Re: Part 1 (of 3): What are major aspects of evolutionary theory?

> > That's right. With no data except the character matrix or DNA
> > sequences, you get an unrooted tree, but there's no way to know where
> > to root it. You need some *other* kind of data to root it.
> Which is what I've been saying all along, ...

We're now in generic agreement. Next is to find out *what* kind of
other data is sufficient to root the tree. You basically say
timestamped fossils are useless. I disagree.

So-far we have discussed duplicated segments of DNA, whose gradual
drift away from each other after duplication shows the arrow of time,
but that works only for very recent duplications where we have the DNA
sequences of ancestors and/or several very near modern
individuals/species which can be used to reconstruct the genomes at the
presumed branch nodes.

> > If you don't know where the tree is rooted, if
> > all branches are unknown-direction, how do you magically know which way
> > a trait evolved?
> Yes, the best way is by looking at its evolution on a rooted tree.
(Which is a circular argument if you're trying to root the tree in the
first place.)

> Nevertheless, there are other ways used.
(Which I'm still waiting for you to state specifically in non-circular
argument.)

> (Weston 1988) states that:- if one character state is possessed by
> all of the taxa that also possess the alternative state

By definition that's impossible. If two states are alternatives, a
species/individual has one or the other, not both.

> For example, open gill slits are possessed by all chordates in at
> least the embryonic stage, but in tetrapod chordates these close early
> in development; thus, all chordates possess open gill slits but only
> some possess both open gill slits (early in development) and closed
> gill slits (later in development). Consequently, possession of closed
> gill slits is hypothesised to be derived relative to possession of open
> gill slits.

I don't accept that argument. When I open a door, the door closes after
me. If I open a door and grab something to hold the door open, it stays
open. The door that closes after being opened is ancestral to the door
that stays open indefinitely.

I can believe that first a mutation caused gill slits to open but not
maintain that condition while the environment changed, so after a while
the changed environment interferred with the gill-open mechanism,
causing them to close. But a second mutation made the gill-open
mecnanism resistant to the change in environment, allowing them to stay
open longer.

For example, the initial gill-open mechanism might have been dependent
on some hormone that is released for some *other* purpose early during
development, and when that *other* purpose is gone the hormone goes
away and the gill-open mechanism begins to fail because of lack of that
hormone. Later the gene for the hormone duplicates, yielding one
version that is expressed only temporarily, and has only the *other*
effect, and a second version that is expressed permanently, which has
only the *gill*open* effect, separating the two effects so each can be
regulated differently.

For all we know, gill slits arose originally in some long-extinct
ancestor that had a growth spurt early during development, which
required extra oxygen, hence the value in having gills temporarily
during that growth spurt, then relying on simple epidermis osmosis for
oxygen input and urine discharge for CO2 discharge during the
relatively low-metabolism adult stage which consisted mostly of sitting
totally silent for days or weeks at a time waiting for prey to wander
close enough to snatch. Later mutations in muscle supported constant
swimming during adulthood, which required more oxygen throughout the
adult time, which favored the chance gene duplication allowing gill
slit regulation to be separated from the growth spurt. Later
development of lungs rendered gills unnecessary during adult life,
allowing reversion back to the temporary-gill state in amphibians, and
eventually a new state of gill slits without actual gills in reptile
embryos.

I agree that in fact the active gills in adult fishlike ancestors came
before temporary gills or mere temporary gill slits in tetrapods. But I
don't agree with the logic to "prove" that above.

> Weston (1994) has pointed out that the direct method can be applied
> to gene duplications (paralogy) to polarise the a and b subunits of
> ATPase based on taxa from the archaebacteria, eubacteria and
> eukaryotes, thus providing a "root" for the tree of life."

Now we're back to gene duplications followed by subsequent divergence
of the two copies from each other. Is there a nice Web page that
summarizes this particular set of data and resultant conclusion,
showing precisely what the root of the Universal Ancestry Tree and
nearby nodes look like, at least in regard to this particular set of
homologous genes?

