Re: EvolutionVsDevelopment

Glenn Morton (
Fri, 24 Jan 1997 23:12:38

Bob DeHaan wrote:
>I find it difficult to respond to your questions and comments piece-meal. As
>I indicated to you in a private post, I will prepare a summary of my position
>on phylo-development and send it out for comment from all. Meanwhile let me
>say a few things about some of your comments and questions.
>I tried to identify your position with regard to some clearly defined
>meanings of the word "evolution." I suggested that "change through time", the
>most inclusive definition of evolution, fairly describes your position. You
> "I would object here. My post was specifically aimed at showing that there
>is NO limit to change via mutation. As such, I did not feel the need to
>raise the natural selection issue in that context."
> I am puzzled. I don't know what you are objecting to. Isn't change
>through time what you are describing for yourself.

No. I was objecting to the characterization of my view of evolution as mere
change through time. Natural selection plays a BIG, BIG role in what changes
are made through time. I do believe in descent with modification as I know
that there is a genetic connection between species and I believe that there is
a genetic connection between genera, families, etc. So I am a raving
evolutionist who happens to believe that the Bible can also be historically

> Second, I have never
>heard the claim made that there is no limit to via mutation, except from
>people like Dawkins and Dennett. Your model may show it where I can't find
>it; where is the empirical evidence to support your statement?

If I put a sequence of letters in a computer, a fast computer, and allow the
locations to be mutated. And if I only allow 4 letters in the game, A,T,C,G,
then any sequence I start with can be mutated to any sequence I want. That is
a mathematical fact! Do this on a computer! If mutation is really random the
computer can mimick it.

"attgccagaatgaacctttga" can be mutated to "tgcgatagatcat..." or anything else
by merely waiting long enough. There is no limit. Lets do it simply. Use the

"ATTG" randomly alter it within a few thousand tries you will have exhausted
every single option. There are only 256 different possibilities. Where is the
limitation that won't allow me to mutate this sequence to ACCG? DNA is merely
a sequence.

Some sequences do not allow a living being to survive. These "bad" sequences
are removed by natural selection. But as long as there is a path through the
sequence space from one sequence to another in which all sequences allow a
living being to exist and reproduce, then mutation can move from one species
to another.

> There are
>boundaries and limits to the amount to change that can be wrought in the
>biological world. The fact that body plans, on the whole have not changed,
>or at least one phylum has not changed into another one since the Cambrian,
>would suggest that mutational change is not limitless.

Considering that a new phylum was found two years ago, would suggest that this
may not be the case. (see Funch and Kristensen,"Cycliophora is a new phylum
with affinities to Entoprocta and Ectoprocta," Nature, Dec. 14, 1995, p. 711.)
It may be difficult to change phylum but it may not be impossible. This
phyla left no fossil record but lives today. One can make a reasonable
argument that it is a new development.

>Further, evolutionary theory calls for a selecting agent. Your limitless
>muational change does not have one. Do you define evolution exlusively in
>terms of mutations sans selection?

No. I didn't say that all mutations were able to give rise to a living being.
I said that if there was a pathway then mutation could take it there. Let me
simplify sequence space because it is obvious that you are not familiar with
information theory. Assume that all life was based on a 2 unit nucleotide.
All genomes were that short. There are 16 possible life forms. In the chart
below all sequences with an X do not produce life and the O do. A being born
with a mutation giving a genome of AA, AG,TC,TG,CT,CC,CG are born dead. If I
create a living being with a genome of AC, I can mutate it successively to
AT,TT,TA,CA,GA,GT,GC,GG. I can successively produce these genomes by random
mutation. What is the limit to this type of change?


In real life DNA you have billions of dimensions and the mathematics are quite
beyond our ability to understand that kind of mathematical space. For the
human genome, every point in such a space can be approached from 3.5 billion
DIFFERENT directions. If there is a path into that point, it can be found by
random mutation, given enough time.
>I dwelt on the transition from aquatic to terrestrial animals and asked the
>question about the fate of the *first* mutation that would initiate the
>change fromm a fin to a leg. I questioned how such a mutation or series of
>mutations would be adaptive since it would interfere with an already
>well-adapted, well-functioning organ, the fin. You did not respond to my
>question. Instead you went on to discuss front paws. What I want to know is
>how you get to front paws in the first place when the very first mutation in
>that direction is presumably inadpative?

