Confessions of a Darwinist


To celebrate Darwin Day, The Virginia Quarterly Review has released Niles Eldredge’s essay “Confessions of a Darwinist”. It will be published as part of a special on Darwin, evolution, and ID in the spring issue.

Read “Confessions of a Darwinist” by Niles Eldredge.


Excellent article by Eldredge addressing so many issues. Thanks for the link. This ‘Darwin Day’ thing is surely ‘growing up’ to become a real contributor to evolutionary science.

What caught my eye was how so many sciences go through a similar cycle:

Evolution in those days was firmly in the hands of geneticists—who were at that very moment collectively like deer caught in the headlights of the onrushing revolution in molecular biology. DNA was threatening the comfortable world of population genetics—and there simply was little intellectual time or psychic energy for genetics-minded biologists to pay any attention to the results of a study on the evolution of a small cadre of long-dead and all-but-forgotten trilobites.

Simple equations seem to explain the world only to fall apart when more detail becomes available. Yet molecular biology also became a powerful test for evolutionary theory, and evolutionary theory survived.

In oceanography, during the early half of the 1900’s there was this idea that most of what explains oceans, circulation, exchange with atmosphere was well understood. Of course the crude temperature and salinity measurements smoothed the vertical profiles. When new instruments show an incredible variability, oceanography was ‘rejuvenated’. Yes, much complexity was added and science was ‘shocked’ but this complexity explained a lot of details… Much of what oceanography understood was the zero’th order variability or the large scale. Genetics, in this case population genetics is a “static overview” of evolution which led some to question how did evolution turn out to be so succesful (aka the representation problem)? The answers were to be found in the variability (evolvability)

When I use ‘static’ I mean that evolution used to look at how variation evolves but the “potential to vary” or variability is a far more interesting concept, and far harder to quantify and measure. See Complex adaptations and the evolution of evolvability by Wagner and Altenberg.


Darwinian evolution, characterized by heritable variation and selection, is not by itself sufficient to account for the capacity to vary, the specific type of variability present, nor for heritability of phenotypic fitness. Rigidity of genotype-phenotype mappings, as often used in evolutionary computation or population genetics, constrains the dynamics of evolution to a small space of possible biological or artificial systems. Open-ended evolution is not possible under such constraints. Yet evolution, by itself, cannot fully explain the advent of genetic systems, flexible genotype-phenotype mappings, and heritable fitness. This presents a challenge both to biologists seeking to understand the capacity of life to evolve and to computer scientists who seek to harness biological-like robustness and openness in the evolution of artificial systems. The sources of variability and its transmission between generations have been identified as key to biological evolvability. Properties such as the facilitation of extradimensional bypass and robustness to genetic variability (Conrad, 1990), heritability of fitness (Michod & Roze, 1999), modularity, as well as robustness to developmental variation (Kirschner & Gerhart, 1998) play important roles in evolvability.

To return to the orginal topic: science often faces ‘crisis’ where new data is waiting for science to catch up to with new hypotheses and predictions. Such moments is where science really shines and where I believe ID really falters. Because it is exactly when science lags the data that ID can intervene by pointing out how our lack of understanding is really evidence of ‘intelligent design’. One need but look at the Cambrian explosion, which by its age is hard to study. Compare how ID deals/dealt with this ‘crisis’ versus how science deals/dealt with it and compare the relative impact on our understanding generated by the two ‘approaches’.

It’s during these “threats” or ‘crisis’ that science shows what it’s worth. And in that aspect ID may serve, as Del Ratzsch points out, at least a function of ‘keeping science honest’. But that function bears little relevance to the larger claims of ID.

I have to admit, I never understood why there is ever said to be a conflict between different rates of evolution at different times, and the fundamental mechanisms of genetic variability and natural selection.

Most of the time, most populations live in niches for which they are extremely well-adapted. Genetic variation is likely to be selected against. We see an appearance of “stability”.

At times, new niches open up. The new niche can be a new environment, or perhaps even an initial change in an organism that fundamentally alters its ability to exploit the same environment. It would often be hard to distinguish between these two “forces”, which would tend to reinforce one another.

