A new paper claims that evolution has stopped in a bacterial species. Is it true?

Several readers called my attention to a new paper by J. William Schopf and colleagues in the Proceedings of the National Academy of Sciences (reference below; free download at link), a paper that has also gotten a great deal of attention in the press. Last week a journalist asked me to comment on it, but I was too busy then to read it. Now that I have, I’m not all that impressed. It’s a decent paper, and doesn’t fail the first test of a science paper—does it tell us something new?—but I don’t think it makes the case that’s gotten the press all excited.

What is that case? The authors claim that their finding—an example of extremely slow evolution (in fact, no perceptible evolution) in a sulfur-producing bacterium—constitutes a test of Darwin’s “null hypothesis of evolution.” That hypothesis, as stated by the authors, is this:

. . . if there is no change in the physical-biological environment of a well-adapted ecosystem, its biotic components should similarly remain unchanged.

In other words, if there’s no selection pressure on the organism, it will not evolve. The authors implicitly equate “change in the physical-biological environment of a well-adapted organism” with “selection pressure on that organism,” and hence with “evolution of that organism,” but that’s not correct. I’ll talk more about this below, but for the time being try to guess how organisms could still continue to evolve in a relatively unchanging environment. Let’s look first at the author’s data, which, they claim, shows no evolution taking place in more than two billion years.

What Schopf et al. did, which is good stuff, is to examine bacterial microfossils in two ancient Precambrian biota and then compare them with modern fossils living in a similar environment. They were concerned with sulfur-using bacteria living on the ocean floor. The oldest formation examined was the “Turee Creek Group Kazput Formation” in northern Australia, a formation that’s ancient—about 2.3 billion years old. That’s old, but it doesn’t take the prize for age, for the oldest known bacteria (cyanobacteria, formerly known as “blue-green algae”) come from about 3.5 billion years ago.

The second group of fossil sulfur-metabolizing bacteria are about 500 million years younger than those from Turee Creek: they’re from the 1.8-billion-year-old Duck Creek formation, also in Western Australia. So we have about 500 million years of potential evolutionary change intervening between the two fossil formations.

Finally, the authors compared fossil bacteria from these two formations with sulfur bacteria that are still with us: a community of sulfur-using bacteria discovered in 2007 in the seafloor off the west coast of South America. All of these bacteria are presumed or known to use marine sulfur compounds, first reducing them to hydrogen sulfide and then oxidizing them to produce elemental sulfur and sulfur dioxide.

So what are the similarities among these three types of bacteria that led the authors to suggest that they hadn’t evolved over 2.3 billion years? There are three. First, all three bacteria lived in communities not in shallow water, but in the sea floor in deeper waters. Paleontologists have ways of telling this, and you can read the paper if you want to know how.

Second, they all have similar morphology, forming long filaments of about the same size. Here is what they look like; the figures show both the ancient bacteria from both formations and the modern collection (“A”). They form long, cylindrical filaments of comparable size

Screen Shot 2015-02-04 at 7.57.27 AM

Third, chemical analysis of the matrix around the fossil bacteria, and of the modern ones, showed that they all lived or live in anoxic (“oxygen free”) communities, produce the same sulfur isotopes, and yield a form of pyrite as a metabolic product. The authors conclude, probably correctly, that the metabolic pathways for sulfur use in all of these forms are similar

So have they remained evolutionarily static, as the authors argue? First, let’s review their claim about these bacteria:

Once subseafloor sulfur-cycling microbial communities had become established, however, there appears to have been little or no stimulus for them to adapt to changing conditions. In their morphology and community structure, such colorless sulfur bacteria—inhabitants of relatively cold physically quiescent anoxic sediments devoid of light-derived diel [daily] signals and a setting that has persisted since early in Earth history—have exhibited an exceedingly long-term lack of discernable change consistent with their asexual reproduction.

They also claim that not only did the morphology and biochemistry of these species remain unchanged, but they didn’t form new species, either, although the concept of bacterial “species,” as Allen Orr and I showed in our book Speciation, is a bit hazy.

But I think the author’s conclusion is premature, and for two reasons.

First, regarding the “null hypothesis of Darwinism,” I think that that notion is wrong—or at least incomplete. Even in an unchanging environment, organisms can still evolve in significant ways.  If new mutations arise that adapt the species better to that unchanging environment, then we will have evolution. For example, the bacteria could evolve more efficient metabolism of sulfur. Alternatively, one could have mutations that simply allow the bacteria to divide faster, giving them a selective advantage over others. Neither of these would be a response to a changing environment, but could still cause evolution. Considerable retooling of the bacteria’s metabolism, DNA synthesis, and so on, could still occur as evolution experiments with various mutations.

There is in fact one natural experiment that showed evolutionary divergence in an unchanging environment. There are salmon in some West Coast rivers (pink salmon, as I recall), that form “year classes”: they have a two-year life cycle and come into the same rivers from the sea to breed. They are basically the same species of salmon, but long ago a few stragglers switched form breeding in “even” years to breeding in “odd” years. Since the breeding seasons of the two classes don’t overlap, they became instantly reproductively isolated from each other in a unique way: they couldn’t interbreed, but at the moment the year divergence began they were genetically identical; and they continued to live in essentially the same environment. Yet over thousands of years the even- and odd-year forms have diverged, to the extent that they are somewhat reproductively incompatible when bred together—the beginning of speciation. This shows that a eukaryotic species living in very similar environments can still diverge genetically, and begin the process of forming new species.

