Coyne is on the Loom

We had Neil Shubin here last week, and now Jerry Coyne is guest-blogging at The Loom. I look forward to the day that I can just sit back and invite prominent scientists to do my work for me here.

Although, I have to say that while Coyne is largely correct, he’s being a bit unfair. He’s addressing Olivia Judson’s recent article on “hopeful monsters”, a concept Coyne and the majority of the biological community reject. I reject it, too, but I think there are some legitimate issues that are associated with the idea that are also all too often and unfortunately discarded.

One point that Coyne handles well: there is a disconnect between the magnitude of genotypic changes and phenotypic effects — a single point mutation can cause amazing morphological changes. As Coyne points out, though, although this can happen, it’s not likely to be a major force in evolutionary change. Dramatic, single-step phenotypic effects are the kinds of things that geneticists select for, but they are also exactly the kinds of things that nature selects against. Evolution is much more likely to sidle up towards a major change by successive smaller steps, since those small changes are less likely to be accompanied by major deleterious side effects. Also, phenotypic outcomes of development should be robust to be advantageous, which typically means that there are many regulatory events cooperating to produce them — and they are therefore buffered by multiple controls.

But please, let’s not always dismiss Richard Goldschmidt when discussing “hopeful monsters”. It really wasn’t that awful an idea. Goldschmidt worked on stable variations in organisms: he studied sex differences (ever noticed that males and females have pretty much the same genes, but different phenotypes?) and metamorphosis (similarly, an organism builds two or more very different morphologies with exactly the same genome). He postulated that there could be specific, well-structured, stable nodes of patterns of gene expression — genes weren’t generally fluid, but tended to lock in to particular states. If he were writing today, he’d probably be bringing up the notion of attractors in chaos theory; the ideas are very similar. In that context, he was proposing a worthy concept that should have been taken more seriously than it was — Mayr’s hatchet job was particularly awful.

The “hopeful monster” concept was not shot down by the synthesis — it was ignored. I think it’s been dismantled by developmental biology, though; what we’ve learned is that the stable morphological types we see in a single species are not simply fortunate stable nodes in a nucleus that can be tuned in different ways, but that each are the product of many generations of slow sculpting by the processes of evolution, and that they are riddled with clumsy kluges that aren’t the outcome of some elegant global pattern switching mechanism, but of a long history of small tweaks.

Now also, Coyne is no fan of evo-devo, and he briefly voices the suggestion that the evolutionary developmental biologists are among the sources of this idea that saltational changes lead to sudden, drastic changes in body plans … but I’m just not seeing that. I am seeing work, for instance, that suggests that Hox duplications have been part of the process of producing additions to body plans, but it’s not a case of “poof, arthropods gain a metathorax in one change” — it’s been quite conventional. It’s more like “poof, arthropods gain an extra Hox gene, which initially adds redundancy and is later shaped by evolutionary processes that confer additional specializations on a segment,” quite ordinary stuff that shouldn’t be at all objectionable to Coyne.

It’s especially peculiar to pin the “hopeful monster” concept on evo-devo, when the one evo-devo expert he quotes, the biologist Sean Carroll, explicitly points out that evo-devo doesn’t support it.

Coyne is also going to be speaking at an evo-devo symposium I’ll be attending in April — I’m going to be very interested to hear what he has to say.


  1. says

    Aren’t you boycotting Coyne for being religious? What kind of an atheistic scientist are you, anyway?

    Watch the Expelled trailer again (or that stupid cartoon), we’re supposed to be expelling anybody like Coyne, damnit. Stein says so, and his concern for freedom is as pure and unsullied as is his lack of the knowledge of science and how it operates.

    Glen D

  2. Peter Ashby says

    I suppose it depends on what you define as a saltation. Is it just the sudden appearance of functional novel structures or do what are effectively duplications count?

    In a note to Nature back in about ’88 Per Ahlberg pointed out that the back end of Coelocanths was mirror imaged around the D-V axis, is that a saltation? How about a partial rotation?

    Then there is the internal position of the scapula in testudinoids, hard to envisage a functional intermediate of that one.

    So I think the problem is that each side of the debate has different ideas of how big a change must be to be a saltation.

  3. Peter Ashby says

    And trophoblast giant cells in mammalian placentas are syncitia which look very much like some virus infected cells. If that is how they came about with the virus becoming an ERV then you can’t be partially infected.

    I’m not saying they are either frequent or common, or trying to make arguments from incredulity. Some molecular evidence for all of the above would be very nice for eg. But to rule them out completely is to close your mind I think.

    We see that in the caricature of ‘hopeful monsters’ which conjures up images of anencephalic human babies and such.

  4. BillW says

    In evolutionary computing, we sometimes see behavior where a new mutation reduces an individual’s fitness in the short term (though usually not drastically) before leading to an increased fitness a few generations later. It’s almost as if to make a useful phenotypic change, the initial genotypic change reduces fitness until it can be “smoothed out.”

    I believe that Richard Lenski at Michigan State mentioned seeing that in some of his model bacteria when I heard him speak a number of years ago.

    Have you seen this sort of mechanism in your work?

  5. Bjørn Østman says

    Bill W., I am doing some work on simulations where I see what you are talking about. I’d like to hear more about what you have seen. Email me:

  6. Bjørn Østman says

    Bill W., I am doing some work on simulations where I see what you are talking about. I’d like to hear more about what you have seen. Email me:

  7. Pete Dunkelberg says

    I agree with PZ’s remarks here, except to note that plants may have viable mutations of large effect more frequently than animals do.

  8. Colugo says

    Can anyone meaningfully define or operationalize “mutation of large effect” or “macromutation” or “hopeful monster”? (As they mean today, not precisely what Goldschmidt – limited by his era’s knowledge of genetics – meant.)

    Do these include relatively trivial changes like short legs and hairless vulture necks, single generation major allometric or heterochonic change, meristic change, topological transformations, and entirely novel features? (If they do, that is a very big box.) How big does a change have to be?

    Are these based solely on phenotypic criteria, without regard to whether these involve major chromosomal alterations, single point mutations, or even solely epigenomic change? What about reliably generated phenocopies after a stable change in environmental conditions?

  9. says

    “How big does a change have to be?”

    I has to be big enough to account for major evolutionary transformations, gaps in the fossil record, the origins of phyla… That’s what macro-mutations are supposed to do ;o)