To give a simplified illustration, we might say that when vertebrates began to be able to colonize land, a period of rapid evolution might have been expected. The selection for the earliest and most minimal abilities to change environments might have been related to a change in the ocean environment, or it might have arisen randomly. The original advantages that such mutations conferred might have been related to factors in the ocean environment - and indeed probably were.

Once vertebrates were able to exploit a new environment, variability that might have been selected against in the old environment might now lead to fitness for a new niche. Hence, one might expect to see “more rapid” evolution.

I suppose this could be summarized by saying that the “rate of mutation” is moderately constant per base pair per DNA replication event, for chemical reasons, with the caveat that DNA repair mechanisms vary a lot. But the “rate of natural selection” may depend a great deal on whether a population occupies a tight niche in a crowded overall environment, or is colonizing a relatively new environment. In the latter case we might expect to see “more rapid evolution”, leading to an appearance of “saltation”.

On “Mutation rate constant over time”

This hypothesis is, from my reading about Homo Sapiens, not borne out by the evidence of surviving-until-today mutations. Indeed, many shallow time predictions via mtDNA analysis do not agree with the paleoarchaeological record.

As an example, Alan Templeton’s recent paper in Yearbook of Physical Anthropology simply refuses to state any dating more recent than 130,000 years ago. How wise of him!

I didn’t even know there was a strong feeling about constant rates of mutation in the biology community. As an outsider physics type, I knew about the many reversals in the Earth’s magnetic field and the concomitant variations in background radiation. Also, all systems that are maintained far from equilibrium are frequently subject to large fluctuations. I guess I just assumed that rates of evolution would not be constant and was surprised when I found out it was an issue.

Highly non-linear systems can be stable for long periods of time, but if perturbed, can snap into another stable configuration until it is perturbed again. I would think that mutation rates themselves would be a kind of semi-stable system, otherwise organisms could not develop a coded heritage to pass on. But under stress, organisms would have to find a new local stable point. If there is enough spread in the system characteristics, some of that spread would overlap the new niches opened up by perturbations in the environment.

This is just an abstract way a physicist would think about this. Hope it doesn’t sound too stupid to a biologist.


I’m certainly not a biologist either, but I recall reading that Niles Eldredge was one of those that noticed the sort of meta-stability you mention. The observation was that a species (viewed as a breeding population) in fact does NOT follow the moving target of a gradually changing environmental niche. Over the course of millions of years, that environment can change quite drastically for geological reasons, changing in elevation, temperature, aridity, insolation, etc.

And the observation is, the species alive at the start of this period were either *already* capable of living with the changes, or they went extinct. They didn’t track any changes. Instead, new (and better-adapted) species branched off on a more or less continuous basis. This is perhaps a subtle distinction: as a new niche opens up, a species does not adapt to fit; instead a new species branches off and rapidly evolves to fit – at which point THAT species “locks in”.

This is Eldredge’s model, which (at least in the past) the “gradual trackers” derided as “evolution by jerks”.

I may have been unclear when I said “the mutation rate may remain relatively constant” (or whatever specific words I said).

DNA replication is a biochemical process, catalyzed by enzymes. The mutation rate can be impacted by the variation of background influences like ionizing radiation, by temperature (especially in non-homeotherms), and the net mutation rate can be impacted by the presence of DNA repair mechanisms, which many if not most modern cellular organisms possess. Humans with DNA repair defects have a much higher rate of certain cancers, presumably due to lack of repair of somatic mutations, and skin cancers predominate, suggesting that solar radiation is a big factor. Undoubtedly, many factors can influence the mutation rate.

Having said all that, DNA replication can more or less only take place in the narrow range of conditions that are conducive to life as we know it. So there are sharp limits on mutation rates, at least if we exclude “mutations” that occur as part of the overall destruction of an organism by something like burning or freezing, and include only those that result in at least a transiently viable post-mutation cell or virus.

What I am really trying to say is that even if the mutation rate is relatively constant over time, we would still expect that the “rate of evolution” at least as perceived by human observers of morphology, would vary over time. When a new environment opens, selective pressure would reduce. As environments become more crowded, and every niche becomes occupied by hyper-adapted species, an appearance of relative “stasis”, at least with respect to human observation of morphology, would result.