Second (and the authors note this in passing), there could be considerable internal biochemical evolution taking place that can’t be detected from simply looking at the fossil bacteria or seeing if they metabolized sulfur in similar ways. After all, bacteria are morphologically simple, and there’s simply not that much room for visible change, especially if you’re constrained to look at fossil bacteria in rocks. And it’s impossible to sequence the DNA of the fossil bacteria or grow them in the lab, so we can’t see how genetically and metabolically different they are.

This is not just a theoretical possibility, for biologists are well familiar with this phenomenon. We know of many cases of “sibling species”: distinct species of closely-related organisms that can’t be told apart through morphology alone, but have diverged considerably in their non-visible characters. For example, I worked on a group of 8 Drosophila species in which females couldn’t be told apart by just looking at them, even under the microscope. (Males differed very slightly in their genitalia). Yet despite their morphological conservatism, the species diverged profoundly in ecology and in their reproductive compatibility: they don’t like to mate with members of the other species, and in many cases the inter-species hybrids were either inviable or sterile.  Genetic analysis showed considerable divergence in the DNA, including in important traits affecting reproduction. The problem is even more severe if you are constrained to look at fossils, in which only the hard parts become mineralized. Important differences in softer parts that could reflect genetic change (granted, not that much of a problem in bacteria) could be missed.

The lesson is that it’s dangerous to use fossils—even bacterial fossils—to conclude that evolution hasn’t occurred.  And the “null hypothesis of Darwinism” is a bit dubious anyway. Yes, species probably change most rapidly when the environment is changing, but there’s no reason why environmental change is a sine qua non for evolution.

The paper by Schopf et al. does show an intriguing case of morphological and metabolic stasis over billions of years, and it probably does reflect a lack of environmental change. But what it doesn’t show is that evolution hasn’t occurred in these bacteria. To know that, we’d have to have them all alive to sequence their DNA and look at their physiology, reproduction, and so on—and that’s impossible for the fossil species.

Schopf, J. W., A. B. Kudryavtsev, M. R. Walter, M. J. Van Kranendonk, K. H. Williford, R. Kozdon, J. W. Valley, V. A. Gallardo, C. Espinoza, and D. T. Flannery. 2015. Sulfur-cycling fossil bacteria from the 1.8-Ga Duck Creek Formation provide promising evidence of evolution’s null hypothesis. Proc Nat. Acad. Sci. USA, online Early Edition. 10.1073/pnas.1419241112.


  1. Posted February 4, 2015 at 9:02 am | Permalink


  2. Dominic
    Posted February 4, 2015 at 9:08 am | Permalink

    Where there is little selective pressure I would have thought that more mutations that were not actively harmful might proliferate? I have not checked the paper but I wonder how many samples of the modern bacteria were sequenced to see if that might be the case.

    • darrelle
      Posted February 4, 2015 at 10:25 am | Permalink

      I had the same thought. But how would mutations that were not actively harmful, neutral mutations, become prevalent in a population? Beneficial mutations, sure, but neutral ones? Conceivably it could happen by chance, but that seems improbable. Perhaps an indication of long periods of unchanging environment, and therefore low selection pressures, is a large variety of different neutral mutations in a population?

      Of course, it is a spectrum, not digital. If there is even a slight benefit then given enough time the mutation should become prevalent in the population.

      • Dominic
        Posted February 4, 2015 at 11:13 am | Permalink

        I was thinking of so-called ‘junk’ DNA…

        • Mark Sturtevant
          Posted February 4, 2015 at 12:43 pm | Permalink

          Bacteria will have rather little junk DNA. But there are 3rd bases of codons for example.

          • Dominic
            Posted February 5, 2015 at 3:15 am | Permalink

            They are too ‘clever’ to! Thanks…

    • Posted February 4, 2015 at 11:27 am | Permalink

      Exactly. To play Larry Moran (and PZ Myers) a bit, not all evolutionary changes are selected for, and neutral mutations and genetic drift are likely to play a larger role at the genetic level than natural selection.

      • Fré Hoogendoorn
        Posted February 5, 2015 at 5:31 am | Permalink

        Given that we’re talking about bacteria here, do they undergo random genetic drift? Sure, you’d get changes due to horizontal gene transfer and neutral mutations where this is possible (third base in codons, etc.), but when bacteria divide, you can’t even tell which one is the parent and which the next generation, not to mention that they would be identical as well.

        • GBJames
          Posted February 5, 2015 at 7:18 am | Permalink

          Drift is within a population over time. So there’s no reason bacteria wouldn’t experience mutations that are not themselves subject to selection and but become fixed by drift just as happens in your fancy-schmancy creatures.

          • Fré Hoogendoorn
            Posted February 5, 2015 at 9:12 am | Permalink

            The definition of random genetic drift, according the WEIT, is: “Evolutionary change that occurs by random sampling of different alleles from one generation to the next. This causes nonadaptive evolutionary change.”

            When a bacteria divides, there is no random sampling, such as is seen in sexual reproduction, just a copying of the entire genome. Hence my question.

            Sure, there may be errors in the copying process, but that is not drift.

