Posted by: Dan | November 14, 2007

Macroevolution and Microevolution

NOTE: This post is a year and a half old now (April 2009 now), and it’s still being linked to by people occasionally reading it as a support for the idea that Macroevolution = Microevolution plus Time. Please see down in the comments – I was shown to be incorrect on that. I was wrong. I misunderstood what evolutionary biologists mean when they use those terms, especially macroevolution. The key missing ingredient was *reproductive isolation*. Please leave a comment if you have further questions, or see John Wilkins’ explanation of Macroevolution: Its Definition, Philosophy and History, on

Via Greta:

In other words, microevolution + time = macroevolution. Larry Moran takes issue with this, citing Gould, Eldredge, and the concept of “Species sorting.” Somehow, despite requests for explanation, Larry will not or cannot explain how species sorting cannot be explained by microevolutionary mechanisms of species competition.

Eldredge (1995) defines “species sorting” (also called “species selection) as “differential speciation or extinction of species within a larger group”; he clarifies that “Some lineages speciate at a higher rate than others, and some species are more prone to extinction than others” (119). These varying rates of speciation and extinction, according to Eldredge, produce definite patterns in the fossil record.

Differential rates of speciation and extinction sounds a lot like, well, a speciation process. Elsewhere, Eldredge and Gould are muddled in their descriptions of how macroevolution cannot be reduced to speciation processes, saying only that “macroevolution must be studied at its own level.” Great, we already know that ‘zooming out’ to view large-scale patterns of many small-scale events is a good strategy, just as a physician would be myopic to diagnose a patient he’s never seen or spoken to based on a single biopsy. But we still say that patients are made out of the cells collected during a biopsy.

Larry goes on to just make an ass out of himself, telling us we’re pitiful for not reading his essay that he spent 10 years writing on macroevolution. Except we did read it, and as Sven DiMilo said in the comments:

I read the essay. In fact, I just clicked on your link and read it again. I’m sorry to report that it didn’t help to clarify my understanding of why the “sufficiency of microevolution hypothesis” ought to be discarded. After a fine treatment of the history of this controversy, there are simply assertions, by you and quoted from Gould and Eldredge, that microevolution is insufficient. But see, that’s what I’m not getting–why is it insufficient?

That higher-level patterns are identifiable in hindsight does not require that new mechanisms are necessary. And statements like “Since speciation is not a direct consequence of changes in the frequencies of alleles in a population, it follows that microevolution is not sufficient to explain all of evolution.” don’t help–it’s the premise I don’t get. Why is speciation more than the simultaneous changes in allele frequencies in two populaitons that lead to reproductive isolation?

Larry continues with assertions, explains that he’s under no obligation to explain his assertions, tells us to go read his essay and leave him alone, and that his critics aren’t getting it (i.e. they’re not being credulous). Oh, and his best explanation “read up on species sorting” falls flat, because he’s skeptical that it’s valid – yep, you heard that right – we’re supposed to buy into microevolution being insufficient for explaining macroevolution based on an argument that he himself finds questionable.

Come on Larry – if it’s microevolution + time + X = macroevolution, then what is X? Transmogrification?

PS – Yes, as Sven also said:

But you’re right–you’re under no obligation to explain or try to convince me. And likewise, I’m under no obligation to buy into your unexplained conclusions just because you’ve been thinking about it for 10 years.

  • Niles Eldredge (1995) Reinventing Darwin: The Great Debate at the High Table of Evolutionary Theory, Wiley.
  • Stephen Jay Gould (1980) The Structure of Evolutionary Theory, Belknap Press.


  1. And here I always thought it was just that I wasn’t educated enough to get it.

  2. That appears to be what Larry would have you (and me) think.

    To think, here’s a fellow scientist, a biologist, an atheist, a rationalist – who thinks that argument from assertion and authority is sufficient. Sounds kinda like Behe’s obnoxious rebuttal to Abbie Smith (ERV) to me…

  3. There is genuine scientific controversy over the existence of higher level processes that are decoupled from natural selection and random genetic drift.

    Anyone familiar with the scientific literature knows this. There are entire books on the subject of which Gould’s “The Structure of Evolutionary Theory” is the most comprehensive.

    Whether or not I personally believe in the existence of these higher level processes (e.g., species sorting) is not relevant to the point I’m trying to make.

    The legitimate scientific statement should not be a dogmatic assertion that macroevolution is simply lots of microevolution. Especially if such a statement is based on an incomplete knowledge of evolutionary theory – as so many such statements are. Instead, it would be more accurate to say there is debate within the scientific community about the sufficiency of evolution. Then you can give your OPINION about which side of the debate you favor.

    The other problem with the simplistic assertion is that it doesn’t jibe with the way evolutionary biologists use the term “macroevolution.” Macroevolution – the scientific version not the creationist version – is a discipline concerned with working out and explaining the exact history of life on Earth. It combines evolution and other natural events, such as ice ages and asteroid impacts, to arrive at this history.

    That’s the sense in which a lot of macroevolutionary biologists object to the simplistic statement that macroevolution is just a lot of microevolution. These evolutionary biologist wonder how the people who make such statements can exlain the extinction of dinosaurs using microevolution. I tried to explain this point in my essay.

