RadioLab distorts some science

September 23, 2018 • 12:00 pm

There’s been a lot of publicity about David Quammen’s new book, The Tangled Tree: A Radical New History of Life, which tells the story of the discovery of a new domain of life, the Archaea, the discovery that chloroplasts and mitochondria are the remnants of anciently absorbed microbes, and, most novel, the recent discovery of horizontal gene transfer (HGT): the transfer of DNA and genes between relatively unrelated organisms. Most of the publicity about the book—to be sure, publicity pushed by Quammen himself—centers on HGT. It is, we’re told, something that radically overturns Darwin’s view of the “tree of life” and of evolution, and even revises our own view of “what it means to be human” (after all, we’re also told, a substantial part of our genome is dead, nonactive DNA from ancient retroviral infections).

A month ago today I reviewed the book in the Washington Post (see my postmortem post here), and was pretty critical of the way Quammen handled HGT. While the phenomenon is fairly common in bacteria, it’s less common in eukaryotes, and hasn’t dramatically altered our view of evolution in either case, much less having changed our view of ourselves. Even horizontally transferred genes must be subject to natural selection if they’re to spread as adaptations, and they certainly haven’t messed up evolutionary trees in eukaryotes, for people are still busy reconstructing the history of life using such trees. (Indeed, the delineation of Archaea as a separate domain of life depends on their being a discernible evolutionary tree.)

Nevertheless, Quammen still promulgates his spiel widely, most recently in a National Geographic interview and, more invidiously, in the RadioLab interview below (click on screenshot). After all, the man has books to sell. But I wish he’d tone down his rhetoric.

Both Quammen and the show’s hosts (Robert Krulwich and Jad Abumrad) seem to promulgate the misconceptions about HGT, and I hope RadioLab isn’t becoming more uncritical. I can’t enact the emotional labor to dissect every error in this 30-minute show, but here are a few bizarre things to which I took exception:

1.) Krulwich says that Quammen’s book shows that HGT and perhaps endosymbiosis (origin of mitochondria and chloroplasts) are “a smack in the face to Darwin’s theory of evolution.” Well, not really. They are revisions of how genetic variation comes to enter a species’ genome, but, as I said in my review, HGT doesn’t really efface Darwin’s “tree of life.” And it hasn’t affected our view of adaptation via the disposition of random genetic variation by natural selection.

2.) In the classic story of the peppered moth Biston betularia, the show tends to cast the spread of the “melanistic” mutation in the moth during industrialization as a refutation of “incremental mutation”. But it’s wrong to say that the old idea was that the color difference was due to several mutations, or happened gradually,. We have long known that the change in color from peppered white to dark was the result of mutation in a single gene, and that the English population became transformed quickly: within a few decades. Further, when pollution controls were passed, the moth quickly reverted to its ancestral peppered form.

The change was not, as Krulwich says, instantaneous. And just because the gene change was due to a “transposable element”” that moved within the moth genome (i.e., not HGT) rather than a change in one nucleotide does not affect the Darwinian nature of the change one bit. It just shows that the “mutation” was a big one involving translocation of a big chunk of DNA into the moth’s cortex gene. No revision of Darwinism needed, especially because Darwin didn’t even know about genetics, so you could equally well shout that genetics itself was a “smack in the face to Darwin’ s theory of evolution”. The whole characterization of the “original” Biston story as “incremental mutation and very slow evolutionary change” is a straw man”—not even wrong.

I wrote to my old undergraduate advisor, Bruce Grant, who worked extensively on Biston. He listened to the RadioLab show and had this to say:

The “review” of the classic peppered moth story is garbled and grossly misrepresents it. No one has ever argued for incremental mutations. There are multiple alleles that produce a range of intermediates collectively called insularia, with the darkest, fully melanic phenotype called carbonaria. It acts as a qualitative autosomal dominant, fully melanic from the get-go wherever it occurs (besides the UK, in North America and Japan). Its frequency increased via directional selection, and its frequency has also decreased by directional selection. Further, these changes have been parallel in both directions on separate continents (US and UK), in concert with documented changes in air quality assessed by suspended particulates ( soot).

The DECLINE in melanism has been much better documented than its increase. There are mountains of data. Those taken by Cyril Clarke include nearly 20,000 specimens collected at one location near Liverpool from 1959 to 2001+. Insularia (intermediates) have always been rare there. Effectively only two phenotypes have been involved: fully melanic carbonaria and the “typical” peppered form. The reduction in the frequency has been gradual, did NOT involve intermediates and fell from well over 90 % to under 5 % in 40 generations. Now carbonaria has all but disappeared there. I witnessed this decline personally from 60 % onwards. It’s real and intermediates (insularia) were always rare there.
Parallel changes have occurred independently in Michigan. All this has been published and is supported by hard data. The RadioLab people do NOT know what they’re talking about.