            • GBJames
              Posted February 5, 2015 at 9:38 am | Permalink

              An error in the copying process is a mutation event, unless I greatly misunderstand the word. Evolution is always a matter of changes in the proportion of various alleles in a population. So I don’t see how it matters whether the reproduction is sexual or asexual.

              Perhaps Jerry can clarify this for us laity.

            • daveyc
              Posted February 5, 2015 at 10:25 am | Permalink

              While drift can occur due to the mechanisms of sexual reproduction, it can also occur completely independently of the process of reproduction. For example, a population bottleneck will cause drift to occur in a bacterial population, just like it would in a eukaryotic population.

  3. Paul Rose
    Posted February 4, 2015 at 9:09 am | Permalink

    Nice capsule view. Could be a good teaching aid.

  4. Posted February 4, 2015 at 9:12 am | Permalink

    Thanks for the clarification. I’d read a couple of press reports and was wondering about the validity of the claim.

  5. Martijn
    Posted February 4, 2015 at 9:13 am | Permalink

    In short: phenotype genotype. Right?

  6. Mark Perew
    Posted February 4, 2015 at 9:20 am | Permalink

    Isn’t this the “Why are there still monkeys?” issue? Evolution and emergence of a new species doesn’t require the extinction of the progenitor species. Or, am I missing something? If I am, please help understand this better.

    • Dominic
      Posted February 4, 2015 at 9:46 am | Permalink

      There are diachronous & synchronous ways of looking at this. If we have A species a that gives rise to species B, even if it still LOOKS as if species A is still around, let’s call it A1, we could not say that the present population of A1 is the progenitor of B. Both descend from A but A1 has possibly changed much less – or at least from morphological standpoints as PCC says above.

      How far do we stretch the existence of a species through time – diachronous? If you plot a graph with a population of modern humans, some have neanderthal genes (human subspecies or separate species depending on your view) & some do not. Some have Denisovan genes & some do not.

      Species are messy!

    • Jonathan Wallace
      Posted February 4, 2015 at 10:02 am | Permalink

      I don’t think it is fair to characterise this (i.e the paper by Schopf et al – I don’t know what the press have said about it and can believe that some reports may have gone way off beam) as the “Why are there still monkeys?” issue. The paper does not seek to argue that modern forms cannot have evolved from primitive ones given that there are still primitive forms in existence which I believe is the nature of the ‘why are there still monkeys’ “argument”.
      The paper does seem to show an interesting example of morphological stasis in a particular type of bacteria over a long time period and relates this to a very stable environment. However, as PCC has made clear there could have been significant biochemical change over this period that is undetectable and furthermore the null hypothesis that organisms won’t evolve in a stable environment is a flawed notion.
      You are correct to say that emergence of a new species does not necessitate the extinction of the progenitor species. In the salmon example that PCC describes the odd-numbered years breeders appear to be on their way to becoming a new species but the even-number year breeders which are the progenitor species continue to live on in the same environment.

  7. DTaylor
    Posted February 4, 2015 at 9:21 am | Permalink

    A really helpful discussion, Jerry. Thanks.

    • Delphin
      Posted February 4, 2015 at 11:24 am | Permalink

      Ditto. JAC lamented the lack of response to these threads. Well, most of us have little of use to add. But I appreciate the clear thinking and careful reasoning in explicating the theory.

      • quiscalus
        Posted February 4, 2015 at 10:26 pm | Permalink

        exactly. I’m not a microbiologist, I have a textbook I purchased for the fun of it, but like so many other things, I’ve not gotten around to reading it (like JAC’s Speciation, and Futuyma’s Evolution, sitting forlornly on the bookshelf) and my initial thoughts were covered in the Prof’s post, specifically the issues surrounding bacterial “species” and that morphology isn’t a dead ringer for determining either species or evolution, which is why fossils wouldn’t be very good bets, as geneticists would tell you (and paleontologist and taxonomists as well, I imagine) but me saying so doesn’t add anything other than to make me feel good for thinking along the same lines. Still, I hope the Prof. appreciates those of us who at least comment to demonstrate that we read and enjoyed the post.

  8. eric
    Posted February 4, 2015 at 9:27 am | Permalink

    The ‘Darwin null hypothesis’ doesn’t seem very, um, Darwinian to me. Here is real Darwin, from OOS:

    Isolation also is an important element in the modification of species
    through natural selection. In a confined or isolated area, if not very large, the organic and inorganic conditions of life will generally be almost uniform; so that natural selection will tend to modify all the
    varying individuals of the same species in the same manner.

    So he’s not saying evolution in such an environment like the vents won’t occur, he’s saying speciation won’t occur. Here’s another:

    although small isolated areas
    have been in some respects highly favourable for the production of new species, yet that the course of modification will generally have been more rapid on large areas; and what is more important, that the new forms
    produced on large areas, which already have been victorious over many competitors, will be those that will spread most widely, and will give rise to the greatest number of new varieties and species. They will thus play a
    more important part in the changing history of the organic world.

    Again, Darwin is saying that speciation and variation won’t occur as much in small isolated areas. This is subtly different, but importantly different, from claiming a species will remain in genetic stasis.

    My search wasn’t comprehensive at all, but the closest thing I could find to the null hypothesis was this:

    That natural selection generally act with extreme slowness I fully admit.
    It can act only when there are places in the natural polity of a district which can be better occupied by the modification of some of its existing inhabitants. The occurrence of such places will often depend on physical
    changes, which generally take place very slowly, and on the immigration of better adapted forms being prevented.