    Microevolution is not sufficient to explain the history of life (= macroevolution) just like relativity and quantum mechanics are not sufficient to explain how our particular solar system came into being as opposed to millions of other possibilities.

    Dan, the concepts of species sorting and emergent properties at higher levels are complicated. They’re not the sort of thing you can easily grasp by just reading one of two papers. At least, I wasn’t able to grasp them so easily – I suppose you might be cleverer than me.

    One of the reasons I don’t want to get into defending these higher level processes is that I don’t fully understand them. Unlike you, however, I have respect for those very smart people who advocate hierarchical theory. I don’t dismiss them as idiots after a superficial look at the concept. There seems to be widespread agreement among evolutionary theorists that species sorting, correctly understood, really is a high level process. The debate is about whether there are any true examples of species sorting.

    “Come on Larry – if it’s microevolution + time + X = macroevolution, then what is X? Transmogrification?”

    I don’t really want to answer this question because it’s set up in a way that does not reflect the true debate. In real scientific terms, you can’t treat microevolution and macroevolution as two sides of an equation because you’re comparing apples and oranges. The way you pose the question presupposes the answer you desire.

    However, here are three kinds of “X” that you should think about.

    X= cladogenesis
    X= mass extinctions
    X= species sorting

  4. So we’re supposed to suspend disbelief because someone high-ranking in the echelon of science says so? I don’t think that I’m being dogmatic – I think I’m being skeptical. Because it certainly appears that all of your suggestions for “X” can be reduced to something demonstrable (at least in concept) in microevolutionary scales.

    Also, I didn’t call Gould, Eldredge, or you “idiots.” I simply said I won’t accept everything you say, just because you say so.

  5. Dan says,

    “I think I’m being skeptical.”

    Are you being skeptical of the idea that macroevolution is just lots of microevolution? Or, do you run around stating that as though it were a proven fact and everyone who disagrees with you is under some obligation to educate you about evolutionary theory?

    “Because it certainly appears that all of your suggestions for “X” can be reduced to something demonstrable (at least in concept) in microevolutionary scales.”

    I think you should write up your ideas and see if you can get them published. It would certainly shake up the field.

    “Also, I didn’t call Gould, Eldredge, or you “idiots.” I simply said I won’t accept everything you say, just because you say so.”

    Good for you. You don’t accept anything I say or any of the quotations from famous evolutionary biologists. You are perfectly free to do so. (BTW, I don’t accept what you say either.)

    Now that you have been alerted to the problem why not go and learn about it before rejecting the ideas of these very prominent evolutionary biologists?

    You don’t sound very “skeptical” to me. You sound as though your mind is already made up and you don’t want to learn about anything that conflicts with your old ideas.

  6. Am I being skeptical of the idea that macroevolution can be conceptually viewed as just microevolution plus time? Come on. That’s the base position. You’re not suggesting that microevolution is not necessary for macroevolution any more than I am.

    I think you should write up your ideas and see if you can get them published. It would certainly shake up the field.

    I’ll ask around.

    You don’t accept anything I say or any of the quotations from famous evolutionary biologists. You are perfectly free to do so. (BTW, I don’t accept what you say either.)

    Shall we teach the controversy then?

  7. Larry,
    By the way, Berkeley’s Evolution 101: Macroevolution says what I’ve been saying all along:

    Macroevolution encompasses the grandest trends and transformations in evolution, such as the origin of mammals and the radiation of flowering plants. Macroevolutionary patterns are generally what we see when we look at the large-scale history of life.

    It is not necessarily easy to “see” macroevolutionary history; there are no firsthand accounts to be read. Instead, we reconstruct the history of life using all available evidence: geology, fossils, and living organisms.

    Once we’ve figured out what evolutionary events have taken place, we try to figure out how they happened. Just as in microevolution, basic evolutionary mechanisms like mutation, migration, genetic drift, and natural selection are at work and can help explain many large-scale patterns in the history of life.

    The basic evolutionary mechanisms—mutation, migration, genetic drift, and natural selection—can produce major evolutionary change if given enough time.

    So I’m really astounded that you would find my comments capable of “shaking up the field.”

  8. Not that I necessarily want to drop a handful of sand into all of these gears, but I would like to suggest that the common assumption that microevolution takes only relatively brief periods of time, but macroevolution takes much, much longer is exactly reversed. Indeed, if one understands the underlying principles of cladogenesis, it is quite literally the case that macroevolution can occur in as little as one generation, whereas microevolution necessarily requires quite a few more.

    To understand why, consider the standard illustrations of cladogenesis and anagenesis that appear in most evolution textbooks (here’s one). As you can see, the purple circles indicate either anagenesis (transformation within a lineage) or cladogenesis (splitting of a lineage, usually into two). In the diagram shown at the link, it should be clear that cladogenesis events take less time than the anagenesis events; that’s why the shading around the former are small circles, while the shading around the latter are elongated ellipses.

