More Recent Comments

Tuesday, May 04, 2010

The Mutationism Myth, II. Revolution

This is the fourth in a series of postings by guest blogger, Arlin Stoltzfus. You can read the introduction to the series at: Introduction to "The Curious Disconnect". The first part is at: The "Mutationism" Myth I. The Monk's Lost Code and the Great Confusion. The second installment is: Theory vs Theory.

Arlin is going to explain to you why everything you thought you knew about mutationism is wrong. You may even be a supporter of mutationism without even being aware of it!

The Curious Disconnect

Our journey to map out the Curious Disconnect— the gap between how we think about evolution and how we might think if we were freed from historical baggage— began with The Mutationism Myth, part 1. Then, in Theory vs Theory, we took a brief detour to distinguish theory1 (grand conjecture) from theory2 (body of abstract principles). Today we are back to the Mutationism Myth and our goal is to probe its claim that the scientific community rejected Darwin's ideas on erroneous grounds.1

The Mutationism Myth, II. Revolution

The Mutationism Myth is a story told in the literature of neo-Darwinism, regarding the impact of the (re)discovery of Mendelian genetics a century ago. In this story, the discoverers of genetics (characterized as laboratory-bound geeks) misinterpret their discovery, thinking it incompatible with natural selection; the false gospel of these "mutationists" brings on a dark period that lasts until the 1930s, when theoretical population geneticists prove that genetics is the missing key to Darwinism; Darwinism is restored, and there is peace and unity in the land.

In typical versions of the mutationism story that we reviewed in part 1, the Mendelians cast a spell on the scientific community, convincing it of a false belief that either

  • Mendelian genetics is inconsistent with the concept of natural selection or
  • selection is irrelevant because mutational jumps alone explain evolution

For instance, Eldredge (2001) writes:

Many early geneticists at the dawn of the 20th century, thought their discoveries of the fundamental principles of genetics somehow cast doubt [on], or rendered obsolete, the concept of natural selection

As noted earlier, a myth is not necessarily false. Some parts of the Mutationism Myth reflect history accurately, and others do not. An underlying truth in the Mutationism Myth is that, as a direct result of the re-discovery of Mendelian genetics, leading geneticists— Bateson, Johannsen, de Vries, Morgan, Punnett, and others— rejected Darwin's theory for how evolution works.

Our goal is to understand why. We must begin with heredity, the heart of the issue.

Re-discovering a lost theory

The re-discovery of Mendel's principles of heredity was nothing short of a revolution, and if you were trained in 19th-century views of heredity, this would be obvious, and there would be no need for me to explain it.

Unfortunately, the chances are good that you, dear reader, have been trained in the principles of Mendelism, and that puts us at a disadvantage. Once we can imagine the purity of hereditary factors, and we learn Johannsen's genotype-phenotype distinction, these principles seem to change our view of the world irreversibly, and its hard to understand what came before. Johannsen's quantitative-genetics experiments on seed weights of the Princess bean, conducted in the first decade of the 20th century, appear to have had more impact on evolutionary thinking than any single study conducted before or since. In the figure below, Johanssen (1903) shows the distribution of weights of beans from a plot planted with a mixture of seeds from pure self-fertilizing lines (the legend says "The variation of the weight of 5494 beans from the 1902 harvest, descendants of all weight classes in 1901") (online source):

The beans from the mixed plot show a nice bell-shaped distribution (figure). Similarly, the beans harvested from pure lines grown in separate garden plots also show nice bell-shaped distributions, though the means differ for each pure line. The key difference is in the results of selective breeding for heavier (or lighter) beans, i.e., planting a new crop using only the heaviest (or lightest) beans: selection shifts the distribution of seed weights in the mixed plot, but has no significant effect on the distribution of seed weights produced by a pure line.

Within just a few decades, neo-Darwinians such as Ford (1938) dismissed Johannsen's results as a logical necessity, as though the experiments proved nothing. Johannsen's studies had changed our understanding so profoundly that Ford was unable to imagine how scientists (mis)understood the world before.

I won't ask you to do what Ford could not, which is to forget genetics.

Instead, I would like to ask you to join me in imagining a different world— one in which particulate inheritance of pure hereditary factors does not apply.

We have been sent to this alien world as evolutionary experts, to consult with its scientists about how evolution might work on their planet. The alien scientists explain that, in their world, the bodies of organisms have differentiated organs composed of diverse cell-like units (CLUs), which swell, fuse, split and exchange material. The CLUs don't seem to have nuclei or central control centers. Instead, they are composed of substances that interact productively and grow, crystal-like (possibly some kind of prion-like protein, we think to ourselves). Different CLUs have different compositions, and thus have different developmental tendencies, e.g., some CLUs have a tendency to aggregate and interact to form a differentiated organ.