What Quammen is probably confused by is Ilik Saccheri’s evidence that transposons may be responsible for the original mutation. But there are still some reservations about that I spelled out during my talk at CoyneFest. I do think that Ilik oversells his work, and shortchanges others in the process, but surely he’d still object to Quammen and the hosts’ interpretation of Industrial Melanism.

3.) The suggested idea that the first billion years of life involved pervasive HGT, and that individuals and species weren’t easily delineated, is just that—speculation. Species and well delineated individuals could have arisen quite early, for all we know.

4.) Krulwich’s claim that HGT shows that evolution “can happen a lot faster than we thought” isn’t really true. A horizontally transferred gene is just a big mutation, and in many cases will sweep through the population just as rapidly or as slowly as a mutational change in the organism’s own DNA. What’s important is how strong natural selection is. And we already knew that evolution by conventional mutational changes in genes can happen quite quickly, even without HGT. (Biston is an example of that, as is the 10% change in beak size in one year in the medium ground finch revealed by the work of Peter and Rosemary Grant and their group.)

5.) Once again we hear the claim that the human microbiome and the presence of viral DNA in our genome means that we have to revise the notion of who we are, and “what it means to be human”. But that’s bogus, too—just a sensationalistic way to sell books. The fact is that I don’t wake up every morning and think, “Geez, my gut is full of E. coli and my DNA full of retroviruses. I have to rethink what it means to be Jerry Coyne!”

6.) There is a clear implication that Carl Woese contributed to work on HGT. This is not the case: his work highlighted in Quammen’s book was in discovering the domain of life comprising the species within the new group Archaea.

7.) Finally, it is overblown to think that HGT and endosymbiosis means that the categories of ‘individual organism’ and ‘species’ become blurry. Is Homo sapiens a real group or an arbitrary construct? The former, for sure. And I am Jerry Coyne, not a mixture of Jerry Coyne and Donald Trump (thank god!).

So listener beware when you hear this discussion on NPR. I wish Krulwich & Co. would have been a bit more critical rather than credulous consumers of Quammen’s sensationalism.

22 thoughts on “RadioLab distorts some science

  1. Thank you for this, Jerry. Announcing the death of Darwin’s big idea has been a cottage industry for a long time, and I fear that Quammen has joined it. And in truth, the evidence for HGT is not all that new – Ford Doolittle and others were writing about it (based on data) 20 or so years ago, and it is a staple of any modern evolution textbook. While it definitely complicates bacterial phylogenetics, it does not, as you so clearly argue, conflict with evolution by natural selection. And finally, with tens of thousands of human genomes having been sequenced, what is the evidence for recent HGT of functional genes? My guess is that there isn’t any.

  2. Quammen may have some antipathy to Darwin, or perhaps, he just realizes that dissing Darwin sells books. In his much acclaimed The Song of the Dodo, Quammen pretty much buys wholesale the old myth that Darwin stole the idea of natural selection from Wallace (although I don’t know if that claim was used as a selling point at the time the book came out).

    On another matter, I’ve long thought that Woese’s “3 domains” of life is a wildly oversold concept. It was long obvious that eukaryotes descended from prokaryotes, and that therefore there was a fair chance that some extant prokaryotes are closer cladistically to eukaryotes than to the rest of the prokaryotes (i.e. that living prokaryotes are paraphyletic). The discovery of which particular bacteria are in fact the ones that are on the stem lineage of the eukaryotes is a great discovery, but it’s not earth-shaking.

    There used to be introductory biology textbooks that were structured around the notion of “How many kingdoms of life are there?” When students got to my classes, one of the goals of the section on systematics was to get them to realize that the question about kingdoms was not only unimportant, but profoundly unimportant. The question of the number of domains is the same. (The “number of kingdoms” bit seems to have gone out of fashion, so I no longer need use this example.)

    1. That’s not how it happened, though.

      Archaea and bacteria differ by quite a lot in basic biochemistry.

      Eukaryota didn’t descend from either – it descended from a fusion of both, which introduced both types of biochemistry.

      It’s just not the case that some branch of Archaea were on their way to becoming eukaryotes and happened to absorb some bacteria that become mitochondria. That event was what triggered the change in direction. What made it possible, in fact.

      It’s absolutely a mistake to consider Archaea as just a kind of bacteria.

  3. According to Wikipedia, Krulwich has no academic background in science. Yet, over the years, he has managed to get several gigs on programs dealing with science topics. I wonder how he did that. He seems to exemplify the belief that anybody can talk intelligently on any subject without a background in it and regardless of how complex it is. Lacking this background in genetics and evolution, it is not surprising that Krulwich can unskeptically accept Quammen’s assertions.

