Extinction, evolution, etc.

From: David C Campbell <amblema@bama.ua.edu>
Date: Mon Aug 01 2005 - 18:11:29 EDT

Trying to heed the posting limit, this addresses several issues from
various more or less related threads:

Cambrian explosion: The chapter by Keith Miller and me in his
Perspectives on an Evolving Creation also addresses the Cambrian
explosion. Ironically, the ID version of the Cambrian explosion relies
heavily on older work by Gould that overestimated the variety and
abrupt appearance of the Cambrian forms, “proving” that it was random
and not following divine direction. Both the ID interpretation and the
“random” interpretation constitute reading one’s views into the
evidence, and they also do not accurately reflect the current knowledge
of the latest Precambrian and the Cambrian. The latest Precambrian has
clear evidence of some of the most primitive phyla (sponges,
cnidarians) and primitive forms akin to other phyla, but generally
having a rather primitive aspect. The early Cambrian includes many
transitions between higher categories (phyla, classes, etc.). Some
forms remain problematic, but many have fit into groups with further
        Although there are aspects of the Cambrian radiation not yet fully
understood, nothing about it is in conflict with an evolutionary model.
 It does contain clear contradiction of claims that phyla, classes,
orders, families, genera, or species cannot evolve from other forms.

>People have tried to appeal to the Christian tradition for this
conclusion but it's a questionable procedure. Precisely the same logic
leads to the YEC dogma "No death before the fall"<

The difficulty in part lies with the difference between compatibility
with Christian tradition (or better, with Scripture, as some traditions
are ill-founded) and necessity. No extinction and no death before the
fall are compatible with Christianity (in ignorance of the physical
evidence) as far as I know; the problems are their incompatibility with
the physical evidence and the unbiblical tactics used to support them.

On the specific issue of extinction, my impression is that it was in
part viewed as an implication that God had made a mistake in creating
something that did not last, akin to the idea that orbits ought to be
circular as the most perfect shape. However, this overlooks the
possibility of creation of a dynamic system, with kinds of organisms
changing along with their environment. Modern anthropogenic
extinctions also reflect the fact that God allows humans to misuse
creation and suffer the consequences.

Jedidiah Morse’s late 1700’s book on American geography, a forthright
and encyclopedic primer for newly graduated seminarians on what to
expect as they take up pastorates in the wilds of the new U.S.A., notes
the possibility that the mastodon is still living in more remote
regions, but treats it solely as an aspect of natural history (might be
dangerous) rather than a theological problem of possible extinction.
That suggests that extinction was not much of an issue by then.

Defining evolution:
> And most people (teachers, students, general public) think that claims
for "E as fact" or for "no controversy about E" means all of it, 1234.
Usually there are no 1-2-3-4 distinctions (or they are very
inadequate), there is just E.<

The problem arises in both directions, as adamant antievolutionists
object to citing 1, 2, 3, etc. as evolution because of the slippery
slope argument. (evolution did not happen so we can’t say that some
evolution happened)

>"God wouldn't create a Linnean pattern" (characterization of an
argument for evolution).

This falsely assumes “either God or evolution”. Since Linnaeus assumed
that the pattern was created, it seems unlikely that many people
believed such an argument. However, the Linnaean hierarcy in fact
turns out to match very well with the expected results of evolution.
On the contrary, it is not clear why separate creation should produce
such a pattern. Separate creation could produce such a pattern, or any
other pattern, but why it would produce one in line with the
expectations of evolution is not evident to me.