    Indeed, if you think about it, what is the actual (i.e. demographic) event that is depicted as cladogenesis (i.e. inside the purple circles)? It’s the divergence of one line from another. And what is the minimum number of individuals that such a process requires? One, especially if that individual reproduces asexually (as can almost all organisms except animals). And over a minimum of how many generations? Again, one.

    Furthermore, the amount of phenotypic change required to split one lineage from another is (as J. B. S. Haldane once pointed out) so slight as to be statistically undetectable. All that has to happen is that the descendants of a newly divergent individual remain divergent (and at least partially reproductively isolated) from the lineage from which they were derived. Then, given sufficient time for anagenetic transformation within that individual’s lineage (and the lineage from it diverged), we will eventually see the kind of divergence that is usually considered to be representative of macroevolution.

    That this is the case should be obvious: mass extinctions caused by global ecological catastrophes can produce massive cladogenetic change in literally a few hours. For example, if some of the more extreme versions of the K-T mass extinction are credible (and there is increasing evidence that they are), then the adaptive radiation of the survivors of this cataclysm can be considered to have begun diverging on the day after the bolide impact. That is, without most of their former arcosaurian competition, such survivors would immediately have begun adaptively radiating, a process that would have been greatly enhanced by the relaxation of selection (on the survivors) that generally accompanies the emptying of multiple ecological niches as the result of a massive ecological change.

    During such a period of “ecological release” the products of the various “engines of variation” among the survivors are not pruned by natural selection, and so lineages descending from “peculiar” individuals (that is, those which would normally have been almost immediately eliminated) would continue to persist long enough for selection to cause anagenetic transmutation among their descendants, while the relaxation of selection would also persist long enough for the “engines of variation” to produce multiple “hopeful monsters” among their descendants, sufficient to produce multiple diverging lines.

    This is not Goldschmidt redux, however. The “neo-hopeful monsters” described above need not be hugely phenotypically divergent to found new clades (although such large-scale divergence is clearly possible). All that needs to happen for such adaptive radiation to take place is for selection to be relaxed for a sufficiently long period to allow for multiple divergent lines to gain a toehold in the newly emerging ecological regimes to which they are beginning to adapt.

    And so, rather than macroevolution being simply microevolution (i.e. selection and drift) extended over deep evolutionary time, macroevolution is a genuinely different process that occurs in the absence of most microevolutionary processes (i.e. the relaxation, not the intensification, of natural selection).

    And Gould could very easily have been correct about “species sorting”, if one assumes that a new species begins with the divergence of a new lineage, beginning with as few as a single individual in as little as a single generation, a single individual that need not be noticeably phenotypically different from its former conspecifics.

    Furthermore, the classical requirement of reproductive isolation (generally considered to be an effect of allopatry) is also inoperative. Although allopatry is certainly effective at providing the conditions for cladogenetic divergence, it is not necessary. All that is necessary is that the newly divergent line remain reproductively isolated from the line from which it has diverged. Plants can do this in a single generation as a result of a suite of genetic mechanisms (most of them involving various forms of polyploidy), but it is not impossible for animals to do so as well, as recent empirical studies have demonstrated.

    Add the newly emerging fields of evo-devo and epigenesis to the foregoing, and it is increasingly clear that macroevolution (i.e. cladogenesis) follows different rules than microevolution (i.e. anagenesis), and that these differences are most noticeable in the fossil record cited by Eldredge and Gould as the basis for their theory of punctuated equilibrium. In particular, the basic program that energized the “modern synthesis” – that is, the reduction of all significant evolutionary mechanisms to a series of linked mathematical models, based on grossly simplified reductions of complex biology to quasi-Mendelian point-like “particles of inheritance” (changes in which drive the variation and divergence of phenotypes) – is impossible to apply in any coherent way to macroevolution. The “modern synthesis” was essentially a “Newtonian” program, whose proponents assumed that the underlying law-like processes (i.e. microevoluiton) are (like physics) both ahistorical and universal. However, it is now becoming clear that the emerging science of macroevolution is both irreversibly historically contingent (and therefore not reducible to mathematical formalisms) and driven by fundamentally different processes than those underlying most of microevolution.

    At least, that’s how it looks to me.
    Allen D. MacNeill, Senior Lecturer
    The Biology Learning Skills Center
    G-24 Stimson Hall, Cornell University
    Ithaca, New York 14853
    phone: 607-255-3357 (Allen’s office)
    “I had at last got a theory by which to work”
    -The Autobiography of Charles Darwin

  9. Larry,
    I solicited Allen’s perspective to help resolve this issue.


    Indeed, if one understands the underlying principles of cladogenesis, it is quite literally the case that macroevolution can occur in as little as one generation, whereas microevolution necessarily requires quite a few more.

    I don’t doubt that this is probably the case, and maybe I need to revise my definition of microevolution as such. Maybe that’s what I’m missing. Whatever the case, I am explicitly arguing that macroevolution could be, at least in theory, resolvable to individual generations. Something occurring in individual generations surely sounds like a microevolutionary event. Couple with the fact that the basic evolutionary mechanisms—mutation, migration, genetic drift, and natural selection—can produce major evolutionary change if given enough time, and that these mechanisms are what underly both macro and micro-evolution, I think that we’re getting caught up in semantics.