Are there any other sets of homologous genes that can be absolutely
rooted in the same way very early before the three domains were fully
formed? It would be interesting to have five or ten such completely
independent examples of very early gene duplication, with hereditory
tree for each, and then see whether they are all aligned the same way
(indicating a whole set of unrelated gene groups following exaclty the
same tree, indicating perhaps whole cells containing these genomes
evolved as whole-cell units), or whether the trees of different gene
groups don't run parallel, indicating that some of their evolution
occurred separately and *later* merged somehow into the whole-cell
clades we observe nowadays.

Um, slight problem: With only three domains of life, there is only one
possible unrooted tree, and all three unrooted trees are satisfied by
that one unrooted tree, so even if all three different unrooted trees
applied to various homologous-gene-duplication groups, there'd be no
way to check anything based on modern DNA evidence. Or am I wrong?
You expert, me confused, please enlighten, kimosabe. Burma Shave? :-)

> This is a form of the ontogenetic criterion. There are arguments
> against it.

Like the one I gave above for gill slits?

> In practice, polarity is almost always determined after the fact,
> i.e. after rooting the tree by outgroup.

Which again begs the question how to know which is the outgroup of an
unrooted tree. Around and around we go, never getting out of the
circular argument: X is a clade, and y is not within X, therefore
y is an outgroup with respect to X, therefore the unrooted tree
of X+y should be rooted between X and y, therefore X is a clade.

> all the other non-outgroup criteria are not commonly used.

So you're saying the only way to root a tree is by circular argument?
Are you a troll??

> Like I've been saying all along, you have to bring additional information.

Like I've been asking all along, what additional information (other
than begging the question with a circular argument of presuming the
outgroup from the start) would be useful for rooting what was initially
an unrooted DNA-sequence and/or character tree?

> you can in fact test your original suspicions by bringing in more
> taxa. If, for example, you root the human tree by using a chimp

OK, at this point I have:
+---Chimp
--+
+---Human

> you can test this by including a gorilla in a new analysis.

How does that test anything? Three trees are equally possible at this point:
+--Gorilla
+--+
--+ +--Chimp
|
+---Human
and the other two you probably can guess.

> If humans still form a single group relative to the chimp when the
> tree is rooted on a gorilla

Huh? The tree is rooted on a branch between two clades, not on a
single-species clade. If one of three species were fossil, not living,
then possibly that fossil species could be the common ancestor of the
living species. But with all three living, you can't root on any of the
single species. Maybe that's not what you meant by "root on"??

> the chimp as outgroup is confirmed.

That makes no sense. With only two *living* species, each is an
outgroup with respect to the other. With three species, the new species
can join an already-existing branch of the tree, as it joined the Chimp
branch in my example above to form a Chimp+Gorilla branch (or as it
might join human to form a Human+Gorilla branch), or the new species
might be the outgroup with respect to the two original species. You
can't just *say* that any new species *must* be an outgroup with
respect to the species you started with. For example, if you start with
Chimp and Human, and then add Neandertal, it would be wrong to assume
Neandertal was outgroup with respect to Chimp+Human.

> Still don't trust that?

I trust that you now have three possible rooted trees with three
species, and have no idea which is correct.

> Bring in more taxa.

Each new taxon could be part of a branch already in the tree, or a new
outgroup compared to everything preceding it. Consider for example if
species were added in the sequence that we obtained their full genome.
We start with E coli and H sapiens, then we add the mustard plant. Is
mustard plant an outgroup with respect to E coli and H sapiens or not?
Then we add a yeast. Then we add Chimpanzee. At each point, is the new
species an outgroup or not with respect to what we already had?
It begs the question to say that we're starting with the tightest clade
and then adding only outgroups at each stage.

