Richard Wein wrote:
>>>You're still evading my question of whether we ever see heritable
>>>Siamese-twinning in humans. If not, then your claim that we see this
>>>mechanism in humans is false.
This is a triviality to the theory in question, something I would happily
scratch from the article if I were convinced you were right about this rare
phenomenon. The big point is simply that Siamese-twinning is a mechanism
for generating morphological complexity, a mechanism that we can understand.
>>I'm unaware of any theoretical causes of Siamese-twinning in humans
>>which are not heritable. It's a morphological phenomenon, and morphology
>>is heritable.
>
>This comment, if you really mean it, shows a profound ignorance of biology.
>I suggest you read a basic text on genetics and inheritance. Morphology per
>se is certainly *not* heritable. Our morpholgy is determined by both genes
>and environment ("nature and nurture"). Of these, only genes are inherited
>(though our parents also play a role in our environment).
Maybe I was using inappropriate short-cut terminology when I said
morphology was heritable. From now on, If someone says I have my
uncle's nose, I will upbraid him for his ignorance. I don't have my uncle's
nose at all, I merely have the genes for it.
>I spent a few hours today browsing the web for info on Siamese ("conjoint")
>twins. Duplication by Siamese twinning is not heritable. The twins do not
>receive two copies of the genome. Their genome is normal. Siamese twinning
>is caused by environmental factors which result in a partial splitting of
>the embryo.
Embryos of all kinds can split and produce multiple births of normal
individuals. Siamese-twinning is not caused so much a result of the
splitting as of the failure of the multiple embryos to fully differentiate.
I don't know where you get the "two copies of the genome" idea; nobody
is claiming that Siamese twins are polyploid.
>For example:
>"If the notochord reaches an area of hypoxia, an area with low oxygen
>content, the notochord will split to go to higher area of oxygen
>concentration. This will cause you to have two notochords. Thus you will
>induce two neural tubes. At term, the embryo will have two heads, two
>brains, two spinal cords, one body. This is called a conjoint twin."
>(http://www.musc.edu/related/comsc/embryo4.htm)
I think there must be an infinite number of mechanical things that could
go wrong with embryonic development to cause such multiplication, and an
infinite number of mechanical problems that would cause differentiation
to be incomplete. I can't believe that none of these things could be caused
by defective genes.
>[In any case, if Siamese twinning was heritable, that would torpedo your
>argument against Dawkinsian duplication. A heritable partial Siamese
>twinning, where only some organs/limbs are duplicated, would be a type of
>Dawkinsian duplication in a modern organism!]
I often refer to Siamese-twinning as a 'crude' mutation. The point is that
a partial Siamese twin consisting of, for example, one vertebra,
positioned within its sibling's vertebral column, is unlikely enough to
be dismissed.
>And I come back to the point that I made earlier. Partial Siamese twinning
>shows that a simple chance disruption during development can result in
>functional duplicate limbs and organs, even in an organism as complex as a
>human. If this can happen when the disruption is by chance, why can't it
>also happen when the disruption is due to a simple mutation?
Duplicate limbs and groups of organs, fine. Gradual serial accretion of new
segments within sets of homologs, no way.
>BTW here's the best web page I've been able to find on Siamese twinning:
>http://www.thefetus.net/sections/articles/Multiple_gestations/Conjoined_twin
>s_structural_anomalies_Jaffe.html
Interesting that you would do so much work on an idea you don't buy. But
it is worthwhile. Teratology definitely has evolutionary clues.
>>>A paleontologist can study fossilized phenotypes. But you have no fossils
>>>of
>>>your proposed organisms. If you can't propose a mechanism by which these
>>>organisms could have evolved, then why should anyone believe that they
>>>existed?
>>I explicitly describe the mechanisms by which the theoretical progenitor
>>comes to be and mutates into the variety of Cambrian fauna. I draw
>>diagrams. What else can I do?
>
>You don't describe the mechanisms. You describe only the morphology. The
>mechanisms of evolution are primarily random mutation and natural
>selection. You hardly address these vital issues.
The genetic aspects of the proposed mutations are trivial and obvious.
Siamese-twinning, a familiar phenomenon, followed by reduction and distortion
through RM&NS, a venerable general principle. You could talk genetics till the
end of time, but what we want to know about is morphology, what we would see
if we had a time machine and could observe evolution.
>>No fossils. I have to agree with Homer Smith that vertebrate origins is
>>a topic for theorists. Otherwise, reconstruction of evolution is just a
>matter
>>of playing with animal figures, arranging them in order to suit yourself.
>>
>>>You haven't answered my point about how symmetry is *maintained* during
>>>evolution of code duplicates, without requiring coincidental identical
>>>mutations in each duplicate.
>>
>>There is so much symmetry within a vertebrate body, I don't see how you
>>can entertain the idea that maintaining it requires unacceptable
>>coincidences. Anyway, your genetic criticism of this morphological theory is
>>underwhelming because you don't even know whether segments are generated
>>from one code segment repeatedly processed, or from many separate code
>>segments being processed.
>You're missing my point. The point of this part of my argument (the context
>has now been lost) was to establish that some segments of your early forms
>must have been generated from one code segment repeatedly processed. Do you
>now agree with that?
I have no problem agreeing with that, but it doesn't seem crucial to my
morphological claims. I really don't know whether my feet differ from my hands
because they are built from code sequences that have been distinct since the
early Cambrian, or whether they are built from the same code and differ due to
the effects of genes that evolved later, genes that regulate development.
>>>>This phenomenon can be viewed as the
>>>>action of genes which regulate the developmental process. So, do I
>>>>admit 'control genes'? Yes, but when they cause mutations which
>>>>apparently have more segments than their parents, they are not
>>>>creating anything new; these are atavisms.