    Something resolvable to a single generation is (by definition) a microevolutionary event, right? The idea that this would be incorrect is extremely counter-intuitive to me.

    As point of fact – you suggest that microevolutionary change cannot be resolvable to one generation, but macroevolutionary change can. Why is it not both? I.e., why cannot most generation-to-generation shifts be subtle micro-scale shifts, while once in a great length of time, a significant change (describable in microevolutionary terms) might occur that becomes fixated, extremely successful, and result in adaptive radiation leading to a new clade?

    If what you’re saying is that gradualism is out – then yes, I’m aware of that. I’m not contesting punctuated equilibrium or the sort. I’m just going with an intuitive understanding of evolutionary biology, and admit that this is not my field precisely. Regardless, both macroevolution and microevolution appear to be man-made descriptions created in hindsight, and the basic evolutionary mechanisms—mutation, migration, genetic drift, and natural selection—are in play. Changing environmental factors and geographies are involved in both, including dramatic shifts that could in concept occur in a day, but only time distinguishes them, right?

    I still don’t get it.

  10. More thoughts:
    Of much greater help has been the TalkOrigins entry on Macroevolution: Its definition, philosophy, and history, particularly the sections on Confusions and those following it. In this section, John Wilkins (the author) explains some of the various and contradictory definitions of micro- versus macroevolution.

    I have a feeling that this is the controversy that Larry is talking about. If so (and please let me know if I’m correct or not), then that’s fine. I’m naturally working under the assumption though that we’re trying to explain natural history by mechanisms (mutation, migration, genetic drift, and natural selection) and environments. You may not be working under the same definition, as intuitively obvious to me as it may seem. As a result, we may be talking past one another.

    Additionally, the TalkOrigins entry has a section answering “Is macroevolution reducible to microevolution?” Again, it comes down to semantics:

    From a philosophical perspective, one might say macroevolution is just a bunch of microevolution. It’s also just a bunch of chemistry. And physics. These are unhelpful answers, so we might find it worthwhile to ask how scientific domains relate to each other. Whenever a scientist or philosopher asks if two theories are reducible one to the other, there are several answers that can be given.

    It goes on to distinguish between types of reductionism:

    The methodological [reductionist] claim that macroevolution (Ma) reduces to microevolution (Mi) is a claim that the optimal solution for investigating evolution is to apply modelling and testing by genetic techniques. And this has been very successful. However, it has not been an unqualified success – developmental biology is not easily reducible to genetics, nor is ecology.

    Yes, in practice the macroevolution of development and ecology based on mechanisms of heredity are too complex to be enumerated solely in terms of genetics. In concept however, they are, assuming that we had the ability to handle that volume of data.

    Moreover, there are many other things involved in development: epigenetic factors (para-genetic inheritance and environmental modulation of genetic effects), cytological inheritance (organelles, cell membranes, ribosomes and enzymes from parent cells, and parent organisms). So genes on their own are not enough to explain why evolution occurs along the pathways that it has.

    Agreed. But I have been referring to hereditary mechanisms, which do not preclude such epigenetic factors.


    [the ontological reductionist challenge against reductionism]: There are entities and processes, they say, that affect macroevolutionary dynamics which are not in their nature microevolutionary. What could these be?

    Well, a list that reductionists would accept includes climate change, geomorphological processes like mountain building, tectonic isolation and drift, vulcanism, extraterrestrial influences like bolide impacts, galactic wobble, precession of the earth’s axial rotation, and possibly even local stars approaching and changing the impact on the earth of comets and other bolides in a cycle averaging around 13 million years. The point the reductionists would make, though, is that everything that these things affect is microevolutionary – only the frequencies of genes in populations, and so on. They serve as the environment in which genes change their frequency (or fail to, and the species goes extinct). What the “player” is in microevolution is the population, comprising organisms, traits and genes; in short, the gene pool. Nothing else is important.

    Nonreductionists will argue, however, that there are emergent processes and entities in macroevolution that cannot be captured ontologically. There are several candidates for these, each challenged by reductionists. The basis for this is a view of evolution as a series of inclusive hierarchical levels, each of which is somewhat independent of the lower levels.

    And more, on emergent properties:

    One of the claims made by nonreductionists is that evolution occurs on emergent properties. An emergent property is one in which the property of a higher level system or object cannot be reduced to the properties of its constituent elements, but instead it “emerges” from the interactions between them (O’Connor and Wong 2002, Mandik 2004). Emergent properties were first proposed by, coincidentally, a friend of Darwin’s, G. H. Lewes, in the field of psychology, but the idea goes back to J. S. Mill in 1843. It is often sloganized as “the whole is more than the sum of its parts”. Emergence was made an issue when applied, ironically enough, to evolution in the 1920s by Jan Smuts and C. D. Broad.