>There are *hundreds* of UCEs -- sequences with 100% identity between
the human, mouse, and rat. Even if these sequences had known function
their sequence identity would be very strange. You'd have to hunt and
hunt to find *known* functional sequences with 100% identity between
these species. This finding of 100% identity in these hundreds of
sequences means that their putative function would have to be
absolutely contingent on *every* residue's identity. And on top of
this, there is no known function, making this finding even stranger. <
The comparison is complicated by the fact that different parts of
“functional sequences” are more or less important. The most common
category of known functional sequences are genes, and genes typically
include more and less critical portions.
If the conserved sequences are important for survival (perhaps in
reproduction and development-how long were the knockouts traced? How
many of the sequences were knocked out? How many in any one mouse?),
then there is a clear evolutionary reason for their lack of variation.
Some of them also could be closely associated with features important
for survival and therefore are shielded from change even though they
themselves are not essential. On the other hand, if they are without
any function, why should a designer stick them into a bunch of things?
It’s not obvious evolutionarily nor IDily why they should exist in the
latter case. However, it’s worth remembering that rodents and primates
are rather closely related. Close genetic similarity between humans
and mice is less surprising than equal human-armadillo similarity would
be, for example. Another possibility is that they have important
functions but that alternative systems exist. This possibility is a
problem for irreducible complexity arguments.
Another possibility is that some of the sequences are some sort of
genetic parasite that are able to maintain themselves.
How long are the identical sequences? Identical short sequences are
much less surprising than identical long ones.
>This is a direct falsification of common descent which predicts that
functionally unconstrained sequences are not conserved. <
No. Common descent predicts sequence similarity. Common descent does
not say anything about the conservation of functionally unconstrained
sequences. Models of molecular evolution predict that functionally
unconstrained sequences should not be conserved. This prediction is
frequently observed to be correct, e.g. spacer regions, pseudogenes.

>Phylogenetic incongruencies are common. A systematic study was done on
the genomes of 5 different light-harvesting bacteria showed dramatic
inconsistencies. Every conceivable phylogeny found support, and no one
phylogeny was significantly preferred (Science, 298:1616).
Mitochondrial sequences are at odds with nuclear sequences. In one
study they clustered frogs and chickens with fish. I gave you the
citations. Alan Feduccia wrote that the growing gap between molecular
analyses and the fossil record "is astounding" (TREE, 18:172).<

Bacteria readily take up DNA, and those major groups for which
distinctive fossil evidence is known (often in the form of distinctive
chemicals rather than recognizable microfossils) typically extend far
back in geologic time. Even among eukaryotes, genetic exchange between
physically associated organisms is occasionally reported. Thus, there
are both means and opportunity for DNA exchange that can throw off
evolutionary analyses. For some bacteria, the DNA is reportedly nearly
randomized relative to other bacteria, though I do not know the extent
to which our sampling of bacterial genomes provides an adequate
measure. Sequenced genomes emphasize (a) species of medical
importance, (b) things that are happy in petri dishes on agar, (c)
random environmental samples. Increasing numbers of C should give a
better idea of the overall diversity of bacteria and the variation
among groups.

>Doolittle is saying that the evolutionary tree model isn't working. <

For relationships among major groups of bacteria, there are some
problems. The evolutionary tree model works very well in many cases.

>About retro viruses, one widespread germline provirus is missing in
chimps and gorillas. Since it is widespread, it must have been in a
distant common ancestor according to common descent. But in that case
the chimps and gorillas should have inherited the provirus. But they
didn't, so CD needs some clever story about an unlikely event that we
can never verify or falsify. <

You don’t have everything that was in the genome of either of your
parents. As the provirus is not important for the survival of
primates, there’s nothing far fetched about its being lost. It is
perfectly true that there is no way to directly test whether it existed
in the ancestor of chimps and gorillas. However, the overwhelming
morphological (including paleontological) and molecular similarities
among great apes and humans fits well with an evolutionary explanation.
 In light of that, the absence of something from a couple of
representatives that is otherwise present across the board strongly
suggests that it was lost.

Again, it’s not very evident why a designer would separately stick a
provirus into a bunch of genomes while skipping chimps and gorillas. A
convincing argument for ID needs to demonstrate having a better
argument, not merely that evolution seems to have some trouble at a
particular point.

