Date: 23 Mar 94  17:23:35
   From: Tim Ikeda
     To: All
Subject: Is there a Denton FAQ?

From: timi@mendel.berkeley.edu (Tim Ikeda)
Organization: Plant Biology/MCB - UC Berkeley

In article <1994Mar22.211218.103764@ns1.cc.lehigh.edu>,
mjb1@ns1.cc.lehigh.edu (MICHAEL J. BEHE) wrote:

> Is there a FAQ somewhere on Michael Denton's "Evolution: A Theory
> in Crisis"?  I looked in the list of t.o FAQs and didn't see a
> title referring to that book. Please excuse me if this information
> is easily accessible. I'm new to the conversation.

If you have the chance, get Denton's book and skip right to Chapter 12
"A Biochemical Echo of Topology"; especially the part where he looks at
cytochrome C homologies.  It's quite bad.  I wouldn't want to spoil
anyone's enjoyment by giving away the "punchline" but a quick glance
at Denton's reproduction of a table from Dayhoff makes you ask
"Why didn't he compare pigeon sequences across the whole column?" Or,
"Why did he choose these particular groups for comparison and leave out
the others?"

For example, he notes that the differences between the bacterial
cytochrome sequence is about the same (about 65%) for horses, pigeons,
tuna, silkmoths, wheat and yeast.  He also notes that the differences
between the silkmoth's cyt. C sequence and that of horses, pigeons,
turtles, carp and lampreys are also similar (about 27%). Then he
asks: "So where are the intermediate forms?  Why don't we see them?"

Basically, I don't think this isn't all that surprising; organisms
that are outside of a particular group might be expected to show
greater sequence divergence. But I wonder why he doesn't make a big
thing about the following data which is in his table:

Differences from horses (Cyt C seq.)
pigeons:   11%
turtles:   11%
tuna:      18%
carp:      13%
lampreys:  15%
silkmoths: 27%
wheat:     41%
yeast:     42%
R. rubrum: 64%
  (C2 seq.)

When Denton talks of searching for intermediate groups it seems like he
expects sequence divergence to stop dead at the time of divergence.
I wonder if this is related to his requirement for unambiguous proof.
He writes (page 55): "To show that any two species of organism are
related in an evolutionary sense, to show for example that one species
A, is ancestral to B, ie A->B or that both species have descended from
a common ancestral source, ie I2->A and I2->B, it is necessary to
satisfy one of the following conditions:  Either *one*, to find a
'perfect' sequence of fully functional intermediate forms I1, I2, I3
leading unambiguously from one species to another, ie A->I1->I2->I3->B,
or I2->I1->A and I2->I3->B, or *two*, to reconstruct hypothetically
in great detail the exact sequence of events which led from A to B
or from a common ancestor to A and B, including thoroughly convincing
reconstructions of intermediate forms and a rigorous and detailed
explanation of how and why each stage in the transformation came about."

[Editor's note: Whoa! Man overboard! I don't think I could establish
my family tree with such criteria!  First, this isn't to say that
Denton is not correct in the strictest sense, but I wonder if anything
could be confirmed that required this level of confidence. Second, I
think that a lot of data has shown up in recent years which doesn't
quite, but almost, passes this test.]

Here's the pattern I think he expects to see for "proof": (where
":" is the sequence of today, and "-" the relative amount of
divergence)


       - group A:
 ------|      - group B:
       -------|                  - group C:
              -------------------|
                                 ----------- "X" organism of interest:
If this were true then you'd expect that the differences B->C and
B->X would be different with (B->X) greater than (B->C).

But in real life, there's no reason to assume this (regardless
of whether the molecular clock is "stable" or the hypothesis is even
right). Things might look like this:

       - group A------------:
 ------|      - group B------------:
       -------|                  - group C---------:
              -------------------|
                                 ----------- "X" organism of interest:
With the above, the B->X and B->C distances could be the same, or not.
However, the X->C and X->B distances are likely to quite diagnostic
(Caveat: Not always).

Another thing that's odd is that Denton accepts that speciation
can occur.  In fact, he cites results where researchers have looked
at the rearrangements of chromosomal segments between drosophila
species as good evidence for direct relationships between certain
species.  So he accepts "homology" at the level of gross chromosome
structure but not at the level of individual genes (I didn't see
whether he mentions anything about conserved intron/exon structures,
but it could be that I missed it on my first quick pass through the
book.  They aren't mentioned in the index, however).

The book was published in 1985.  The most recent reference in Chapter
12 is from 1980.  No reference in Chapter 7 "The Failure of Homology",
where Denton talks a bit about developmental biology, is dated
post-1977 (with the exception of an _Encyclopaedia Britannica
reference).  So maybe some "new" molecular data is available. ;^)

Regards,
Tim Ikeda
timi@mendel.berkeley.edu