From: Chris Cogan <ccogan@telepath.com>
>>Richard
>>Could you please clarify what you mean by Darwinism, and in what respect
you
>>disagree with it, as there seems to be a great deal of confusion about
this
>>term.
>
>Chris
>Well, I don't really have a good one, because Darwinism is not, strictly
>speaking, a well-defined concept, but I think it *would* exclude merging of
>two organisms (as you describe below). And, pure and perfect Darwinism is
>definitely gradualist, though Darwin has been said to have been reluctant
>to accept strict gradualism (which itself needs defining, in my opinion).
>But as Gould points out. gradualism is clearly not essential to the basic
>process of cumulative variation and selection.
If that variation is random (as you define it below), then gradualism (in
the sense that I define it) *is* essential, because substantial increases in
functional complexity within a single generation are too improbable. (I
should have written "functional complexity", instead of just "complexity",
since any old random collection of parts can be complex.)
Maybe my use of the word gradualism is only confusing the issue. I thought
it was useful to have a word to sum up the limitation on the kind of changes
that can be made by natural evolution in a single step, and there doesn't
seem to be any other word available.
>Richard
>>I think that few people use the term to mean Darwin's understanding of the
>>process of evolution, because we know now that Darwin got some things
wrong,
>>and, used in this sense, no-one is a Darwinist.
>>
>>For the record, I use Darwinism to mean the theory that evolution occurred
>>by purely natural (non-intelligent) processes, and that the processes
which
>>are responsible for the evolution of complexity are random variation and
>>natural selection.
>
>Chris
>With a suitable definition of "random" (i.e., random with respect to
>whether a variation is good for the survival and reproduction of the
>resulting organism (if any).
Yes, thank you. That's a very good point.
>>Richard
>>This definition of Darwinism implies gradualism, in the sense that large
>>increases in complexity cannot occur in a single generation. However, it's
>>important to note that we're talking here about genotypic complexity,
>>because most random variation occurs at the genotypic level, and a small
>>change in the genotype can result in a big increase in phenotypic
>>complexity. For example, an organism may have a complex gene which is
>>unexpressed and a simple mutation may "switch on" this gene. Also (Cliff
>>please note), the merging of two organisms by symbiosis (as is thought to
>>have occurred in the evolution of the eukaryotic cell) is not a violation
of
>>gradualism, because it does not involve the creation of new complexity,
but
>>simply the combination of existing complexity from two organisms into one.
>
>Chris
>Though it may not be a violation of gradualism, it certainly *does* create
>new complexity; the resulting new organism may be significantly more
>complex than either of the two original organisms, at least in principle.
Yes, more complex than *either* of them, but not significantly more complex
than *both* of them considered together. That was my point.
>I
>think it will generally only *work* when the new genome is only slightly
>more complex than the original, because of the improbability of two quite
>different genomes producing a biologically viable result.
>
>The complexity of the overall *situation* will not necessarily be more
>complex (or at least not much more), because, though each initial genome
>might be simpler than the resulting genome, they are *two* genomes. If we
>join the strings "aabababccde" and "ouioliljooo" we get
>"aabababccdeouioliljooo," which is more complex than the initial strings.
>But the initial strings as a pair are more complex than the resulting
>string *because* there are two strings.
Well, I'm not sure whether adding one string on to the end of the other is
an increase in their combined complexity, or a decrease, or no change. But
the point is that it's a small enough change to be considered gradual (by my
definition), and to not be too improbable.
>If the two initial cells were similar enough, then, though the resulting
>cell would still be more complex than either of the initial cells (assuming
>redundant genetic bits were cast out), it need not be *much* more complex.
>We might merge "abcde" with abcx" to produce "abcdx," for example (and
>"abcdx" is more complex than either of the original strings). This sort of
>merging would be compatible with gradualism. But, it would not necessarily
>be the case that all such mergings would produce such slightly more complex
>results.
But, in those which produced significantly more complexity, most of that
complexity would not be functional. (This apparent disgreement is probably
just a result of my failing to distinguish between mere "complexity" and
"functional complexity".)
Also, I think you're overlooking some possible scenarios.
1. Two cells might merge without combining their genomes. Cell A somehow
finds its way inside cell B. Cell A continues to divide. When Cell B
divides, each descendant may take with it one or more descendants of A. This
may continue indefinitely without any genomic change. (This is why I was
careful to refer to "random variation" and not "random mutation".)
Eventually, however, the genomes adapt to the situation. Note that, even in
the modern cell, mitochondria still have their own genomes.
I'm not familiar with the work of Margulis and others who've proposed the
symbiotic origin of eukaryotes, so I don't know if this scenario is anything
like theirs.
2. Combining of genomes doesn't necessarily mean some sort of complicated
interweaving of their DNA. It could be just a matter of adding the two
genomes side-by-side, as separate chromosomes, or attaching one to the end
of the other. Given relatively unstructured organisms, such as cells, it may
be possible to get a viable result in this way.
3. Parts of the genome can be transferred from one organism to the other.
IIRC it's well-established that genetic material from one organism is
sometimes inserted into the genome of another by natural causes (e.g. by
viruses?). And, of course, this has been done by genetic engineers too.
These transfers can add significant functional complexity to the target
genome, but, again, this is not the creation of new complexity, just a
transfer of complexity from one organism to another.
Richard Wein (Tich)
This archive was generated by hypermail 2b29 : Sun Aug 27 2000 - 05:13:12 EDT