We are skeptical of the alleged lack of nuclei, so we explain the "nucleus" concept to the aliens and propose that CLUs actually have a spatially localized store of information that controls growth and development. The aliens listen carefully and ask clarifying questions in order to understand our hypothesis. Then they tell us that they know we are wrong. Alien scientists long ago developed a method of splitting CLUs which showed that the separated parts of CLUs largely retain their potential for growth and development, even if the CLUs are cut in multiple pieces. Thus, the alien scientists had demonstrated that CLUs and their substances have a hereditary aspect, but the potential for heredity seems to be dispersed in the substances, not centralized in a nucleus.

During the life of an organ, CLUs may come and go. CLUs circulating in the body are harvested continually in the reproductive organs, where substances are extracted to form minute reproductive corpuscles, RCs, whose role in reproduction is similar to gametes. However, the RCs or reproductive corpuscles don't have the 1-copy-of-each-factor neatness of earthly Mendelian gametes. The growable substances in the RCs are variable in amount, thus RCs vary in hereditary potencies. Furthermore, the composition of CLUs circulating in the body reflects the totality of what is happening in the body: because the body is continually growing and reacting and changing, the RCs are changing, too. In particular, the composition of the RCs tends to deviate more strongly when the organisms are stressed or face unusual conditions.

While some of the alien organisms are asexual, others have tri-parental reproduction that involves mixing of RCs from different parents. Each of the 3 parents makes a contribution of RCs, typically equal in size, though in some species, one type of parent contributes much more than the other two. When the parental RCs come together, the substances in them seem to mix or blend.

A different kind of evolution

The alien scientists have outlined the basis of heredity on their planet, and they are looking to us expectantly for ideas about how evolution is going to work. We were hoping to gather more facts, and particularly to hear from other experts about the diversity of life, and so on, but the aliens are eager for our ideas right away. What can we infer about evolution in a bottom-up manner, from an understanding of heredity?

We see immediately that it will be possible to apply some concept of "selection" in this world, but its going to be awfully slippery. We reach into our conceptual toolbox, and the first thing we find is the concept of "selection coefficient". But thats not useful on the alien planet, because there is no stable genetic entity to which one may apply the selection coefficient— everything in the alien world with a bearing on heredity seems to be variable in potency and to be subject to blending. Heredity depends on the differential growth of continuous substances, modulated by their differential incorporation into RCs due to conditions of life, and so on. The alien world lacks the algebraic neatness of pairwise combinations and pure factors.

In fact, our hearts sink as we realize that, because of this blending-together, it might be impossible for evolution to start from a single hereditary variant, as would be possible on earth starting with a single Mendelian mutant. The distinctive features of the individual variant would simply diffuse and blend.

But our discouragement is only temporarily. Yes, it would have been simple and easy if hereditary factors emerged discretely, combined in simple ratios, and maintained their purity during reproduction— but who said science was supposed to be simple and easy?

We are undaunted. We are determined to discover some way to apply the principle of selection. In fact, given that the RCs deviate more strongly under unusual conditions, we note with enthusiasm that extra hereditary variation will emerge just when it would be helpful to provide fuel for adaptation to new conditions! Due to hereditary blending, one variant individual, with a variation in a favorable direction, would not be enough.

But thats not a problem. In fact, to treat it as a "problem" is wrong-headed, because this alien world is not a world of discrete heredity anyway! Instead, on the alien planet, heredity is a bulk process, like the flow and mixing of liquids. The hereditary substances flow (metaphorically) in new directions every generation, and selection can get some leverage from these fluctuations of hereditary potency, even if there is not any single discrete particle to grasp. Selection would guide these fluctuations, building them smoothly from generation to generation. Possibly we could develop a mathematical formalism for this process by adapting the breeder's equation of quantitative genetics, although the shifting of hereditary potencies from one generation to the next would be problematic. An even more radical thought occurs to us: Lamarckian evolution can't happen in our world, but in the alien world, it just might be possible due to the way the RCs reflect what is going on in the body as it experiences its environment.

If you have followed me thus far, congratulations! You are one of the re-discoverers of Darwin's lost theory of evolution!

Evolution without mutation

Sadly, when I refer to a "lost theory", its not a joke, because Darwin's "Natural Selection Theory" (not to be confused with the principle of natural selection2) is largely unknown to contemporary scientists. During the Darwin bicentennial last year, I lost track of how many times "Darwin's theory" was explained by reference to "selection and random mutation" or some such anachronism.