I beg to differ that I did answer your question. I did not ignore it. Part
of our problem is that our terminology is quite different. The nature of a lot
of mutation is that it changes a whole lot of things all at once. To assume
that 1 mutation produces 1 small change and another mutation produces 1
another small change is an outdated view of how mutation and development work.
1 mutation can turn off a whole lot of systems. Gould says,

"One promising solution to this dilemma recognizes that certain kinds of small
genetic changes may have major discontinuous effects upon morphology. We can
make no one-to-one translation between extent of genetic change and degree of
alteraion in external form. Genes are not attached to independent bits of the
body, each responsible for building one little item. Genetic systems are
arranged hierarchically; controllers and master switches often activate large
blocks of genes. Small changes in the timing of action for these controllers
often translates into major and discontinuous alterations in external
form."~Stephen J. Gould, Hen's Teeth and Horses Toes, (New York: W. W. Norton
& Co., 1983), p. 181

As I noted, 3 mutations perform most of the morphological change between two
species of monkeyflowers. (H.D. Bradshaw Jr., S. M. Wilbert, K. G. Otto and D.
W.Schemske, "Genetic mapping of Floral Traits Associated with Reproductive
isolation in monkeyflowers (Mimulus)," Nature, 376 Aug. 31, 1995, p. 765)

>On the other hand, if one assumes that there was a developmental program for
>the transition from fins to paws, then the first step does not depend on

I would prefer not to assume until there is some evidence to make such an
assumption necessary.

> The program simply drives the animal through the incipient
>stages, come hell or high water, until the final stage of paws, or claws or
>hooves has been attained. That doesn't mean that there wasn't any shaping of
>the incipient stages along the way by natural selection. Developmment theory
>holds that the main line of change was developmentally driven, with side
>lines and adpative embellishments added by natural selection.

If as you indicated in your previous post that the entire paw development
program was incorporated in the genome of the panderichthys, then what role is
there for natural selection to perform? The program will produce paws
regardless of what the environment does.
>Your next question is the tough one. Did the germ line of the original
>vertebrates contain all the genomic developmental programs for all the
>subsequent fish, amphibians, reptiles, mammals, humans? Logically my
>position calls for it. That's not to say that I'm comfortable with this
>position. At this point I am assuming that the germ line genome of Pikaia,
>one of the presumed first vertebrates, was rich in genomic developmental
>programs even though its individual somatic genome was quite simple. A
>small, simple body carrying a germ line that was loaded with developmental
>In defense of this position, I refer to a hint made by Vefmeij in his
>commentary on Wray, et. al.'s, article "Molecular Evidence for Deep
>Precambrian Divergences Among Metazoan Phyla" in Science of 25 Oct. 1996.
> Vermeij said, "Contrary to some perceptions of the Early Cambrian,*a large
>array of developmental patterns* was already available when the momentous
>spurt of the latest Neoproterozoic and earliest Paleozoic eras ensued"
>[referring to the Cambrain explosion] (p. 526). It was developmental
>programs that were synthesized in the Precambrian according to Vermeij's
>comment which were then organized and ordered into the array of Cambrian

Maybe. But that does not mean that the entire developmental information for a
human was available to Picaia. Unless you can answer the question of why the
human developmental plan was immune to mutation over the past 560 million
years, I can't see how your view can be viable.
>My second comment is that the structural developmental programs may not have
>required as much space in the germ line as might be supposed since the genome
>of mammals is made up of a large amount of silent genetic material. I
>hypothesize that the developmental patterns suggested by Vermeij may not have
>been physically very large or taken up large amounts of genomic space. I do
>not know of experiemental evidence for developmental regions.
>Another difficult problem I encounter is how to account for the sequence of
>expression of the developmental program throughout millions of years. Why
>did fish come before amphibians, which came before reptiles, etc? I have no
>answer for this question.