    In this writer’s opinion, an emergent property is simply a property that we have trouble computing or predicting from a knowledge of the constituent parts, but this simple dismissal is insufficient. We have to identify the following aspects of the matter:

    E: The environmental factors in which a species exists – for example, geological and climatological changes

    O: the properties of the organisms – for example their traits and capacities severally

    I: the interactions between the organisms – for example, the lineages of heredity at the gene, haplotype, genome and developmental levels of organization. Also, the issue of organisms changing their environment through a feedback process known as “niche construction” affects both E and I (Oyama et al. 2000). We can set up the reductionist position like this:

    Reductionism: I(O & E) → Ma

    Ma is therefore the result of the union, in some way, of E, O and I. This can be massively complex and give rise to “sudden” changes [note 3], or hold the evolutionary process in a state of stasis for long periods. Whether or not one wants to call this “repeated rounds of microevolution” or not (Erwin 2000) is open to debate. And even if it is, we still need to know the models for how they relate, and what Mi covers.

    Yes, Larry, there are legitimate scientists who think that such reductionism is not valid. They posit some unknown or poorly-defined factor that happens, but outside of the standard generation-to-generation framework. That’s why others have offered the saltation, orthogenesis or (mockingly, in my case) transmogrification options. These appear to go against everything I’ve ever learned in biology – because continuous heredity throughout the epochs is assumed, and anything occurring between generations and involving heredity of some sort is necessarily microevolution, by my definition.

    This holistic description of emergent properties as something not expressible in generational timescales is what I take you to be arguing for. Hence my ardent criticism. If I’m mistaken Larry, then please say so.

    Allen, does this make more sense? (thanks for the comment, BTW!)

  11. Perhaps I should add Wilkins conclusions, which match my own quite well:

    Ontologically, all the objects of Ma are accounted for by the objects of Mi plus the objects and processes of E. However, we can’t just assume all the processes of Ma are just the aggregate sum of the processes of Mi – this needs to be shown. Methodologically, we can not predict the outcomes of Ma from a knowledge of the states of Mi plus E. This is not because the outcomes of Ma are not the result of Mi and so on, necessarily, but because we cannot compute in a reasonable time those outcomes – too many variables, conditions, and interconnections (Dupré 1993, Rosenberg 1994).

    So, contrary to Larry’s blowhard assertion, there’s nothing unconventional about my position at all. The alternative, however, in Wilkins’ words (since mine get a snarky response from Larry):

    The alternative, nonreductionism, posits that there are properties and processes going on that cannot be reduced to E, O, and Ialone. There are some other things happening, call them M, that need to be added into the mix.

    Nonreductionism: M & I(O & E) → Ma

    The arguments in biology are therefore concerned with what the set of Ms might be, and how they operate.

    To follow that line of thinking though, “M” appears definable as “I(O & E)”, and the equation collapses into the reductionist form.

  12. I think you’re right: this is basically a problem of semantics. The “classical” definition of microevolution (from the “modern evolutionary synthesis”) is “change in the allele frequencies in a population over time.” This is why when teaching evolutionary theory, evolutionary biologists generally introduce the topic of microevolution by presenting the Hardy-Weinberg-Castle genetic equilibrium law, and then explore the microevolutionary consequences of violating one or more of its five necessary preconditions.

    This means that microevolution is driven by processes that cause changes in allele frequencies in populations over time. The “big three” are natural selection, sexual selection (often, but not always, considered separately from “pure” natural selection), and genetic drift. There are others, of course: meiotic drive is an exotic example.

    Note that mutation by itself is generally not considered to be a microevolutionary process. This is because mutation alone almost never changes allele frequencies. This is because the mutation rate is, of necessity, extremely low (anything higher than about one mpm is generally fatal or strongly selected against). Therefore, significant changes in allele frequencies as the result of mutation alone would only happen in such tiny populations that they would be effectively equivalent to (and therefore indistinguishable from) random genetic drift.

    Another way to think about this is that micromutation is about changes in allele frequencies within anagenetically changing clades, whereas macromutation is about changes in gross morphology between cladogenetically diverging clades. Defined in these ways, I hope it clear to see that neither one can be reduced to the other; they involve fundamentally different processes over fundamentally different time frames (although, as my earlier post on the subject was intended to point out, the time scales involved are actually the reverse of what most people assume).

    In sum:

    mutation (all of the 47+ processes listed in my post at The Evolution List) is the “fuel” for the vehicle,

    micromutation (i.e. natural selection, sexual selection, and genetic drift) is the “engine” for the vehicle,

    • the environment is the “driver” of the vehicle,

    macroevolution is the track laid down on the landscape by the vehicle,

    time is the “landscape”, and

    • an evolving population is the “vehicle”.

  13. Dan asked:

    “…microevolutionary change cannot be resolvable to one generation, but macroevolutionary change can. Why is it not both?”

    Given the definition of microevolution as “changes in allele frequencies in a population over time”, then macroevolution (defined as the divergence of a population into two or more diverging clades) is essentially an instantaneous process, whereas microevolution necessarily requires extended periods of time.

    We still, of course, have a problem with instantaneous extinction (such as that caused by an asteroid collision). Such an event causes massive, essentially instantaneous changes in allele frequencies over large populations, by instantly eliminating whole aggregations of alleles (by obliterating their “carriers”). However, I would argue that this can be integrated into the definitions listed by simply considering extinction (especially mass extinction by catalysm, or “neo-catastrophism”) as an example of macroevolution.