      1. Are there any major media outlets that do not pander to religion? In the US the percentage of their market who are religious is large and the antipathy toward atheism substantial. I suspect it’s mostly a matter of marketing.

        1. I don’t think it is just marketing. The fact that we have such a large number of religious folk here means that a great many of the people at places like NPR are themselves believers. And believers all possess a level of credulity that leads to this sort of thing.

  4. Darwin and evolution still sells. Besides those that would like to see it fail or stumble, many more seek to draw attention to their work via titles that exaggerate the controversy or attempt to create one. It is simply a matter of aggressive marketing.

  5. Twitter is ablaze with Quammen. He comments on any post having anything to do with his book and with evolution. I just don’t get the novelty of HGT. Perry Marshall, in his book Evolution 2.0 also acts as it’s something new. I emailed him to say that my old biology 101 text, 40 years old now, has plenty to say about HGT. I’ve argued with these guys that’s it’s just genetics. We’re getting better at it and we incorporate the finding into evolution. There’s a better understanding but certainly nothing has been overturned.

  6. The suggested idea that the first billion years of life involved pervasive HGT, and that individuals and species weren’t easily delineated, is just that—speculation. Species and well delineated individuals could have arisen quite early, for all we know. … Is Homo sapiens a real group or an arbitrary construct?

    As a backdrop, some of the best people have recently illuminated that question I think (I still need to read their paper), and while phylogenies customarily push the root node as far back as it can go they use several methods to arrive at a posterior credible dating interval [ , Fig, 3].

    The root node that delineates a lineage split is < 40 Myrs after the Moon forming impact, so at most 1/20 the suggested 'pervasion'.

    Eukaryotes split recognizably within the other Archaea, while the primary endosymbiosis likely happens some 500 – 1,500 Myrs after that.

    This can be wrong of course. But at least the current ideas are a little bit less of speculation than Quammen et al imply.

  7. I’ve had a book on my shelf for a while (The Third Domain by Tim Friend) and I might now have to read it and see how he treats the historical work of Woese and HGT. Perhaps it is more objective.

    1. This is a good book – I do recommend it. It’s been a while since I read it, but I do recall a lot about Woese, as well as about the unique biology of the modern archaea. Not so much about HGT.

  8. I read the book more for fun the scientific edification. The struggles and humaneness of researchers well known but a,ways as interesting as a novel. I was impressed as much by Woese’s sturdy effort to keep going as by his lat phase.
    Perhaps the HGT seemed more a hypothesis for the future than a real thing led me to read the book wrong. Fascination in. Hypotheses if always more rewarding than buying them.

    The last point. Are you sure?

  9. I disagree with the title “RadioLab distorts some science”. Specifically the word “some”. The hosts always seem eager to latch on to the most sensationalist and debatable interpretations of the science they report.

  10. I greatly appreciate the knowledgeable and credible critiques of books and media coverage of evolution and Darwin. Without these it would be hard for someone like me (i.e. someone with basic understanding of evolution and who has read many books about it, including those of Ernst Mayr and others of the 20th century) to sort out speculation from fact. Most general readers will not be able to do that but it is crucial that it be identified and sorted out from fact. Thanks to Jerry and to the commenters for doing this important work! (I do distribute a lot of articles and comments on this blog to my personal listserv, in order to keep them informed).

  11. HGT and endosymbiosis is “a smack in the face to Darwin’s theory of evolution.”

    When Jerry replied: “Well, not really.” I thought that was quite a weak rejoinder.

    But later he wrote: “No revision of Darwinism needed, especially because Darwin didn’t even know about genetics, so you could equally well shout that genetics itself was a “smack in the face to Darwin’s theory of evolution”.


    It is disappointing that certain authors and even scientists muddy the waters of scientific progress and understanding just to sell books or attract funding. The world does not need more of this.

  12. Hype and exaggeration are unfortunately all too common in science. In the public forum, it’s not only books but also press releases and talk shows. In the literature itself, it’s overuse of the words “novel” and “paradigm” (as examples). There’s pressure to do this, not only to secure funding, but also for status and promotion. This seems to be especially true in cancer research. In evolutionary biology, it provides fodder to the creationists. Forums such as this one provide some correction.

  13. I’m not surprised.

    While I initially enjoyed RadioLab — even though it approached my limit for cutesy pop science — their episode a year or two back supporting the hunting of endangered species in Africa was over the top. It was truly a deplorable and ignorant smear job on conservation, chock full of straw men.

    That sort of facile anti-environmental contrarianism is annoying, and dangerous. No thanks.

Leave a Reply