>The finding of nonhomologous development means that similar structures,
in similar species, come from different genes or embryonic development
pathways. How could this be on CD? <

Nonhomologous development does occur, but many cases of similar
structures in similar species coming from the same genes and
developmental pathways are known. The use of different genes and
pathways can readily occur through evolutionary change. Your argument
about conserved regions claimed that things have to change
evolutionarily; this claims that things can’t change evolutionarily.

>Then there are complexities, such as protein synthesis which uses the
DNA code. Protein synthesis in the cell is a phenomenal process
involving hundreds of macromolecules performing feats of copying and
translating, proofreading, checking, etc. And we have no idea how it
could have evolved.<

Various ideas do exist. There are hints from the genes and cell
function. Examples include the similarities between various tRNAs that
point to origin from a gene duplication event. That duplication must
have taken place in an organism with a simpler protein-building
machinery with fewer amino acids. Again, the role of RNA in the cell
and its ability to function as an enzyme suggests the possibility of an
organism relying mainly on RNA rather than proteins. Increased
proofreading, checking, etc. are useful but not necessary and can
readily be added on over time.

>For this, and myriad other complexities, evolutionists can only provide
broad speculation that has no basis in reality. This is not like
complaining about the mechanic who has not fixed your car yet because
he is busy. These structures are very well understood and evolutionists
have spent countless hours trying to figure out how they could have
evolved. Highly trained scientists with powerful computers cannot
figure out how a dumb process could have done it.<

As noted above, there are bases in reality, though I’m not sure what
sort of evidence you want. Very little evidence about details of
biochemical pathways has any chance of getting into the fossil record,
so models of how they may have evolved must be based on examination of
existing organisms (for similarities, patterns, possible relict
features, etc.), evidence on the environmental conditions of the time
(which, though probably not exactly like the Miller-Urey setup,
probably were more favorable to the formation of complex organic
compounds than some have claimed), and modeling of possible
evolutionary pathways.

We still have a lot to learn about the basic structures. Remember that
DNA sequencing is less than 30 years old and the development of
computer technology to handle much of this is also at most a few
decades old. Critical advances are much more recent. Only recently
has there been much success in predicting protein structure from
sequence, for example. Furthermore, evolution is not a big research
priority nor funding priority. Money closely follows the prospect of
immediate commercial usefulness. Despite this, there are many ideas
about the evolution of biochemical systems and various research
programs going on. The mechanic is still figuring out what the parts
of a car are and what they do. The fact that he has not yet inferred
full details of the Detroit assembly line (especially if the assembly
line has evolved into something dispersed across several countries by
the time the mechanic tries to figure it out) does not disprove the
existence of GM.

My own research focuses on the evolution of the Mollusca. There are
many transitional forms, many good series in the fossil record, and
good agreement between morphological and molecular analyses. I do
occasionally run across things that don’t match up. Some of them
reflect misinterpretations or mistakes. Some of them reflect real
problems that have not been sorted out yet. Some discrepancies reflect
previous models based on very little evidence. However, the overall
match with evolutionary expectations is excellent.

>You say that abrupt appearance of fossil species are not evidence
against evolution<

Fossil preservation depends on many factors. The key issue for
evolution is the existence of the organism to get fossilized. However,
several more factors affect its chances of getting fossilized,
including the durability of the organism, the general suitability of
the environment, and any unusual conditions. If it is fossilized, the
beds must not be eroded away, metamorphosed, or otherwise destroyed.
Finally, the fossil must be found and studied. In light of the
patchiness of the fossil record, many things should be expected to have
a sudden appearance. If the change happens in spurts rather than
gradually and continually (i.e., a more punctuated pattern), as is the
case for some organisms at some times, then the chances of catching the
change occurring are even lower. (Note that this is not something made
up due to lack of evidence, but rather is based on specific examples in
which little or no change was observed over a long period of time,
followed by observed examples of rapid change, and then another stable
interval.) Another problem is an artifact of classification.
Assigning Linnaean names, while very useful in providing points of
reference, emphasizes the recognition of differences and drawing lines.
 Things that don’t fit so well are often neglected, especially when the
main function of naming is to help recognize distinct forms. This is
the case for biostratigraphic work, which, as the main commercial
application of paleontological taxonomy, has been a major driving and
funding source for naming fossils. Also, separation of workers easily
produces apparently abrupt changes when going from one person’s work to