Unfortunately, a good theory needs to be able to answer questions like this.
If you are wanting to replace evolution, you also need to be able to have a
theory which makes predictions which evolution does not make and which then
turn out to be true. I would encourage you to develop solutions.
>Were these phyletic developmental programs immune from mutations, you ask? I
>would say yes, just as the developmental programs for the body plan
>apparently are. They have lasted with little or no change for hundreds of
>millions of years. Or they may have unusually effective repair mechanisms.
That may not be true. Developmental mutations occur all the time in living
systems. How do you handle polyploidy? For those who might not know
polyploidy is the process whereby a plant doubles its chromosome number and
becomes a new species overnight. Wheat and corn are polyploids. How do you
fit that into a developmental program? Does the program know when to make a
polyploid? How do you account for the change in chromosome type and number in
a species of rat which apparently evolved on Mauritius, over the past 400
years? (T.H. Yosida, et al, "Mauritius Type Black Rats with Peculiar
Karyotypes Derivedfrom Robertsonian Fission of Small Metacentrics,"
Chromosoma, 75: 51-62(1979).)

>But the position that a series of developmental programs drove the main
>changes in various phylas suggests that the relatively sudden breaks between
>the various taxonomic levels in a given phyla was produced by discreet
>programs, acting in a more or less holistic manner, not by gradualistic
>changes (or even punctuations) that is the hall mark of natural selection.

Natural selection works on changes that have already occurred. Natural
selection is not the cause of the changes but the arbitor of the value of the
previously made changes. The changes do not have to be gradual. Polyploidy
isn't gradual. That toad with his eyes on the roof of his mouth that I
mentioned in the last post did not get there gradually. His parents did not
have eyes on their noses.
>Still, many mysteries remain.
>You asked, "Are you denying that natural selection operates? Is not the
>development of immunity by disease causing bacteria a result of natural
>selection?" I do not deny the operation of natural selection. I only say
>that it works on a sliding scale. Early in the Phanerozoic, developmental
>programs were in the driver's seat, and natural selection came along for the
>ride, pitching in wherever it could. Starting in the Cenozoic, our current
>era, natural selection gradually took over the reins of biologic change and
>became the dominant force in the organic world because developmental programs
>were all already in place and there were no more left to be expressed.

So where does that new phylum of cycliophora come from?

> Development also works on a sliding scale. It starts out high, and ends
>low. NS starts low and ends high.

When you get right down to it, we are no more complex biologically from many
invertebrates. Our cells perform the same function with similar structures,
they have eyes, mouths, digestive systems, blood, etc. There is none of theis
high and low business. None of this higher and lower animals as far as I can

>Has the unfolding ceased, you asked. From my perspective, yes. There are a
>finite number of developmental programs in a phyletic germ line, after which
>no further developmental unfolding can take place. In fact, the germ line of
>phyletic lineages begins to enter into a time of increasing decline and
>disorder, after the developmental programs are played out. Note, for
>instance, the large number of genetic diseases ascribed to the human germ
>line, which does not suggest a lineal germline full of potential for the
>future, but rather one that is gradually losing its steam and decling.
>Did the immune system develop or evolve? The same developmental principles
>apply. The early structure of the immune system arose exclusively through
>developmental programs. (Dare I mention that the functioning of the immune
>system is one of Behe's irreducibly complex biological systems? I have
>argued with Behe that he should claim development as the biological process
>underlying his irreducibly complex biological systems, rather than leaping
>immediately to an Intelligent Designer.) The fine tuning of the immune
>system was done increasingly by natural selection after the basic structure
>and function was laid down by development.
>I'll have to get a copy of Gilbert's Developmental Biology in order to
>address vertebrate embryology. Can you give me the publisher.

Sinauer. He may have a subsequent edition out I have the third.


Foundation,Fall and Flood