    So, natural selection can only act when there is room for improvement, which depends on the currently existing ecosystem. Okay. But is this the null hypothesis? Not really. Evolution still occurs, what’s happening is really that the variants nature selects are the unmutated ones. But mutation and selection as mechanisms still occur.

    • eric
      Posted February 4, 2015 at 9:33 am | Permalink

      Also, a citation and copy edit apology: sorry for the excessive line breaks. The quotes are from the Project Gutenberg plaintext version (6th Ed), which has line breaks. I deleted the ones I saw but some snuck through because they aligned with the natural line breaks in the comment editor box. Next time I’ll copy it to Word and clean it up first.

    • Posted February 4, 2015 at 1:20 pm | Permalink

      Sorry to ask, but I want to try to grasp what your comment said.
      In your last paragraph, are you saying that in an environment in which there is no “room for improvement” selection and mutation occur but are less perceptible?
      The reason I ask is that it has always been my (admittedly very limited) understanding that evolution is a sort of “if it ain’t broke, don’t fix it” process. My high school biology teachers explained that this is why sharks have been around for 400+ million years, because they are well adapted and thus have no need to evolve.
      I’m starting to think Mr. Adams & Ms. Burton were wrong, or at least didn’t have a complete answer.

  9. Posted February 4, 2015 at 9:29 am | Permalink

    Type 2 error.

  10. GBJames
    Posted February 4, 2015 at 9:29 am | Permalink

    Thanks for this post, Jerry. I had seen the claim in the press and expected it to be wrong/incomplete but your discussion makes it clear.

  11. Randy Schenck
    Posted February 4, 2015 at 9:30 am | Permalink

    Excellent to get this understanding. To read some of the news on this, would leave us to believe that evolution had stopped on this particular bacteria. This is a good example of news and journalism giving the public a not so good look at science.

  12. Jimbo
    Posted February 4, 2015 at 9:35 am | Permalink

    Great review. I was also underwhelmed by the conclusion. Agree with point 2: genetics has moved us WAY beyond comparisons based on morphology. Further, subtle changes in morphology is easier with big critters (eg sharks) and more subtle with prokaryotes, especially fossilized ones.

    That said, can a genetically uniform bacterial “species” evolve with no environmental changes whatsoever? Great question and we appear to have the answer: yes. Check out the elegant experiments of Lenski who bred 60,000 generations of E Coli over decades.

    Result: very rare spontaneous changes to metabolism occurred. It seems to me that DNA copying errors, environmental mutagens, and radiation will always result in unavoidable genetic mutation and drift. Interesting aspect of this work is the possibility that deep ocean organisms experience very little radiation (no uv, maybe X-rays and cosmic rays).

  13. Paul S.
    Posted February 4, 2015 at 9:37 am | Permalink

    Obviously I don’t understand evolution as well as I thought I did, so I’ll ask a novice question. Is it possible that this bacteria has both survived for Ma as is and also produced mutated offspring that may have continued to evolve and or become extinct?

    • eric
      Posted February 4, 2015 at 10:16 am | Permalink

      Well I think Jerry’s (and other people’s) points is that staying morphologically “as is” is not good evidence that the population stayed genetically “as is.”

      Having said that: yes, its possible that the species in question produced daughter species which then migrated away or were not preserved in the fossil record for some other reason. I don’t know whether that’s likely or not, maybe someone else has a better understanding of the fossil records they used and how comprehensive (of the time and region) they can be considered to be.

  14. Posted February 4, 2015 at 9:41 am | Permalink

    Everyone seems to be assuming that evolution equals natural selection. But we know that evolution at the molecular level never stops, as neutral mutations become fixed at a fairly constant rate. Even without any selection, the genomes of modern bacteria must be very different from the genomes of ancient ones. Or don’t the authors count neutral evolution as real?

    • Jacques Hausser
      Posted February 4, 2015 at 10:03 am | Permalink

      Yes. But selection itself can itself be canalizing and play its part, that is, in such a biotope, it can eliminate lineages which depart from an optimum – that would explain the morphological and metabolical stability of purely asexual “species”.

      • Jacques Hausser
        Posted February 4, 2015 at 10:04 am | Permalink

        remove the first “itself”

    • daveyc
      Posted February 4, 2015 at 12:20 pm | Permalink

      Exactly. This very obvious (to a biologist) point seems to be frustratingly missing from the headlines I’ve seen.

  15. Jim Knight
    Posted February 4, 2015 at 9:44 am | Permalink

    At least the authors couched their terms in the title: “…provide PROMISING evidence…” This paper points to new, possibly testable, directions for future research.

  16. Daniel Engblom
    Posted February 4, 2015 at 9:56 am | Permalink

    Very interesting and clear discussion Jerry, thank you!

  17. Ken Pidcock
    Posted February 4, 2015 at 10:01 am | Permalink

    Professor Coyne mentions how difficult it is to define bacterial species. Let someone try their hand at describing what bacterial speciation would entail!

    • Posted February 4, 2015 at 10:44 am | Permalink

      This is tripping me out. I assume the speciation puzzle is because they reproduce by division and not sex? Are they then all the same species with different morphologies and food sources? And are the other levels in the taxonomy similarly conveniences based on those traits? Or is it just that the word “species” is itself only strictly meaningful in animals that have sex?