Darwin had no such theory. Given Darwin's assumptions that inheritance is blending (not particulate), that the germ-line is responsive to external conditions (not isolated), and that hereditary potencies shift gradually every generation (not rarely and abruptly, from one pure, stable state to another), it is physically impossible for a rare trait, having arisen by some process, and conferring a fitness advantage of (for example) 2 %, to be passed on to offspring by a stable non-blended hereditary factor, thus conferring on the offspring a 2 % advantage, and for such a process to continue for thousands of generations until the previously rare trait prevails. We may think of evolution in this way: Darwin did not.

Instead, Darwinism 1.0 (Darwin's conception of evolution) is an automatic process of adjustment to altered conditions, dependent on a rampant process of "fluctuation" yielding abundant "infinitesimally small inherited modifications" in response to the effect of altered "conditions of life" on the "sexual organs" (Chs. 1, 2, 4 and 5 of Darwin 1859). Fluctuation was not rare and discrete, but shifted hereditary factors continuously and cumulatively each generation, producing visible effects in "several generations" (Ch. 1 of Darwin 1859). Muller (1956) referred to Darwinian fluctuation as "creeping variation". I have called it "variation on demand", and I also think that, to understand Darwin's view, its helpful to think of heredity and variation as processes mediated by fluids (liquids or gases). Darwin's critics, and quite a few of his friends such as Huxley and Galton, believed that individual "sports" (mutants) could be the start of something new in evolution, but this was not part of Darwin's theory, which invoked blending inheritance and held fast to natura non facit salta.

For those who would like to get more of a flavor of Darwin's view from his own writings, I have included a few passages below in an appendix. Readers may wish to go further by browsing online sources via the links provided. Others may wish to take a colorful look at Darwin's laws of variation from the Virtual Museum of the Origin of Species.

To account for his principles or "laws"3 of variation, Darwin proposed a "gemmule" theory for the mechanism of heredity, where "gemmules" are somewhat like the RCs or "reproductive corpuscles" in the fictional alien world described above.

Although Darwin's "Natural Selection" theory invoked Lamarckian effects, the fluctuation-selection process that Darwin called "Natural Selection" was recognized immediately as its mechanistic core. Only this core mechanism remains in the reformed view of Weismann and Wallace— "Darwinism 1.2" for our purposes—, which expunged Lamarckism and relied on selection of ever-present fluctuations, a process understood (in Darwinism 1.2) as the exclusive and all-powerful driving force of evolution.

Developing a new view of evolution

In fact, the "Mendelians" did not reject the principle of selection. Instead, they rejected "fluctuation" as the basis of evolutionary change for exactly the reason we would expect, namely that these fluctuations are not heritable. Johannsen's experiments were influential because they suggested that the fluctuations that emerge reliably every generation, i.e., Darwin's "endless slight peculiarities which distinguish the individuals of the same species and which cannot be accounted for by inheritance from either parent or from some more remote ancestor", are non-heritable and cannot be the basis for evolution by natural selection.

This is precisely the reason that geneticists gave, explicitly, for rejecting Darwin's view. For instance, in his 1911 book Mendelism, Punnett (of the "Punnett square" one studies in Genetics 101) explains the new "basis of evolution":

"The distinction between these two kinds of variation, so entirely different in their causation, renders it possible to obtain a clearer view of the process of evolution than that recently prevalent. . . Evolution only comes about throught the survival of certain variations and the elimination of others. But to be of any moment in evolutionary change a variation must be inherited. And to be inherited it must be represented in the gametes. This, as we have seen, is the case for those variations which we have termed mutations. For the inheritance of fluctuations, on the other hand, of the variations which result from the direct action of the environment upon the individual, there is no indisputable evidence. Consequently we have no reason for regarding them as playing any part in the production of that succession of temporarily stable forms which we term evolution. In the light of our present knowledge we must regard the mutation as the basis of evolution— as the material upon which natural selection works. For it is the only form of variation of whose heredity we have any certain knowledge.

It is evident that this view of the process of evolution is in some respects at variance with that generally held during the past half century. " (Punnett, 1911, p. 139-140; online source)

Punnett rejects "fluctuations", defined as "the variations which result from the direct action of the environment upon the individual".