  14. Dan says,

    So, contrary to Larry’s blowhard assertion, there’s nothing unconventional about my position at all.

    I never said that your position was unconventional. Quite the contrary, I’ve said all along that there is scientific controversy about the sufficiency of microevolution. In a scientific controversy there are two valid scientific positions and the correct answer hasn’t been decided. Both positions are “conventional” in this case.

    Those who ignore the controversy by claiming that there’s only one scientific explanation of macroevolution are not doing good science.

  15. Please excuse the incorrect formatting in the previous response. The quotation from Dan is just the first paragraph.

    I’m still trying to figure out what codes are allowed in the comments and the trial-and-error process is a bit frustrating.

    Why doesn’t WordPess allow you to review your comments and edit them before posting? Or am I just not seeing the option that allows you to do that?

  16. Larry,
    No, wordpress doesn’t allow preview of comments. I can (and will in a moment) go back and edit comments to correct for formatting errors. But I agree, it would be nice if those features were available to you.

    Also, to my position – I was focusing on your comment that perhaps I should have my ideas published. I realized that you were mocking me, and I mocked back.

  17. It’s interesting that even Darwin tended to think about these two subjects separately. He used the term “descent with modification” to refer to the process(es) we now refer to as macroevolution”, and referred to the mechanisms by which this happened “natural selection” and “selection in relation to sex”. Ironically, although Darwin is mostly known among non-scientists as the “discoverer” of the first (which we now refer to as “evolution”, a term he avoided because of its teleological connotations), he was himself mostly concerned with the mechanisms by which it happened. This was because many other people had previously asserted that “descent with modification” had occurred, including Buffon, Lamarck, Chambers, and even his own grandfather, Erasmus Darwin. Charles Darwin was mostly obsessed with the processes that drove evolution, rather than the fact that it had occurred.

    The other major mechanism driving “descent with modification” — genetic drift — wasn’t discovered until the early 20th century, when it was originally referred to as the “Sewall Wright effect” in honor of its discoverer. We now recognize at least three types of genetic drift: the founder effect (Ernst Mayr gets most of the credit for this, in his theory of peripatric speciation), the genetic bottleneck effect (essentially a variant of the founder effect), and “classical” genetic drift (essentially random sampling error in very small, inbred populations, usually of animals).

    The distinction between macroevolution and microevolution has a surprisingly long pedigree. Most people assume that it must have been made about the time of Eldredge and Gould’s famous paper on punctuated equilibrium (i.e. the early 1970s). However, there is good historical evidence that some evolutionary biologists were using precisely these two terms as early as the turn of the 20th century, as pointed out in John Wilkins’ essay on the subject at Talk/Origins. Many people assume the distinction was derived from the distinction between micro- and macroeconomics. However, although it bears some superficial resemblance to those two terms, at base they are fundamentally different (i.e. non-analogical) concepts.

    Personally, I think it’s more useful to distinguish between anagenesis and cladogenesis, rather than microevolution and macroevolution. However, this involves a lot of semantic groundwork, and is therefore only really useful when one is formally teaching evolutionary biology.

  18. All of this is very interesting and informative; thanks to all participants.
    Just a couple of probably superfluous observations:
    1. It never makes any sense to speak of selection (a classically micro-evolutionary process) outside of the context of environment. That’s why I view events like mass extinction, severe climate change, post-catastrophe ecological release, faunal interchange, etc. as just exaggerated instances of section–they are changes in the environmental selection pressures impinging on populations, not in the populations themselves. Some of these environmental changes–asteroid slams into habitat, home pond freezes solid for 2000 years–are large and/or fast enough that existing genetic variation is insufficient to drive adaptation, and extinction results. Others–new competitors and predators migrate over new land-bridge, all competitors and predators wiped out during period of protected estivation–may be extremely rapid environmental changes to which a population can nevertheless adapt. Or not. But either way, for a population in question it’s just microevolution exaggerated.
    2. Macroevolutionary patterns can, essentially by definition, only be studied in hindsight. It takes a long, long period of fossil accumulation to be able to point and say “punk eek!,” or even (in most cases) “cladogenesis!” If, however, we had a time machine and could go back to study these populations in real time, there would be nothing to see but classical microevolution, plus populatiuon fission by vicariance or dispersal.
    3. The only real contender, it seems to me, for a macroevolutionary process that does not and cannot reduce to microevolution is this business of “species sorting.” But that is only because it was invented for that exact purpose–by requiring emergent properties for species, it is automatically meaningless at the level below, populations. So yes, IF species sorting exists, it would have to be regarded as a non-micro macro proccess. Because that’s precisely why Gould thought it up! The real question is, then, what are these emergent properties that species have but not populations? And I submit that nobody has thought of one.

  19. Sven DiMilo says,

    The real question is, then, what are these emergent properties that species have but not populations? And I submit that nobody has thought of one.

    Really? What’s the logic behind this statement? Why do you reject all the examples that Vrba, Stanley, Eldgredge, Gould, and others have proposed in dozens of papers published over the past thirty years?

    Gould discusses them in his book The Structure of Evolutionary Theory. Why not pick one of the examples he describes and tell us why you reject it out-of-hand?