>>
>>>I wish you wouldn't keep stating your hypotheses as if they were facts!
>>
>>I could try to write like Darwin, always throwing in disingenuous
>>subjunctive expressions like "...might we not believe...?" but it would
>>be tiresome for all concerned.
>
>On the contrary, it's vital to distinguish hypotheses from facts, even if
>that means a little extra verbosity. What's really tiresome for me is having
>to keep checking whether your statements are hypotheses or facts.
That's called 'looking for errors', a pretty normal thing, I would think.
>You misunderstood me (refer to the original context). I'm saying that both
>your model and mine require the duplication of segments by Dawkinsian
>processes. But the segments duplicated in your model are less complex than
>the segments duplicated in mine.
Only whole bodies are duplicated in my model; these then become segments.
Why is a whole body less complex than an internal segment?
>>>Your model involves partial duplications too, unless your 4 proto-limbs
>>>were all added simultaneously.
>>
>>They are added simultaneously.
>
>This requires an astonishing degree of coincidence.
No, just Siamese-twinning. The 'limbs' are entire bodies at first.
>OK. I thought you were talking about organs. I have no problem with nearby
>muscles being co-opted into the formation of an eye. If you're not involving
>other organs, then I think we have a pretty similar idea of the evolution of
>the eye. By the way, I assume you are allowing duplication of
>light-sensitive cells and their nerve connections. (I just want to point out
>again that you do require Dawkinsian duplication of simple structures.)
There is an obvious yet ultimately perhaps difficult distinction in my
thinking,
a distinction between structures which may vary a lot in number and those
whose variation in number is limited by the principle of reduction and
distortion.
I focus on gross skeletal morphology and segmentation. I don't say anything
about cells, hairs, or other such 'uncounted' structures.
>>I only argue that complexes of segments came into being in various
>>patterns and combinations, and that some of these combinations were
>>successful.
>
>It's not enough to rely on random success. For changes to be retained and
>improved on, you need a selective advantage. It's not plausible that a
>relatively complex structure like your proto-vertebrate evolved by random
>mutation without natural selection.
I totally believe in RM&NS. NS of advantageous morphology.
>Of course, I don't know any more than you know how and when your fusion
>occurred.
>>I have all the duplication occurring in the early Cambrian through
>>Siamese-twinning of entire bodies, followed by reduction winnowing
>>out unnecessary structures. This conforms to the evidence; it fits the
>>recognized pattern of reduction and specialization.Your model allows
>>duplication of specific internal structures without duplication of the
>>whole organism. A difficult thing theoretically, and not supported by
>>the evidence.
>>
>>>You didn't address the point about *fusing* of organisms.
>>
>>At first the Siamese-twins are independent physiologically; while
>>enjoying the advantage of greater size and the potential for improved
>>locomotion, they can fuse at their leisure. I don't see this fusion as
>>difficult for creatures with identical genomes.
>
>You're talking about fusion of individual organs (not the whole organism)
>into a functioning single organ. This involves the same kind of difficulties
>as duplicating an organ.
Not at all. This is the well-understood process of gradual reduction and
distortion, a perfectly good Darwinian/Dawkinsian phenomenon.
>>Various configurations formed, and only a few succeeded. Developing
>>coordinated movement through gradual evolution doesn't seem problematic.
>>With these outriggers in place, any kind of fortuitous twitching might move
>>the creature faster than its competitors.
>
>Even though they're pointing in different directions? Get real!
Random mutations create variations, selection chooses the ones that work.
You don't think natural selection can pick out the structure that has the best
configuration? The range of possible directions a structure could point is
too infinite or something?
>Coming back to the subject of extra toes and fingers (polydactyly), I did
>some research. It's interesting that polydactyly is associated with over 30
>different congenital disorders, having a wide variety of effects
>(http://www.drgreene.com/970519.asp). One of these, Trisomy 13, results
>from the duplication of the whole of chromosome 13. This doesn't
>sound like reactivation of an atavistic gene.
Why not? Supernumerary segmentation is common in teratology.
But why don't we see wholly new kinds of segments going in new
directions? Why the limitation to elaboration along certain existing
axes in such duplicative ways? Where are the more free-ranging
mutations? Understanding this constraint on morphology is crucial.
If variations in phenotypes are a matter of regulation, of and limiting
and distorting the development of a primordial pattern, the constraint
is understandable. If evolution is simply Dawkinsian and open-ended,
there is no explanation for the apparent constraints.
>Furthermore, an additional thumb is possible
>(http://www.medmedia.com/o15/113.htm). Are you going to argue that an
>ancestor had two thumbs per hand? What's more, it isn't necessarily the
>whole thumb that's repeated, sometimes it's just part, or even just the tip.
>Yet all of these variations seem to be viable, i.e. they at least have a
>functioning circulatory system. It's odd how all these different atavistic
>genes function so well in conjunction with the other genes for a modern
>hand, which must be very different from the atavistic precursor hand.
Some of us would say it's interesting how these atavistic *structures*
function so well in conjunction with other *structures*. The primacy of
the gene is an idea of yours and Dawkins, not of mine. But why is it
odd that structures than once functioned together should do so again?
If a brand-new structure fit right in, that would be odd.
There's no point in talking about the recent development of the thumb.
It's just another digit, distorted in development in accordance with
regulatory factors, call them genes if you like. I would argue that
ancestors had many segmented appendicular radials, structures
ancestral to digits. So I guess you could say they had a hundred
thumbs on each limb.
--Cliff Lundberg ~ San Francisco ~ 415-648-0208 ~ cliff@cab.com
This archive was generated by hypermail 2b29 : Tue Jun 20 2000 - 04:06:21 EDT