It wasn't about rejecting natural selection: Punnett identifies mutation as the "basis" of evolution precisely on the grounds that it provides "the material on which selection works". While TH Morgan (1916) often avoided the phrase "natural selection", as in the following passage, he clearly is not rejecting a role for differential effects of fitness

"evolution has taken place by the incorporation into the race of those mutations that are beneficial to the life and reproduction of the organism" (p. 194) (online source)

This "mutationist" view was merely the start of a new way of looking at evolution. In the next installment, we'll find out what sort of understanding of evolution emerged among this new generation of evolutionists inspired by Mendelian principles. We'll see that, contrary to the Mutationism Myth, the period between the discovery of genetics and the origin of the Modern Synthesis in the 1930s was not a dark period of confusion at all, but a period of innovation that gave rise to key elements of the genetics-based understanding of evolution that persists today, including new ways of understanding selection.

Looking ahead

This post raises several issues that will receive attention in future posts of The Curious Disconnect. For instance, the mutationists rejected "Natural Selection", the theory1 of Darwin, but not the "concept of selection" (as mistakenly asserted by Eldredge, above). In a later post, we will explore how the ambiguity in "natural selection" covers a multitude of sins (e.g., Charlesworth, 2005), and we'll consider ways to speak (and think) more clearly.

A second issue is the cult of personality that has developed around Darwin, which instills in so many scientists the desire to align themselves with Darwin and label themselves "Darwinists" while ignoring Darwin's actual views. Rather than reject or defend Darwin's actual theory, the cultists make personal excuses for Darwin ("he couldn't have known!"), as though science were about judging persons rather than evaluating theories. In a future post, we'll explore the distorting influence of the Darwin Fetish.

A third issue has to do with the structure of Darwin's theory, and more generally, how we determine the structure of a theory, and how the parts fit together. This will become important when we evaluate the deeply problematic claim of the Modern Synthesis to have reconciled Darwin's view with genetics. In essence, the architects of the Modern Synthesis will claim that "the maintenance of abundant infinitesimal variation in the gene pool" replaces "fluctuation" while leaving the rest of Darwin's theory unchanged.


Darwin espoused a theory of evolution, not merely a principle of selection. If this theory merely asserted the principle of selection, then no possible finding in genetics could contradict it. In fact, Darwin's theory invoked the principle of selection, in the context of a mechanism he called "fluctuation", to account for most of the actual facts of evolution, leaving a residue to be explained by other means (Lamarckian and Buffonian effects)— other means that Darwin's followers soon rejected as untenable, leaving only the fluctuation-selection process.

Thus, prior to the discovery of genetics, Darwin's theory was understood correctly to rely on continuous hereditary variation that Darwin called "fluctuation", and that was induced by environmental conditions, not inherited from parents. The Mendelians argued that, if we wish to understand the "the basis of evolution— the material on which selection works", we must look to mutation, not to Darwin's "fluctuations", because variations induced by conditions are not heritable.

Little of this is understood today, because "Darwinism" or "Darwin's theory" has been redefined, and the original meaning of "Darwin's theory" has gone done the proverbial memory hole.


Charlesworth, B. 2005. On the Origins of Novelty and Variation. Science 310:1619-1620.

Darwin, C. 1859. On the Origin of Species. John Murray, London.

Darwin, C. 1883. Variation of Animals and Plants under Domestication. D. Appleton & Co., New York.

Eldredge, N. 2001. The Triumph of Evolution and the Failure of Creationism. W H Freeman & Co.

Ford, E. B. 1938. The Genetic Basis of Adaptation. Pp. 43-56 in G. R. de Beer, ed. Evolution. Clarendon Press, Oxford.

Johannsen, W. L. 1903. Erblichkeit in Populationen und in reinen Linien. Gustav Fischer, Jena.

Morgan, T. H. 1916. A Critique of the Theory of Evolution. Princeton University Press, Princeton, NJ.

Muller, H. J. 1956. On the Relation between Chromosome Changes and Gene Mutation. Brookhaven Symposia in Biology 8:126-147.

Punnett, R. C. 1911. Mendelism. MacMillan.

Appendix: Darwin's principles of heredity

Three passages below illustrate Darwin's view of the emergence of hereditary variation. The first indicates that the emergence of hereditary variation occurs on the scale of a few generations— no waiting around for mutations— and that the fluctuations build up cumulatively:

"It seems clear that organic beings must be exposed during several generations to new conditions to cause any great amount of variation; and that, when the organisation has once begun to vary, it generally continues varying for many generations." (Darwin, 1859, Ch. 1; online source

Darwin knew that "sports" (mutants) could have heritable effects, but he imagined that infinitesimal fluctuations were even more likely to be heritable:

"If strange and rare deviations of structure are really inherited, less strange and commoner deviations may be freely admitted to be heritable. Perhaps the correct way of viewing the whole subject would be, to look at the inheritance of every character whatever as the rule, and non-inheritance as the anomaly" (Darwin, 1859, Ch. 1; online source).