  20. Well, it’s certainly true that I do not know all of the examples that have been proposed. That’s why I used the verb “submit” instead of “assert.” I’m willing to be shown the error of my ways, seriously.

    Gould discusses them in his book The Structure of Evolutionary Theory. Why not pick one of the examples he describes…
    I won’t participate further in these discussions until I have had the chance to do that. Fair enough?

  21. Sven,
    I agree with what you’re saying – the wording with micro and macro, in the context of common descent, suggests a certain degree of continuity. This is the difficulty that I was having, and I don’t think that Larry is the best explainer of this concept. As noted above, I had to go talk to Allen before Larry’s view made any sense.

    What Larry fails to do is explain the semantic differences here. While we intuitively translate macroevolution and microevolution as “big change” and “little change” respectively, that’s not how they were defined by the modern synthesis.

    Here, Larry’s describing microevolution in the classical sense, where allelic frequencies shift slightly over time but it’s still the same species. Macroevolution, in contrast, is where a significant new trait gets fixated into the population, such that the old species name and description no longer fits. (the difference between anagenesis and cladogenesis) This terminology creates a discontinuity in population dynamics – instead of a gene pool constraining the taxonomic phenotype, you have one or two subtle changes that establish a new gene pool altogether.

    These classic definitions make no sense, however. For people who are familiar with studying evolution from the standpoint of heredity and divergence, there’s only some divergence and a lot of divergence, respectively, and no easy line in the sand between micro and macro. But that’s the rub – we’re used to the gene-centric view of biology, not the species-centric view. Contrary to what Larry is saying, his view doesn’t introduce any new mechanisms of change at the gene level. The population dynamics do change however.

    At least that’s how I think it works – Hope that helps.

  22. Another way to look at the difference is to consider what the outcome of the two processes are. As I posted earlier, microevolution is defined as those processes that cause changes in allele frequencies in populations over time (duration unspecified): natural selection, sexual selection, and genetic drift. In classical Darwinian terms, microevolution (i.e. selection) was believed to be the source of evolutionary adaptations. According to the “modern synthesis”, microevolution also includes the effects of random drift (i.e. non-adaptive microevolution), plus several other processes of limited applicability (e.g. meiotic drive, etc.).

    By contrast, macroevolution is defined as those processes that cause divergence between two or more phylogenetic lines (beginning with as a few as a single divergent individual): allopatric and (perhaps) sympatric cladogenesis. In classical Darwinian terms, macroevolution was believed to be the result of non-adaptive divergence, usually resulting from increased sterility between diverging populations (underlying genetic cause unspecified). According to the “modern synthesis”, macroevolution is the result of vicariance, combined with gradual random genetic change, eventually rendering interbreeding between isolated populations difficult or impossible.

    Dobzhansky (along with other founders of the “modern synthesis”) proposed “diversifying selection” as a contributory cause for the divergence of phylogenetic lines. However, it is now increasingly accepted that selection plays little or no direct role in cladogenesis. On the contrary, selection for both prezygotic and postzygotic isolating mechanisms presupposes that divergence has already resulted in partial sterility between diverging populations.

    Again, the two terms —microevolution and macroevolution — generally do not operate via the same mechanisms (selection & drift vs vicariance) nor do they produce the same effects (adaptations versus the origin of new clades).

    It is supremely ironic that Darwin’s most famous book, and the foundation for all of modern biology, was badly mistitled: it isn’t really about the origin of new species (and certainly not via natural selection. On the contrary, the first half of it is mostly about the origin of adaptations by means of selection, while the second half is mostly about descent with modification. Once again, we have the distinction between anagenesis (i.e. change driven by selection and drift) versus cladogenesis (i.e. divergence driven by vicariance), going all the way back to the Origin.

  23. In this context it also helps to know that Darwin was writing a book he had been planning to call Natural Selection, which was mostly going to be about the origin of adaptations (i.e. microevolution), when he was goaded by his desire for priority to publish a “brief abstract” of what would have certainly been a multi-volume work about it (similar to Lyell’s Principles of Geology) upon receiving a letter from Alfred Russell Wallace about the very same subject. It is not entirely clear why he shifted the focus of the title of the book to the “origin of species”, although one suspects that catching the eye of the scientific community (not to mention the literate public) was a primary motivation. Most people read it expecting to have the origin of species explained, and find that it mostly explains the origin of adaptations (and the divergence of already existing species).

  24. Dan says,

    I agree with what you’re saying – the wording with micro and macro, in the context of common descent, suggests a certain degree of continuity. This is the difficulty that I was having, and I don’t think that Larry is the best explainer of this concept. As noted above, I had to go talk to Allen before Larry’s view made any sense.

    Whenever you encounter a new idea that conflicts with your previous notions, it’s always a good idea to seek out another source. Usually it takes two or three different sources to convince you that something important is going on.

    What Larry fails to do is explain the semantic differences here. While we intuitively translate macroevolution and microevolution as “big change” and “little change” respectively, that’s not how they were defined by the modern synthesis.

    Actually, as I explain in my essay, ….