      Does the same species vaguerie apply to viruses?

      • Torbjörn Larsson, OM
        Posted February 4, 2015 at 11:45 am | Permalink

        I’m a layman here, but due to conjugation et cetera, I think one can say that bacteria do have variants of “sex” where they exchange gene material that goes into the future lineages. How complete and analogous they are re the concept of biological species, I don’t know.

        But I think the situation is like this. A biological species barrier doesn’t make much sense for bacteria. I think, though I haven’t read Jerry’s texts, that he identifies “species” based on that – it is a very useful classification.

        But there are 25+ attempts of making more or less useful species conceptions. (One can google up lists describing them.) That which I have seen applies most usefully for bacteria is “ecological species”. However see below on diversity and ecological successions in bacterial mats, it looks to me from Des Marais talk that genome classification is much more useful. How to define “species” or at least classes from genome sequencing, I don’t know.

        • Torbjörn Larsson, OM
          Posted February 4, 2015 at 11:48 am | Permalink

          “That which I have seen applies most usefully for bacteria” – That which I have seen claimed to apply most usefully for bacteria…

      • Posted February 4, 2015 at 12:02 pm | Permalink

        There are proposals by some biologists and philosophers of biology to generalize the notion of species. (David Hull, I believe, comes to mind.)

    • Posted February 17, 2015 at 7:37 pm | Permalink

      The Ecological Species Concept is the definition to go by (Van Valen, 1976) when talking about bacterial species (imo).

      I’d also like to point to a paper I published last year on that problem: Trade-offs drive resource specialization and the gradual establishment of ecotypes, Østman B, Lin R, Adami C (2014) BMC Evolutionary Biology, 14:113.

  18. Kevin
    Posted February 4, 2015 at 10:05 am | Permalink

    Li-5 has a half life of about 10-24 seconds; U-238, 10+15 (>4 billion years). That’s 40 orders of magnitude for the same physical phenomena: radioactive decay. How is this bacteria, in any way, outside the realm of not being affected by evolution? And by that, I mean its environment.

    Physics can know precisely what the ‘environment’ can be: a collapse of the wave function or a connection to thermal reservoir. Laws of physics drive evolution on a thermal scale. Put any bacteria in a cold dead rock for a billion years and it will still have to contend with statistical physics.

  19. Genghis
    Posted February 4, 2015 at 10:17 am | Permalink

    Did Lenski’s citrate-eating E.coli look any different from the original strain?

    • peepuk
      Posted February 4, 2015 at 12:15 pm | Permalink

      Yes, bigger cell sizes and more rounded shape.

      • Jimbo
        Posted February 4, 2015 at 12:23 pm | Permalink

        But how do the two fossilize?

  20. Posted February 4, 2015 at 10:21 am | Permalink

    I was hoping when I read the headline that Schoph et al had something more than bacterial morphology to go on! I’ve had a figure showing striking morphological similarities in ancient and modern bacteria in my History of the Earth lecture for at least the past 10 years.

    I expected to see new genetic evidence of evolutionary stasis. And so I am still wondering: What happens when you sequence apparently identical bacterial ‘species’ taken from locations that are widely separated by both geography and inhospitable habitat? E.g., particular thermophilic bacteria that are found in the isolated hot springs on the sea floor off the coast of South America and the hot springs in Yellowstone National Park? I would think that a comparison of ‘junk’ DNA to estimate divergence times and the more active DNA to see how much metabolic or other processes have changed would be necessary to document stasis.

    • Posted February 4, 2015 at 11:03 am | Permalink

      a community of sulfur-using bacteria discovered in 2007 in the seafloor off the west coast of South America

      This was the same point that stuck out for me: if the South American sulfur-using bacteria were chosen because their environment was most like that of the fossil bacteria, that makes sense – but then one would be surprised if morphology and the surrounding matrix were much different (same guys metabolically + same single food source => same output). So are there any genetic differences across the same-looking type of bacteria in different locations?

      And on a related question, how similar in morphology and matrix might two known-to-be-distinct forms of bacteria appear to be. As a Republican might say, “I’m a notascientist,” but this question overlaid with your question would inform at least the probability and degree of evolutionary change (or stasis) in the subject bacteria.

      This paper seems to me would be good news for baraminologists: Look! Finally one that’s exactly as it was on the ark! I guess Noah’s high-pressure molten sulfur tanks were inadvertently left out of the final edit.

      • Torbjörn Larsson, OM
        Posted February 4, 2015 at 11:29 am | Permalink

        This reminded me of two pieces of information I have picked up recently:

        – Mass sequencing of ocean water has shown a huge diversity:

        “„The pelagic and benthic communities strongly differ on all taxonomic levels“ says Lucie Zinger, „with less than 10 % of the bacterial types occurring in both ecosystems.“ Alban Ramette explains this observation as follows: „ The various realms may offer habitats with very different conditions which would impose strong selective pressure on microbial communities. The authors also reported that deterministic factors of community change were identified to be productivity (i.e. the formation of biomass) and geographic locations. In addition, the heterogeneity and dynamic nature of coastal and vent habitats explains the the high community variability of these ecosystems.“”

        [ http://www.mpi-bremen.de/en/Microbial_Diversity_in_the_Oceans.html ; my bold]

        – Biomat communities are very diversified with ecological roles, where the exact spieces differ depending on environmental cues.