Darwin learned about heredity the hard way: by exchanging hand-written letters with hobbyists and stockmen who bred pigeons, sheep, dogs, and so on. Below he is describing an experiment in domestication of ducks from wild eggs, based on information provided by Mr. Hewitt, a source referenced by Darwin many times in his works:

"Mr. Hewitt found that his young birds always changed and deteriorated in character in the course of two or three generations; notwithstanding that great care was taken to prevent their crossing with tame ducks. After the third generation his birds lost the elegant carriage of the wild species, and began to acquire the gait of the common duck. They increased in size in each generation, and their legs became less fine. The white collar round the neck of the mallard became broader and less regular, and some of the longer primary wing-feathers became more or less white. When this occurred, Mr. Hewitt destroyed nearly the whole of his stock and procured fresh eggs from wild nests; so that he never bred the same family for more than five or six generations. His birds continued to pair together, and never became polygamous like the common domestic duck. I have given these details, because no other case, as far as I know, has been so carefully recorded by a competent observer of the progress of change in wild birds reared for several generations in a domestic condition. "(Darwin, 1883, p. 293; online source)

Thus, Darwin is describing subtle variations that emerge in response to new conditions, and that emerge immediately or, at least, within a few generations. He saw hereditary fluctuation as an effectively continuous process, i.e., a process that can be subdivided arbitrarily in time and in outcome because it is the summation of infinitesimal increments. Adaptation can happen rapidly and reliably because organisms start to vary immediately upon encountering new conditions. Similar variations will be manifested in many individuals (as in the case of the ducks above), so that multiple members of a "race" may emerge and interbreed simultaneously with, or prior to, selection. This avoids the problems posed by the swamping effect of blending inheritance (Darwin did not believe that a solitary variant could begin an evolutionary change). Darwin's principles of variation are roughly that

  • hereditary variation emerges in response to "altered conditions of life" (e.g., domestication);
  • the process is so rapid and productive that visible effects appear in one or a few generations;
  • continuous ("infinitesimal", "insensible") fluctuations occur in virtually all characters;
  • some effects are definite or reliable ("all or nearly all the offspring of individuals exposed to certain coditions during several generations are modified in the same manner"), while others are "indefinite" (isotropic);
  • definite effects reflect mainly internal (developmental) causes, but also external (environmental) and Lamarckian causes ("effects of use and disuse").


1 An updated version of this post will be available at

2 Don't blame me for this egregious ambiguity, which we will address in a future post.

3 Today we would call these laws "principles" or "generalizations". "Laws" in 19th century science are empirical generalizations, reached by the method of Baconian induction: collect lots of facts and distill them into generalizations.


Marcello Pucciarelli said...

Darwin is often said to be a "pluralist", and this quality is also apparent in some of his less celebrated insights:

"It should not ... be overlooked that certain variations ... frequently recur owing to a similar organisation being similarly acted on ... In such cases, if a varying individual did not actually transmit to its offspring its newly-acquired character, it would undoubtedly transmit, as long as the existing conditions remained the same, a still stronger tendency to vary in the same manner. The conditions might indeed act in so energetic and definite a manner as to lead to the same modification in all the individuals of the species without the aid of selection." (page 105 of the 5th edition of the Origins, online at

My questions, for the planned posts: given that The Origins does not mention pangenesis and that, on the other hand, Darwin allows for environmentally induced variations to dispense with natural selection (though only in extreme cases), how integral is the "fluctuational" view of variation to Darwin's theory1 of evolution?; and, is Darwin's a theory1 of evolution, or a theory1 of adaptation? Or, maybe, we have two theories1: a theory1 of adaptation based on natural selection and a theory1 of evolution built around the first with the addition of a whole mess of discredited hypoteses.

Arlin said...


Thanks. You ask 2 questions, one about direct effects of the environment, and the other about the significance of pangenesis, Darwin's mechanistic theory of heredity.

The idea of a direct effect of the environment that changes the members of the species en masse (i.e., in parallel) is associated historically with Buffon, a French naturalist. Darwin's theory1 (as expressed in the _Origin_) relied on 3 means (we might say "mechanisms") of modification: "natural selection" via fluctuation and selection, direct effects a la Buffon, and effects of use & disuse a la Lamarck.