    When the principle tenants of the Modern Synthesis were being worked out in the 1940’s, one of the fundamental conclusions was that macroevolution could be explained by changes in the frequency of alleles within a population due, mostly, to natural selection. This gave rise to the commonly accepted notion that macroevolution is just a lot of microevolution. Let’s refer to this as the sufficiency of microevolution argument.

    The founders of the Modern Synthesis are responsible for promoting the idea that macroevolution is just lots of microevolution. They pretty much ignored the different definitions of macroevolution in order to reinforce their opinion that genetic changes in populations were the only significant player in evolution. Later on some of them changed their minds.

    I’m sorry that I failed to explain the semantic differences. It was the main point of my essay. I guess I’ll have to re-write it. What part of the following paragraph confused you?

    Levington then goes on to draw a parallel between microevolution and macroevolution on the one hand, and physics and astronomy on the other. He points out that the structure and history of the known universe has to be consistent with modern physics, but that’s not sufficient. He gives the big bang as an example of a cosmological hypothesis that doesn’t derive directly from fundamental physics. I think this analogy is insightful. Astronomers study the life and death of stars and the interactions of galaxies. Some of them are interested in the formation of planetary systems, especially the unique origin of our own solar system. Explanations of these “macro” phenomena depend on the correctness of the underlying “micro” physics phenomena (e.g., gravity, relativity) but there’s more to the field of astronomy than that.

  25. Larry,
    It’s not that you didn’t describe the issues. You just didn’t focus on something that my preconceived notions couldn’t reconcile. Which is okay – you didn’t write your essay for me specifically. Case in point, the paragraph that you quote from your essay: it is common knowledge that the Universe is composed of particles called atoms, which obey fundamental and reducible laws of physics. You didn’t illustrate any discontinuity there.

    Reference to cladogenesis alone didn’t do it for me either. The strength of the cladogenesis argument is when it is contrasted with anagenesis. By contrasting the two more concisely, it becomes easier to pinpoint the difference.

    Similarly, the description of microevolution as something that takes many generations to be visible, whereas macroevolution needs only one in principle, was an eye-opening juxtaposition. All it may take is one pair of organisms to found a new population, and this is a distinct process from that of allelic frequencies that merely fluctuate. Or, if you prefer – microevolution is the description for relative stasis, whereas macroevolution is the dramatic punctuated shift and the consequences of such when repeated on geological timescales.

    That, plus Allen linked to a helpful depiction of cladogenesis vs. anagenesis that helped me a lot (I’m a visual thinker).

    And one last thought on what could be better about your essay and response to comments (specifically on the sufficiency of microevolution). I was asking “what other mechanisms are needed then” and you were trying to respond with cladogenesis as a mechanism. I don’t think that it’s a mechanism at all. Cladogenesis and anagenesis are distinct patterns that are consequences of the basic genetic mechanisms, not new mechanisms of heredity themselves. Case in point, you started off your blog post with this:

    There is legitimate scientific debate about whether macroevolution is more than just lots of microevolution or whether macroevolution encompasses mechanisms not seen in microevolution. It’s the sufficiency of microevolution argument.

    And no, there are no new mechanisms in macroevolution – they’re different patterns of the same mechanisms of heredity – selection, drift, mutation, etc. So calling it the “sufficiency of microevolution” argument goes against what you’re trying to say. Natural selection as a mechanism IS sufficient to explain common descent and everything that followed from that, but it generates two different patterns of change.

  26. Dan wrote:

    “I don’t think that it’s a mechanism at all. Cladogenesis and anagenesis are distinct patterns that are consequences of the basic genetic mechanisms, not new mechanisms of heredity themselves.”

    but then went on to write:

    “Natural selection as a mechanism IS sufficient to explain common descent and everything that followed from that, but it generates two different patterns of change.”

    Here, precisely, is the problem. Neither natural selection nor the various other processes included under the rubric of microevolution, nor the various processes included under the rubric of macroevolution are mechanisms. On the contrary, they are outcomes produced by the interactions between processes.

    According to this view:

    Natural selection is the outcome of the interactions between:
    • variation,
    • heredity, and
    • fecundity,
    which together produce
    • differential reproductive success

    Sexual selection is the outcome of the interactions between:
    • variation,
    • heredity, and
    • asymetrical parental investment
    which together produce
    • non-random mate selection
    which in turn produces
    • differential reproductive success

    Genetic drift is the outcome of the interactions between:
    • variation,
    • heredity, and
    • effective breeding population size
    which together produce
    • inbreeding depression and
    • random changes in allele frequencies in populations


    Macroevolution is the outcome of the interactions between:
    • the intensification and/or relaxation of selection
    • changes in adaptive zones, and
    • previous historically contingent macroevolutionary events
    which together produce
    • historically contingent changes in inter-taxon composition and diversity

  27. Got me there Allen. Thanks for the correction.

  28. […] from other WordPress sites on which I had commented.  He had stepped in to add some clarity on a post written by Dan Rhoads at Migrations. science evolutionPopularity: 1% [?] SHARETHIS.addEntry( { title: “Evolutionary Pluralists”, url: […]


%d bloggers like this: