tell.
Nevertheless, it falsifies your claim that "a Creationist doesn't
believe there's any clade that includes more than one species".
> In the case of three species, A, B and C, if A and B can each
> hybridize with C, then it suggests that all three are of the same
> created kind -- whether or not A and B can hybridize with each other.
>
>Then by that logic, if there's a chain of pairwise hybridizable links
>A-B-C-...-X-Y-Z then all of A thru Z are of the same kind? Suppose we
>find that I can potentially mate with somebody five years older than
>myself, and that person can potentially mate with somebody five years
>older still, etc. by five-year links all the way from myself today to
>some distant ancestor 400 million years ago. So by that logic, I and
>that distant incestor are of the same kind? And likewise a parakeet
>could mate with another parakeet a year or two older, and by 1-2 year
>links that chain could also be carried back to 400 million years ago,
>in fact to the common ancestor of all birds and all mammals. So all
>birds and mammals are of the same kind? Perhaps all eukaryotes are of
>the same kind via that transitive definition. Of what value is such a
>definition?
>
> it is
> overwhelmingly likely that horses, asses and zebras (six species of
> Equus) are the descendants of the one created kind which left the Ark.
>
>But the Ark was just something in a story, not anything which really
>existed. The whole Flood story was a fiction.
Nevertheless, it falsifies your claim that "a Creationist doesn't
an...@sci.sci
> >each vocal anti-evolutionist stands on that issue?
> <http://www.talkorigins.org/faqs/homs/compare.html>
Let me take a look:
Specimen Cuozzo
(1998) Gish
(1985) Mehlert
(1996) Bowden
(1981)
Menton
(1988)
Taylor
(1992)
Gish
(1979) Baker
(1976)
Taylor
and Van
Bebber
(1995) Taylor
(1996)
Lubenow
(1992)
ER 1813 ER 1813
(510 cc) Ape Ape Ape Ape Ape Ape
Java Man Java
(940 cc) Ape Ape Human Ape Ape Human
Peking Man Peking
(915-
1225 cc) Ape Ape Human Ape Human Human
ER 1470 ER 1470
(750 cc) Ape Ape Ape Human Human Human
ER 3733 ER 3733
(850 cc) Ape Human Human Human Human Human
WT 15000 WT 15000
(880 cc) Ape Human Human Human Human Human
Sorry, that layout is too difficult for me to understand.
Maybe I'll look at it again another day and try to figure it out.
Maybe if I look at the HTML source it'll give me a clue how it's
supposed to be organized, but not tonight, bedtime now.
.
Augray
Are you by chance (and this is a serious question) visually impaired?
Because to me, the table seems pretty straight forward, although I can
imagine that the table related tags in html which are used on that
page might cause confusion. Here's a similar representation of the
same data:
| | | | Bowden | |
| | | | (1981) | |
| | | | Menton | Baker |
| | | | (1988) | (1976) |
| | | | Taylor | Taylor | Taylor
| | | | (1992) | and Van | (1996)
| Cuozzo | Gish | Mehlert | Gish | Bebber | Lubenow
Specimen | (1998) | (1985) | (1996) | (1979) | (1995) | (1992)
------------------------------------------------------------------
ER 1813 | | | | | |
(510 cc) | Ape | Ape | Ape | Ape | Ape | Ape
------------------------------------------------------------------
Java | | | | | |
(940 cc) | Ape | Ape | Human | Ape | Ape | Human
------------------------------------------------------------------
Peking | | | | | |
Man | | | | | |
(915- | Ape | Ape | Human | Ape | Human | Human
1225 cc) | | | | | |
------------------------------------------------------------------
ER 1470 | | | | | |
(750 cc) | Ape | Ape | Ape | Human | Human | Human
------------------------------------------------------------------
ER 3733 | | | | | |
(850 cc) | Ape | Human | Human | Human | Human | Human
------------------------------------------------------------------
an...@sci.sci
Thank you for the compliment and agreement.
I'm actually surprised I didn't notice it myself years ago.
> My Bible says "And there was evening and there was morning, one day
> .... a second day .... a third day." It is quite clear.
Yes, a clear description of three complete day/night cycles of light
before the Sun was created. The ancient Hebrews themselves had lots of
experience out in the "elements" with their animals, and might have
been bright (pun) enough to figure out that maybe that intensely bright
and hot thing was the source of the diffuse light called "day". If they
had made up their own myth they might have come up with some story that
was consistent on that point. But no, they followed the Babylonian or
other earlier myths, *buzz*, plagiarism loses in this case.
In any case it shows not only that the myth of Moses being a real
person who got those stories from the Creator was a bald-faced lie, but
whoever canonized those stories wasn't even bright to notice the
contradiction and fix it for posterity, and the many re-tellers prior
to canonization weren't very bright either. In all those generations,
not one person was able to fix that obvious contradiction.
I can understand that if two myths are passed down by separate
families, or written by different witnesses, and then these stories
from various sources are later compiled into a book, such as the two
chapters of Genesis that deal with the Creation, or the two Gospels
that deal with the birth of Jesus, the compiler of the anthology might
not know which is correct, so might just include both to avoid losing
whichever later turns out to be the right one. But to have a logical
progression of verses within a single coherent story be
self-contradictory like that is absurd. How can I, after 3000 years, be
the first to have noticed that stupidity?? Oh well, it took until
Galileo before anybody noticed that when you throw something it doesn't
go straight out and then at some point suddenly stop and fall straight
down, but instead it follows a smooth curve whereby there is no
particular point at which it switches from going-out to falling-down
behavior. And it took until Darwin&Wallace before anybody figured out
that there's no clear boundary between one species and the next, either
in time (Homo erectus->sapiens) or at a single time in the midst of a
speciation event (Chuman, then Chimp-clade of chuman and Lucy-clade of
chuman single species, then Chimpanzieman and Humanlucyp ring-species,
then Chimpanzie and LucyHuman finally separate). And it took until
Einstein or later before anybody figured out that most of that randomly
flickering stuff you see when your eyes are closed or at night is
really there, individual quantized photons of light at low sampling
rates where the receptors in your retna are reporting each individual
photon to your brain much as a Geiger counter reports each individual
ionizing event as a "click" when the overall rate is low.
Hey, did anybody remember to celebrate Wallace's birthday four days ago?
Sorry, I forgot until now.
.
Sam
heard was at the time early compilations were made and the different sources
merged, those doing so didn't want to seperate groups of people who had
slightly different recorded versions of basically the same story, so they
were both included in the compilation to satisfy both groups/sources.
Leaving one or the other out would have served to isolate people, where the
point was trying to bring them together. Much of the research into biblical
origins correlates styles/wording/etc between many of these duplicate
tellings and has generally seperated and attributed much of the Old
Tesitiment to four different authors. Of course the die hard Christian
continues to believe it was dictated to Moses.
<an...@sci.sci> wrote in message
news:7436b$43c68f97$c690c02a$11...@TSOFT.COM...
an...@sci.sci
> | | | | Bowden | |
> | | | | (1981) | |
> | | | | Menton | Baker |
> | | | | (1988) | (1976) |
> WT 15000 | | | | | |
> (880 cc) | Ape | Human | Human | Human | Human | Human
Ah, yes, that's much easier to understand. Thanks.
I'm curious. Did you format all that data manually (perhaps using a
text editor with macros), or do you have a program that automatically
converts the HTML syntax to that format?
So any chance of somebody sponsoring a public debate between some of
those people regarding any of those disputed opinions?
.
Augray
>> Here's a similar representation of the same data:
>> | | | | Bowden | |
>> | | | | (1981) | |
>> | | | | Menton | Baker |
>> | | | | (1988) | (1976) |
>...
>> WT 15000 | | | | | |
>> (880 cc) | Ape | Human | Human | Human | Human | Human
>
>Ah, yes, that's much easier to understand. Thanks.
>
>I'm curious. Did you format all that data manually (perhaps using a
>text editor with macros), or do you have a program that automatically
>converts the HTML syntax to that format?
I did it manually.
>So any chance of somebody sponsoring a public debate between some of
>those people regarding any of those disputed opinions?
Don't hold your breath. Such a debate would force the participants to
actually confront the evidence.
an...@sci.sci
> explanation I've heard was at the time early compilations were made and
> the different sources merged, those doing so didn't want to seperate
> groups of people who had slightly different recorded versions of
> basically the same story, so they were both included in the compilation
> to satisfy both groups/sources. Leaving one or the other out would have
> served to isolate people, where the point was trying to bring them
> together. Much of the research into biblical origins correlates
> styles/wording/etc between many of these duplicate tellings and has
> generally seperated and attributed much of the Old Tesitiment to four
> different authors.
That explains why different books, or maybe even different chapters
within one book, would conflict with each other. But that's no excuse
for Genesis 1, a sequential narrative, obviously composed by a single
person, yet self-contradictory from one stanza to another.
That Genesis 1 seems like a cross between the "telephone" game (from
Art Linkletter's TV program) and the infinite-add-on thread on Usenet
newsgroups. Imagine one person writes the first sentence or paragraph
of a story. Second person adds a second such part consistent with the
first part. Third person sees *only* the second part, not the first
part, and tries to add a consistent third part. Etc. Each nth person
sees only the (n-1)th part and tries to add the nth part consistent
with it. Finally all parts are shown together, for a big laugh.
.