Darwin saw "natural selection" as the predominant means of modification of species, with the other 2 playing a minor role. His followers quickly abandoned the ideas of Lamarck and Buffon, leaving only "natural selection".

So, the answer to your question is that "fluctuation" is indeed integral to Darwin's "natural selection" mechanism, and thus (in turn) integral to Darwin's view.

The fact that pangenesis is not mentioned in the _Origin_ is not relevant to establishing the integral role of fluctuation. Conversely, the lack of a causal mechanism for the law of gravity is not a hindrance to Newton's theory of planetary orbits (which accounted for Kepler's conjecture of elliptical orbits).

The important thing is that Newton got the law of gravity right (even if he didn't know what caused gravity), and that Darwin got the laws of heredity wrong (and didn't know what caused heredity). There is no such thing as a correct mechanism of hereditary fluctuation because the law itself is wrong-- fluctuations as defined by Darwin (see the specific quotations in the text) are not heritable.

Anonymous said...

I'm not convinced that characterising Darwin's view as dependent on essentially un-inheritable 'fluctuation' is entirely accurate. The word does not appear in the text, and he does come out pretty clearly as favouring a Lamarckian view. This is wrong, but a Lamarckian mechanism would fit the bill as a source of heritable variation. He openly admits to not actually knowing these 'laws'. The lack of a causal mechanism was as little a hindrance to Darwin as Newton in formulating principles derived from observation and deduction.

Variation is integral to Darwin's Natural selection mechanism - variation alone in individual lives; heritable variation in the multiple.

Owing to this struggle for life, any variation, however slight and from whatever cause proceeding, if it be in any degree profitable to an individual of any species [...] will tend to the preservation of that individual, and will generally be inherited by its offspring. The offspring, also, will thus have a better chance of surviving. [...] I have called this principle, by which each slight variation, if useful, is preserved, by the term of Natural Selection, in order to mark its relation to man's power of selection.

Arlin said...


You are right that "fluctuation" is not used this way in the _Origin_ (though Darwin used it elsewhere), where the preferred term is "indefinite" "variation" (or "variability"). After the discovery of genetics, "fluctuation" or "fluctuating variability" were the terms used by geneticists.

To achieve clarity, we avoid ambiguous tokens (e.g., "variation") and focus on the underlying concepts. Did Darwin say what he meant by "indefinite variability", a phenomenon that later writers called "fluctuation"? Yes, he did. I used Darwin's meaning, and I showed that geneticists did, too.

Superficially, it might seem that your claim:

"Variation is integral to Darwin's Natural selection mechanism"

salvages Darwin's theory and is different from

"Fluctuation is integral to Darwin's Natural selection mechanism"

but thats just a superficial appearance. Whats important is not what you mean by "variation", but what Darwin meant. So, lets substitute his definition:

"Endless slight peculiarities which distinguish the individuals of the same species and which cannot be accounted for by inheritance from either parent or from some more remote ancestor" are integral to Darwin's Natural selection mechanism.


Arlin said...


Sorry for being long-winded. Did that address your concern?

I don't think your quoted passage contradicts what I said, once we confront the maddeningly ambiguous use of "natural selection" by Darwin and his followers. Here, Darwin is stating that a principle of selection could apply to heritable variations of any type-- including the monstrous "sports" that held no significance in his theory1. That is, he is expounding the selection principle (theory2).

But in his theory1, Darwin puts his money on "fluctuation" ("indefinite variability"). Real theories1 take risks, and this was one of Darwin's.

Anonymous said...


Thanks for taking the trouble!

NS is essentially a theory of phenotypes. Darwin was happy to deal with infinitesimal gradations in those, and so are we - phenotypes (often) 'smooth out' the underlying digital reality, just as music sounds continuous from the 1's and 0's on a cd.

Selection*** doesn't need inheritance, but obviously it won't do the job Darwin needed it to - adaptation - without. Of course, the now-known disconnect between genotype and phenotype allows for infinitesimal change in either to be coupled with huge change in the other.

But still, 'fluctuation' (infinitesimal variation) would work with NS IMO. He saddled himself with a difficulty by natura non facit saltus, perhaps, but NS can still bias a spectrum, given heritability. If your final quote is restrictive to environmental effects alone, then fair enough, but it isn't how I read him. He seems to think of variation arising in response to need, perhaps.

*** I'd note the differing uses of 'selection' in biology. I've seen drift called non-selective selection, and in similar vein "NS is the process that causes change in allele frequency" ... which doesn't leave much distinguishing work for the s-word to do, for all that s & d work in tandem!

Arlin said...


Thanks. You said "If your final quote is restrictive to environmental effects alone, then fair enough, but it isn't how I read him."

I'm having trouble seeing what "reading" of Darwin will squeeze heritable effects into his conception of fluctuation, given their induction by changes in "conditions of life"-- which for me raises a red flag that he is talking about environmental variation. For instance, how do you "read" the passage about ducks that I gave in the appendix, or the passage below?

"It seems clear that organic beings must be exposed during several generations to new conditions to cause any great amount of variation; and that, when the organisation has once begun to vary, it generally continues varying for many generations." (Darwin, 1859, Ch. 1)


Marcello Pucciarelli said...


I think I'm getting close to understanding your views, though I'm getting a slightly different picture.

In your first reply to Allan you talk of "endless slight peculiarities which distinguish the individuals of the same species and which cannot be accounted for by inheritance". I think that the first clause (individual differences) is the essential one, while the second (unaccounted for by inheritance) is accessory. It suggests an idea of novelty, which seems at odds with the fluctuational view.

Confronted by Fleeming Jenkin's objections, Darwin sees that the probability of a variation being passed to the next generation depends both on its utility and on the number of individuals already possessing it. So, he concludes against the importance of any "single variation" -- including but not limited to the "monstrous" sports: "I did not appreciate how rarely single variations, whether slight or strongly-marked, could be perpetuated." (Origin, 5th ed, p. 104)

This suggests that the favorable variations (individual differences) must already be present in at least a small subset of the population. Thus, inheritance would indeed account for their presence.

I'm not saying that Darwin consistently held this view, where an individual variation is put against the background of the current generation, not of previous generations. Anyway, it is consistent with a fluctuational view because it does not require the introduction of novelty (and of a mechanism for their origination) from one generation to the next.

Anonymous said...


Hmmm. "x is integral to Darwin's Natural Selection mechanism".

I think we can agree that selection is integral - this demands variants to select among.

Given that his core analogy relates to iterative domestic breeding, I think we can agree that heredity is integral also.

But on whether his mechanisms of variation and heredity are integral, we might agree to disagree. He flounders, for sure (you feel like shouting across the years at him: "Look! Over here!" Dawkins notes how close he came to spotting particulate inheritance).

I take your point, that without an effective 'clutch' coupling Variation, Selection and Heredity, selection can do no work, and Darwin's 'fluctuational' clutch would slip.

His problem is sexual organisms (or Biology, as many biologists seem to view it!). If your aliens were asexual, their evolutionary world would be more sustainable.

Marcello Pucciarelli said...

Here is a simple attitudinal test. How should we treat the issue of Lamarckism in Darwin? Answers:

"Darwin allowed Lamarckian mechanisms of variation, but..."

1. other mechanisms of variation can be - and have been - plugged-in to serve as the raw material of natural selection.

2. Darwin denied that Lamarckian mechanisms can play a significant role in evolution.

3. Darwin's own view of variation is not directional.

All three answers are correct, but they have different implications:

1. variation could be anything that can be subject to natural selection, and we don't care about what Darwin (or anybody else) really thought.

2. Darwin envisioned mechanisms, or modes of variation, other and more important than Lamarckian ones. But we don't care very much about which mechanisms or modes of variation (they are quite obviously outdated by modern standards).

3. Darwin's own views of variation are definite enough to exclude some mechanisms from the core of his theory. Thus, we may do to them what we usually do to scientific ideas generally -- accept or reject.

We can replace Lamarckian variations with discontinuous changes and redo the exercise.

Arlin said...

Marcello, you wrote:

"I think that the first clause (individual differences) is the essential one, while the second (unaccounted for by inheritance) is accessory. It suggests an idea of novelty, which seems at odds with the fluctuational view."

Both clauses are Darwin's words. They are part of his theory whether you like it or not.

But-- just out of curiosity, why does new variation seem to you to be at odds with the fluctuational view that Darwin described in his own words?


Arlin said...

Marcello, thanks again for your participation. This kind of extended dialog is the only way to develop consensus and its the sort of thing that I was hoping to happen when I started this blog.

You wrote:

"This [conclusion that single variations can't be the start of something new in evolution] suggests that the favorable variations (individual differences) must already be present in at least a small subset of the population. Thus, inheritance would indeed account for their presence."

Again, I'm trying to explain Darwin's theory, as expressed by Darwin and as understood by scientists a century ago. So, if Darwin wants to invoke the emergence of hereditary fluctuations due to the effect of "conditions of life" on the sexual organs, thats his theory, and if Marcello wants to claim that all of the variation was there before, thats a different theory.

Its important to understand that in Darwin's view, indefinite variability (fluctuating variability) is a bulk process in which hereditary potencies (of the race) flow out in all directions (directions in morphospace). Reproduction blends components. These are two of Darwin's laws of heredity. In a system dominated by such laws, heredity does not provide any inertia: as soon as members of the race move into a new environment, they fluctuate, and these fluctuations all blend together during reproduction to make a new mixture. The race has no fixed "genotype". Instead, it just has whatever is the current blend or mixture of hereditary potencies, all of which will flow and blend again in the next generation. In this view, heredity cannot be the reason that a race would stay the same for millennia-- its selection that keeps things in place (no wonder Darwin thought it was so strong!!).

Many individuals may flow in the same direction, resulting in a local "race". This is Darwin's answer to the objection of Fleeming Jenkins. Darwin made this possibility clear in the very same passage that Marcello quoted in his first comment: "The conditions might indeed act in so energetic and definite a manner as to lead to the same modification in all the individuals of the species without the aid of selection." (page 105 of the 5th edition of the Origins, online at


Marcello Pucciarelli said...

thank you for taking the time for answering my objections. Sorry for my inability to make things short.

We agree that Darwin was an externalist about variation: altered conditions of life cause variations to occur; more importantly, altered conditions of life trigger a "tendency to vary" which, once set in motion, would not stop and could be transmitted. When Arlin says that "heredity does not provide any inertia" I read him as saying that heredity can transmit the tendency to vary, and no force is necessary in order to produce further changes in a varied/variable trait.

So far, I think we agree. But we are not finished yet, because (IMO) we need to reconcile this view with the "classical" view of inheritance as pertaining to traits. Arlin's message seems to be that if we want to understand Darwin's theory we need to forget about the inheritance of traits: the apparent stability of traits is the product of selection, not of inheritance.

I'm not following this advice because there are many references in Origin to a "principle of inheritance": "like produces like". Darwin (IMO) used the principle to make the case for natural selection stronger -- in Arlin's terms, it provided the inertia that was missing from Darwin's own model of heredity. So I don't think that Darwin held the principle as true, or explanatory enough -- he just saw that it did a good job in securing the role of natural selection.

On the other hand, in Variations Darwin attacked the principle directly, showing its problems and how they could be solved by pangenesis. There is a tension here, and not the only one. A quick search on the online texts of Origin shows that when Darwin writes of new characters, they are either acquired or failing to be transmitted. I may be wrong, but I couldn't find a positive example of a coherent process of origination (through "blind", undirected variation) and transmission/selection. On the contrary, references to new characters and their transmission abound in Variations.

So, I was wrong to maintain that favorable individual differences must already be present in the population. But I still hold my contention that by emphasizing the importance of individual differences (in Origin) Darwin is telling us that it doesn't matter whether such
variations are new compared to the previous generation. A "single variation", on the other hand, would be a true novelty, and in answering Jenkin's objections Darwin made clear that his bets were on individual differences and not on single variations (including small ones).

Notice that we can maintain a fluctuational view based on individual differences, as a process involving quantitative changes in a morphospace with a given number of dimensions. This is why I said that the introduction of novel traits (new dimensions) would be at odds with the fluctuational view.

Recapitulation and further speculations: there are tensions in Darwin's views of variation and natural selection, probably due to the fact that they are both externalist. Darwin managed somehow to prevent his views of variation to overflow into natural selection, thanks to a handy "principle of inheritance" that he actually didn't see as explanatory. By the way, the principle is not true. The standard view is that a non-blending model of inheritance would have served Darwin better, but I agree with Arlin that we cannot just plug in the right mechanism of inheritance and expect the theory(1) to work.

Arlin said...

Marcello wrote:

"Arlin's message seems to be that if we want to understand Darwin's theory we need to forget about the inheritance of traits: the apparent stability of traits is the product of selection, not of inheritance."

That is NOT AT ALL a fair summary of what I said! Of course I understand that in Darwin's view "like begets like", and I said nothing to contradict this. In Darwin's view, like begets like because the hereditary substances are passed from parent to offspring.

My point was that, to understand Darwin's view, its helpful to think of the substances that mediate heredity and variation as fluid substances that flow and blend.

The Mendelian theory is not like this. Its discretized, in the sense that there is a pure and stable genetic type or genotype that is passed on. Unless the hereditary factor undergoes a rare mutational change, its perfectly unchanged and unblended and unaffected by environment.

In the Mendelian view, selection need only to eliminate the rare mutants in order to keep the population in place. In Darwin's view, the population fluctuates and remixes itself every generation.