        “He will show how several previously unknown rRNS gene sequences of bacteria and eukarya were identified, indicating that these mats can extend our understanding of the diversity and early evolution of benthic microbial communities. He continues to catalog the diversity of lipid biosignatures, whose fossil equivalents can record the diversity of ancient microbial ecosystems.”

        Des Marais thinks of biomats as “rainforests”, because they are as productive as those are, and “cities”, because they are as diverse as the inhabitants. The interesting piece here stars about 30 minutes into his talk.

        Note that Des Marais show that the mat inhabitants somehow contrive to do stuff that they usually can’t do in a similar environment. (Reduce sulfur in a highly oxygen rich environment.)

        [ https://www.youtube.com/watch?v=n-Qb9om_vLk&list=PL7B4FE6C62DCB34E1&index=4 ; Microbial mats and Earth’s Early Biosphere – David Des Marais (SETI Talks)]

        So indeed, are there any genetic differences across the same-looking type of bacteria in different locations? The data seem to suggest there is, and that this is nearly as old a situation as life is.

      • eric
        Posted February 4, 2015 at 12:14 pm | Permalink

        And on a related question, how similar in morphology and matrix might two known-to-be-distinct forms of bacteria appear to be.

        In principle, very different. I don’t know how well this extrapolates to bacteria, but some varieties of onions have five times the amount of DNA as others. The Onion Test is typically used as an argument against the “no junk DNA” position, but it also highlights how very similar species or variants within a species can show large genetic differences, even without large morphological differences.

  21. Posted February 4, 2015 at 10:47 am | Permalink

    Very interesting post.

    Your comment:

    Even in an unchanging environment, organisms can still evolve in significant ways. If new mutations arise that adapt the species better to that unchanging environment, then we will have evolution. For example, the bacteria could evolve more efficient metabolism of sulfur.

    Is also the very first thing that struck me. Seems to me that what is usually most interesting (and mutable) about bacteria is their chemistry (chemical “machinery” and pathways). And they can’t get the details of the chemistry in the fossil bacteria.

  22. Torbjörn Larsson, OM
    Posted February 4, 2015 at 10:48 am | Permalink

    Thanks for this. I saw the press release, and I wasn’t that happy after a first read. Unfortunately the link is paywalled for me, but Jerry’s analysis makes me happier.

    I must confess Schopf and images of filaments always make me cringe a little. He, by all means not alone, was very optimistically pushing those as signs of early life. The so called Apex ‘microfossil’ debate; as I understood it from taking my first astrobiology course those were later rejected because there are so many geological processes making these types of structures. We need to use a comprehensive list* of biogenecity criteria to test these fossils vs “self-organizing structures”. [“A fresh look at the fossil evidence for early Archaean cellular life”, Brasier et al, Phil. Trans. R. Soc. B, 2006.]

    I see from Brasier’s paper that some filaments that are older than 2 billion years remained as accepted finds, because they were consistent with biotic origins. (At least as of -06.)

    * I’ve seen papers where they want to make consensus lists on specific microfossils such as bacterial etched tracks in volcanic glasses, because of the problematic false positives.

    • Torbjörn Larsson, OM
      Posted February 4, 2015 at 10:52 am | Permalink

      I should say that it is the _juxtaposition_ of Schopf and images of filaments that makes me cringe…

  23. CB
    Posted February 4, 2015 at 11:16 am | Permalink

    Thankyou for the very clear run down of evolutionary principles.
    Without a genetic comparison it is impossible to say anything definitive. Who knows maybe it is a case of convergent evolution and the ancient species died out long ago.
    It almost seems like magical thinking to claim no evolution-change is a universal and it is impossible that NO change has occurred which brings the question of what we are willing to call evolution-simple genetic changes or whatever- pigs to giraffes. Evolution has become somewhat of a buzzword-but in my view it is all evolution-small genetic changes with no discernible phenotypic change nevertheless offer a different background for further inevitable change.

  24. Posted February 4, 2015 at 11:20 am | Permalink

    Reblogged this on Santa's Reindeer and commented:
    Really interesting take on evolutionary theories….

  25. CB
    Posted February 4, 2015 at 11:24 am | Permalink

    One further thought. Maybe the structure of the genome of the current organism would give some clues about changes-if there are duplications and metabolic networks of some complexity etc.

  26. Dominic
    Posted February 4, 2015 at 11:26 am | Permalink

    It has a limited niche – just one way to live – so although other ‘ways of being’ (a creature) have changed little superficially, eg thinking bivalves around for 500 my, they have had lots of niches to fill & so grown to a great variety. The salmon have essentially occupied one niche at separate times so do not compete.

    • darrelle
      Posted February 4, 2015 at 12:57 pm | Permalink

      RE the salmon not competing, I don’t think that can be considered a given. Though they are not occupying the same specific areas of their shared ecosystem at the same time, they are still sharing the ecosystem. Time sharing should certainly result in a less concentrated resource load on the ecosystem, but there would still be limits.

      Also, each population could affect the ecosystem in ways other than direct competition for resources, and those changes could result in selection pressure on the other group.