The question of how to measure the “rate of evolution” is actually a profound one. My point is merely that, to me, the idea that it should always subjectively appear gradual and constant does not derive from the fact that genetic diversity, sometimes acted on by natural selection, is the mechanism.

Thanks Harold and Flint. What you said makes sense.

I didn’t know how large a range mutation rates could cover, but I couldn’t imagine a physical system in which they were actually constant. And, if they were absolutely constant, I would think many organisms wouldn’t be able to adapt quickly enough to relatively rapid changes in the environment (obviously, those that don’t become extinct). So the organisms that survive drastic changes in the environment would more likely be those that had more variable mutation rates. The long periods of stasis would just mean that the organism was stable enough that it didn’t need to capitalize on this variability. In fact, I would suspect that changes in temperature, background radiation, etc. could affect environment and mutation rates at the same time.

In another abstract sense, a small perturbation in the mutation rate can produce larger (and perhaps more timely) changes in observable morphology if the selection pressures are great enough. I suppose that is because many DNA mutations (the ones that don’t kill) affect multiple characteristics in an organism, and give the organism additional adaptability to find a niche if one opens up. So perturbations in the mutation rate are amplified in the organism, just like many non-linear systems.

Pardon me if this is a bit naive. Just trying to wrap a physicist’s mind around some evolutionary concepts. I appreciate any insights you can share. Thanks.

By the way, I find these discussions much more enjoyable than the flame wars in some of the other threads.

Yeah, this place needs a system where there’s some positive drag coefficient on flames. I suggest a /. style system. Here’s what I said on After the Bar Closes about it:

What I like about the /. system, I think would work here if it could be implemented–people who have been around long enough to be trustworthy are given small powers of moderation. They can demerit comments for being offensive or garbage or whatever. Individual viewers of the site can select what level of comments to see. If you particularly want to see the worst comments, you can, but everybody else wouldn’t have to. At /., they have so many comments that their system has 7 tiers, but here we’d only need two: regular and garbage.

Flint. Just to clarify what I said. I understand that the “parent” species doesn’t necessarily have to change, but can disperse “daughter” species that “lock in” as you said. I was thinking that a constant rate of mutation seemed “unphysical” for a biological system, and was trying to understand how a variable rate might be an advantage. I think Harold clarified the point of variability for me.

Mike Elzings -

I believe there is some evidence that species in some rather variable environments can actually be selected for higher mutation rates than we generally see elsewhere. I guess the pressure in those environments is literally that if your many, many kids are more variable, you have a statistically better chance that some of those kids will survive to have kids. than if all of the kids are a lot like you. Whether this leads to relatively rapid branching off of new species, I have no idea. (And I said “relatively”).

Having said that, I now realize that instead of saying that the mutation rate is constant, I should have said that the range of mutation rates in organisms that survive and reproduce is constrained.

Not only is it constrained by the fact that mutation only takes place in the narrow range of temperatures, chemical environments, and pressures that even the most uniquely adapted life can withstand, it is also constrained because there has to be some “minimum threshold” of fidelity to the original template to allow viable offspring to be produced at all.

And as I said above, to no significant disagreement, and yet I repeat, that doesn’t mean that what we humans perceive as the “rate of evolution” need be equally constrained over time and place.

Thanks Harold. Well said. I find evolution extremely fascinating. It’s like having complex, non-linear systems within complex, non-linear systems, within … etc. Systems like this are actually a stable way to dissipate gradients in the flow of matter and energy within certain ranges of gradient. Of course, if gradients are too high, the systems get destroyed, but then other patterns develop that handle the steeper gradients more effectively. If the gradients are too small relative to the internal energies within a system, nothing interesting develops. Life seems to exist on a knife edge, and this is what makes it so interesting. I know how to grow snowflakes on a computer, but this is orders of magnitude more computationally intensive.

le kie beis speca me lienga . … . hlp?

This is cool, you have to try it. I guessed 53566, and this game guessed it! See it here -

About this Entry

This page contains a single entry by Reed A. Cartwright published on February 12, 2006 3:22 PM.

Those pathetic pesky details again was the previous entry in this blog.

Evolution Sunday/Darwin Day is the next entry in this blog.

Find recent content on the main index or look in the archives to find all content.



Author Archives

Powered by Movable Type 4.361

Site Meter