> If you bring in all the primates and still the humans form a single
> group, then either all other primates are descendants of humans or
> humans are a clade.

Huh?? We never disputed that each living-today species is a clade,
disjoint from any other living-today species. That has nothing to do
with rooting the tree of primates, which must be along some branch
connecting two clades of species, not within a single living-today
species.

> Now, you may consider both these hypotheses equally sensible without
> fossil evidence. But I would not.

No, what I consider equally sensible is that the root lies between
humans and all the other species, or that the root lies between African
apes and all the other species, or that the root lies between apes and
all the other species, etc. Whereever we break the unrooted tree into
two pseudo-clades, the true root must lie at the break point (in which
case both are true clades) or within one of the two pseudo-clades (in
which case the other is a true clade). So given any place we break the
unrooted tree into two halves (to place a supposed root), at least one
of the halves of the tree must be a true clade. So we're always
breaking the tree between some true clade and "all the other species",
hence the apparent monoteny of wording of my above partial list of
alternatives.

At this point in my reply I was going to find a cladogram of all
primates on the Web, then take away the root and re-arrange it to a
nice symmetrical unrooted tree, and challenge you to state what
evidence (not authority) you have to show where the root belongs. But I
couldn't find any such cladogram. I spent a couple hours researching
the question, starting with various tolweb pages as a skeleton
cladogram, then filling in details from other sources but getting
totally stuck, no cladograms shown anywhere for most of primates. I
then resorted to reading English descriptions of the various taxa, and
manually building each tiny part of the tree from that descriptive
text, such as "the sub-order ... consists of 3 infra-orders, namely ...
which has ..., ... which has ..., etc." Several descriptive taxa which
appeared in tolweb didn't appear in any of the descriptions, so I had
to look them up individually, but finally I got the complete tree all
put together. Please check it. Note that several nodes aren't fully
resolved. Is that the best cladistic data we have currently, or have
the nodes been fully resolved and I just haven't been able to find that
data?

--Primates
|--Strepsirhini (non-tarsier prosimians)
| |--Lemuriformes
| | |--Cheirogaleoidea
| | | `--Cheirogaleidae (dwarf and mouse lemurs)
| | `--Lemuroidea
| | |--Lemuridae (true lemurs)
| | |--Lepilemuridae (sportive lemurs, weasel lemurs)
| | |--Palaeopropithecidae
| | `--Indridae (woolly lemurs, sifakas)
| |--Chiromyiformes
| | `--Daubentoniidae (Aye-aye)
| `--Lorisiformes
| |--Lorisidae (lorises, pottos)
| `--Galagidae (galagos, Bush Babies)
`--Haplorhini
|--Tarsiiformes (tarsiers)
`--Anthropoidea
|--Platyrrhini (New World monkeys)
| |--Cebidae (monkeys)
| `--Callitrichidae (tamarins and marmosets)
`--Catarrhini (Old World monkeys, apes and humans)
|--Cercopithecidae
`--Hominoidea (apes)
|--Hylobatidae (gibbons and siamangs, total appx. 11 species)
| |--Hylobates agilis (Agile Gibbon)
| |--Hylobates concolor (Black Gibbon)
| |--Hylobates hoolock (Hoolock Gibbon)
| |--Hylobates klossii (Kloss' Gibbon)
| |--Hylobates lar (White-handed Gibbon)
| |--Hylobates moloch (Moloch Gibbon)
| |--Hylobates muelleri (Mueller's Gibbon)
| |--Hylobates pileatus (Pileated Gibbon)
| `--Hylobates syndactylus (Siamang)
`--Hominidae
|--Pongo pygmaeus (orangutans)
`--African Apes
|--Gorilla gorilla (gorillas)
`--chuman
|--humans (Neandertal et al in this clade)
`--chimps
|--Pan troglodytes (chimpanzees)
`--Pan paniscus (bonobos, aka as pygmy chimpanzees)
..

.

Quantcast