      • GBJames
        Posted February 4, 2015 at 1:59 pm | Permalink

        The question is not if there are “limits” in some absolute sense, it is whether such limits as exist are set by the other population. In this case, I don’t see what they might be.

        • darrelle
          Posted February 4, 2015 at 3:14 pm | Permalink

          Yes, that is what I meant. I don’t know enough to make any informed guesses for this particular case. Just speculating a “for example” possibility.

          1) The salmon spawn results in a high level of some waste product that affects the levels of a microbe in the water, and the levels take some time, a couple of years, to return to what they were before the spawn.

          2) The salmon population splits. Early after the split the numbers of each group are low enough that the microbe level that spikes after a spawn doesn’t spike as high and returns to normal in less time.

          3) The two populations increase and the microbe level spikes higher after a spawn and takes longer to return to normal. Long enough that it is now higher at the start of a spawn than it has ever been in the past. And the microbe has some affect on the eggs of the salmon.

          • Dominic
            Posted February 5, 2015 at 3:18 am | Permalink

            Good points… ta.

          • Dominic
            Posted February 5, 2015 at 3:22 am | Permalink

            …& we have not even mentioned kin selection!

  27. JH
    Posted February 4, 2015 at 11:39 am | Permalink


    Thanks! That was an easy to understand translation of the paper.

  28. Posted February 4, 2015 at 11:43 am | Permalink

    Jerry, a most interesting post, I rank it in your top ten. It is a post that gives us laymen food for thought.
    Bacteria is/are, as you pointed out, somewhat different from us eukariotes, bacteria are geared to one thing, and that is direct reproduction. Was it Jaques Monod who said that a bacterium ‘wants’ to be 2 bacteria?
    Never mind who said it, it appears to be so, Bacteria are morphologically highly streamlined and metabolical wizards. Their morphology is basically spherical (cocci) or short rods (bacilli). They can clump, form chains (as above), make duet (e.g., gonococcus) or just stay alone. That is only 4 possibilities. There is indeed not much to be concluded from bacteria remaining in morphologically identical strains/chains over a billion years. We cannot (at least with the present technology) know how they altered -or not- their biochemistry. Your point appears very sound to me.

    Thank you for the salmon -mind boggling. I erroneously thought that sympatric speciation was a nearly exclusive insect way.

  29. Benjamin Branham
    Posted February 4, 2015 at 11:58 am | Permalink

    Great follow-up and clarification, Dr. Coyne. I had wondered on what your thoughts on speciation without selective pressure would be. Very insightful and educating, as expected.

  30. daveyc
    Posted February 4, 2015 at 12:13 pm | Permalink

    According to the headlines I have read, evolution at the molecular level apparently does not count as “evolution.” As somebody whose primary research interests include molecular evolution, this is very frustrating indeed.

  31. peepuk
    Posted February 4, 2015 at 12:50 pm | Permalink

    Reminds me of Sir David Attenburrough :

    “humans have stopped evolving physically and genetically because of birth control and abortion”

    I’m a big fan of him but this is seems to me nonsense.



    • alnitak
      Posted February 4, 2015 at 4:21 pm | Permalink

      Sir David missed on that one. Descent always comes with modification. The only question is what the environment (including us) is favoring.

  32. Mark Sturtevant
    Posted February 4, 2015 at 12:56 pm | Permalink

    It is still an interesting paper, and worth thinking about, IMO, but even put into its most favorable light this will not of course overturn the general fact that morphologies have and are evolving.
    I wonder how long it will take for creationists to misinterpret this one, and trumpet that scientists have discovered that evolution does not happen.

  33. Posted February 4, 2015 at 1:07 pm | Permalink

    It’s a decent paper, and doesn’t fail the first test of a science paper—does it tell us something new?

    Where does “replication” fit into your definition of a science paper?

    • Posted February 4, 2015 at 2:48 pm | Permalink

      Of course results should be replicated at some point because doing so increases confidence in a supported hypothesis. Also of great value is when attempts to replicate fail to reproduce results. That too helps improve understanding.
      I think of the roles of ‘new’ science and ‘replication’ science to be like the roles of growing roots in a plant. They grow into new territories providing new sources of water and nutrients, but also roots should enlarge and generally improve how they anchor the plant to reinforce what the plant has done.
      So replication is also crucial for any description of how science is best done.

  34. Marilee Lovit
    Posted February 4, 2015 at 1:17 pm | Permalink

    Thanks very much for this post. Very interesting and accessible to amateurs like myself. More than anything else, this clear explanation of research in topics of evolution, is what keeps me coming back to WEIT.

  35. Gregory Kusnick
    Posted February 4, 2015 at 1:24 pm | Permalink

    What evidence is there, apart from morphology and habitat, that these three bacterial species are even part of the same lineage? How do we know that the modern ones aren’t descended from an independent colonization of a similar niche, and the apparent morphological and metabolic similarities are due to convergence?

    The very fact that the fossil species were found on dry land in Australia tells us that their deep-sea habitat wasn’t geologically stable over billions of years.

    • Posted February 4, 2015 at 2:59 pm | Permalink

      I do not think we can be certain. But I suspect (do not know that were we to learn about the metabolic pathways of bacteria living in these environments billions of years ago, we would see that they are very similar to the modern ones due to homology (shared ancestry) rather than to convergent evolution. It strikes me, in the gut mind you not by lots of known precedent that I can think of, that elaborate metabolic pathways that were honed by natural selection billions of years ago to work in a continuously stable environment would persist in that environment rather than be replaced by some upstart pathway trying to muscle in by convergence.