Thus, many examples of abrupt appearance or apparently (based on
published data) abrupt appearance in the fossil record will be expected
under an evolutionary model. However, these are not expected to be
randomly distributed, and many examples of good evolutionary precursors
are expected. This matches well with the observed patterns. Things
with good fossil records that are well-studied generally have many good
series of fossils and transitional forms. Things with poor fossil
records or ones that have received little study of evolutionary
patterns have fewer. At higher taxonomic levels, the chance of
sampling is higher, and the pattern is correspondingly stronger. The
chances of determining the ancestral species to a randomly chosen
species may not be very high. However, the chances of having a pretty
good idea of its antecedents at higher taxonomic levels is good. It’s
like tracing your genealogy. Some cultures have better records than
others. It’s much easier to find someone who is probably a close
relative than having absolute certainty that you have located your
great great great great great great great great granfather.

For example, the early Cambrian does have the first appearance of many
forms, some with no close antecedents known from the Precambrian.
However, we do see a lengthy fossil record of bacteria, and a somewhat
less lengthy but still long record of eukaryotic protists. Sponges and
cnidarians are among the most primitive animals and are the first
distinctive animals known in the Precambrian. The Precambrian and
early Cambrian forms are generally primitive in comparison with later
forms. Radiation of various lower taxonomic levels can be traced over
time. E.g., I could not confidently identify particular Cretaceous
fossils as being the ancestors of eastern North American amblemine
freshwater mussels. However, I can identify many Cretaceous fossils as
being freshwater mussels of some sort. Going back to the Triassic,
there is good evidence of the derivation of freshwater mussels from
trigonoid bivalves, a relationship also supported by all molecular
analyses that have included both groups. Going back through the
Paleozoic, the trigonoids can be traced through successively more
primitive forms to lyrodesmatid bivales in the Ordovician. Details of
the relationships of various Ordovician bivalves are not clear (and
barely studied), but lyrodesmatids fall into a general group, supported
by molecular and morphological data. In turn, this group can be linked
with the group of oysters, scallops, etc. on morphological and
molecular grounds, though a specific fossil common ancestor is not
known. The few known Cambrian bivalves are not obvious as direct
ancestors of these groups, but they are relatively generic in many
ways. There are good transitional intermediates between a more
cap-shaped shell and the bivalve form. Kimberella, in the Precambrian,
has mollusk-like features. Mollusks and various other phyla such as
annelids share some morphological features (especially in early
development) and are molecularly close together. In turn, these
resemble other bilaterian animals in molecular and morphological

>you say that phylogenetic incongruities are due to incomplete data<

Some reflect incomplete data. Some reflect inappropriate data. E.g.,
long branch attraction in which essentially random similarities are
mistaken for phylogenetic information. Some reflect misinterpretation
of the data. E.g., molecular clock assumptions that are unjustified by
the evidence; assumption of homology in the case of convergence or
paralogy. Some reflect genuinely incongruous patterns. E.g.,
horizontal gene transfer, differential survival of ancestral
polymorphisms, hybridization.

>It has no problem with early mammal fossil showing up all over the
As they occur in the Triassic, when Pangea was still attached,
migration is quite plausible.

Dr. David Campbell
425 Scientific Collections
University of Alabama, Box 870345
Tuscaloosa AL 35487
"James gave the huffle of a snail in
danger But no one heard him at all" A.
A. Milne
Received on Mon Aug 1 18:12:52 2005

This archive was generated by hypermail 2.1.8 : Mon Aug 01 2005 - 18:12:53 EDT