      • Gregory Kusnick
        Posted February 4, 2015 at 5:40 pm | Permalink

        But is there any such thing as a continuously stable environment over billions of years? New seafloor is constantly being created, and old seafloor subducted or uplifted, on timescales shorter than that.

        So what I’m suggesting is not that new species displace old species from stable habitats, but rather that new species colonize newly formed habitat, and then coevolve with their habitat as it ages. The fact that they end up looking the same would then be a result of selection pressures exerted by that evolving habitat.

        Or is it really the case that the hypothesis of an unbroken line of descent in an unchanging environment is such a slam-dunk that there’s no need to consider alternatives?

  36. Posted February 4, 2015 at 2:55 pm | Permalink

    Red Queen Hypothesis, anybody? I would expect selective pressure simply from the viruses who attack those bacteria (unless there are none?).

    • Ken Pidcock
      Posted February 4, 2015 at 3:22 pm | Permalink

      With respect to viruses, I think it’s widely accepted that there are never none.

  37. Chinahand
    Posted February 4, 2015 at 3:16 pm | Permalink

    I just wanted to simply say thank you for this post and the work you put in to write it. Science for the win!

  38. alnitak
    Posted February 4, 2015 at 4:17 pm | Permalink

    These bacteria are still evolving, we just can’t detect the changes. Descent with modification is an iron rule, a zoo keeper’s dilemma, because no matter what the keeper does, the animals change as they replicate. Slow evolution is quite possible, but only perfect DNA replication can stop the river out of Eden in its tracks. Near-perfect replication means slow change, a very slow current, but still it moves.

  39. Diana MacPherson
    Posted February 4, 2015 at 5:13 pm | Permalink


  40. quiscalus
    Posted February 4, 2015 at 10:11 pm | Permalink

    Does this make me a bad person, when upon attempting to read this I got sidetracked thinking “Married sex proves Darwin’s null hypothesis”.

    • gravelinspector-Aidan
      Posted February 5, 2015 at 7:50 am | Permalink

      You’d have to ask … Emma?
      I’ll get a ouija board.

  41. Posted February 5, 2015 at 3:31 am | Permalink

    It is clear that the paper claim that bacteria stopped evolving is not supported by the evidence.

    There is very little morphological differences among bacteria when compared to the vast morphological varietes found in multi-cellular organisms. To claim that bacteria are not evolving because they are not changing their morphology is nonsense.

    Bacteria evolve all the time, even without selective pressure and radiate even in the same environment, wihtout geographical isolation (simpatric divergence).
    Furhtermore, bacteria eagerly acquire new DNA by lateral transfer, so the genetic difference even between two E. coli strains (same species!) could be higher than 20%.

  42. gravelinspector-Aidan
    Posted February 5, 2015 at 7:49 am | Permalink

    I tried following the link at the end of Prof Coyne’s article, but got asked for a PNAS log in (which I don’t have. Though … I might have this afternoon. Depend’s on an Oil company’s site licenses.). So I’ve still not read the paper.

    That’s old, but it doesn’t take the prize for age, for the oldest known bacteria (cyanobacteria, formerly known as “blue-green algae”) come from about 3.5 billion years ago.

    I think that this is a reference to Schopf’s mid-1990s work, also on Australian specimens. While the dating is valid to within a few tens of millions of years precision, this paper was debunked in about 2002 by the Martin Brasier (sadly, recently deceased ; he wrote a cracking autobiography a couple of years back which makes him sound a really fun guy to do field work with. I don’t know if he’d finished volume 2). While Schopf’s 3.5Gyr claim may still be valid, re-examination of the original specimens by Brasier and his students failed to relocate a number of the specimens originally described (they are very microscopic ; this may well be simply a clerical error), recorded morphology in some recovered specimens which Schopf claimed wasn’t seen, and most importantly a geothermobarometry analysis of the mineral assemblage in Schopf’s samples showed temperatures and pressures considerably higher than Schopf’s environmental interpretation. This effectively pushed the formation of the specimens well down into the throat of a hydrothermal system and well beyond the known limits of life and protein (enzyme stability. I forget the exact numbers, but it was well in excess of 100 Kelvin higher than any known life forms today.
    These don’t completely invalidate Schopf’s claim of 3.5 Gyr life forms, but the work was considered seriously undermined.
    Since other specimens of less ambiguous life forms are securely dated to 3.2 Gyr, and controversial (Mosijzes – however his mane is spelled – C-13 anomalous graphite from Greenland) “signs of life” date to 3.9 Gyr, this doesn’t greatly affect the overall story of the origin of life on Earth. But it did damage Schopf’s reputation.
    From what I’ve seen quoted from the actual paper, Schopf has been considerably more cautious in his statements than the journalists commenting on his (et al) work. but I haven’t found a copy of the paper (Prof CC, could you oblige?) to actually read and comment on.

2 Trackbacks/Pingbacks

  1. […] Jerry Coyne has blogged the same topic: […]

  2. […] https://whyevolutionistrue.wordpress.com/2015/02/04/a-new-paper-claims-that-evolution-has-stopped-in… […]

%d bloggers like this: