Newly found: the world’s oldest fossils!

August 23, 2011 • 5:27 am

The Earth is 4.54 billion years old.  How long did it take life to arise after that?  While there are tantalizing hints of life from Greenland dating back about 3.75 billion years ago (“bya”; these hints come from carbon deposits that may have been formed by living organisms), there’s nothing like a fossil to establish the existence of life.  And, of course, the earliest organisms were very simple, like bacteria.  Fossils of bacteria or any single cell—”microfossils”—are hard to detect in the rocks, for small particles, like carbon-encrusted mineral grains, can easily look like bacterial “microfossils.”

This difficulty in distinguishing true bacterial fossils from simple inclusions led to a big scientific kerfuffle in 2002, when Bill Schopf claimed to have found the world’s oldest fossils, of cyanobacteria (once called “blue green algae”), in Australian rocks dated 3.465 billion years ago. That would have made them the world’s earliest fossils, but Martin Brasier, of Oxford, took exception, claiming that these were mere particles of heated graphite. There were even claims that Schopf had selectively published his data, leaving out photographs of the “bacteria” that didn’t look so bacterial.  You can read about the 2002 controversy here.  Time has come down on the side of Brasier, with most paleobiologists agreeing that the objects Schopf described were not living creatures.

Ironically, though, Brasier is the senior (last) author on a new paper by David Wacey et al. in Nature Geoscience (see report at the BBC here) that does show, pretty conclusively, the oldest known fossils on Earth, dated at 3.4 billion years old.  This makes them only a tad younger than the fossils claimed by Schopf but 200 million years older than the oldest reliable microfossils known up to now.  And the fossils described by Wacey et al. are much better documented  than were Schopf’s, demonstrating that prokaryotic organisms had already evolved only a billion years after Earth had formed.  These are also bacteria, of course, but their association with pyrites, and their morphology, also show that they were sulfur-metabolizing bacteria.

The new microfossils were found in a sandstone formation in Western Australia; the photograph from the BBC shows the study site, and notes that the new fossils were found at the base of these ridges:

Here are some photos of the microfossils from Figure 1 of Wacey et al.; I’ll add the caption below so you can see what’s going on (click to enlarge; clicking again makes it even bigger):

Figure 1.  Examples of spheroidal/ellipsoidal microfossils from the SPF (samples SP9D2, SPE1, SPV3a–c). a,b,e, Clusters of cells, some showing cell wall rupturing (arrows in a,b), folding or invagination (arrow in e). c,d,h, Chains of cells with cellular divisions (arrows). f,i–j, Cells attached to detrital quartz grains, exhibiting cell wall rupturing and putative escape of cell contents (arrow in f), preferred alignment of cells parallel to the surface of the quartz grain (arrows in i), and constriction or folding between two compartments (arrow in j). g, Large cellular compartment with folded walls (arrows).

As the authors note, “determining the biogenicity [biological origin] of putative Archaean microfossils is notoriously difficult.”  How do we know that these things are real remnants of bacteria and not just inclusions or artifacts?  There are several independent lines of evidence, none conclusive but together building a very solid case:

  • They look like cells, being cell-shaped, cell-sized, and forming chains of spheroids that look like chains of both well-established fossil bacteria and modern bacteria.  Some can even be seen “dividing” or expelling their contents after cell damage (see figure above).
  • The variation in size of the bodies is small—smaller than you’d expect if they were abiological inclusions.  A uniformity of size, however, is expected if they’re all members of one living species.
  • The cell “walls” of the microfossils, too, are of uniform thickness, unlike that of artifacts like silica grains coated with carbon.
  • The geochemistry of the bacteria and surrounding rock supports the idea that these are true organisms.  This involves not only the isotopic nature of the carbon, but the presence of nitrogen, a crucial biomarker, within the cell walls.

One of the more exciting features of these fossils is their co-occurrence with small grains of pyrite, a compound of sulfur and iron that is a byproduct of bacteria that metabolize sulfur.  This supports a biological origin not only because the tiny pyrite grains are found in conjunction with the microfossils, but because such grains are also seen in association with modern sulfur-metabolizing bacteria.

I judge the paper a very good piece of work: the authors are careful in their conclusions and state explicitly that no single feature of these objects prove that they’re bacterial microfossils, but that the total weight of evidence strongly supports that conclusion. I agree.  So we have the oldest fossils on earth, which means that simple life, not yet having the complexity of “true” cells, must have evolved well before 3.5 billion years ago.  And perhaps the earliest cellular life metabolized sulfur rather than oxygen.

Does this mean that these bacteria were members of the Archaea, primitive single-celled organisms that often inhabit extreme environments, rather than members of true “bacteria”? Many biologists think that organisms with “true cells” (members of the Eukaryota, like us) are more closely related to the group Archaea than to true bacteria.  As far as I can see, we can’t determine whether the cells of Wacey et al. are real bacteria or Archaea. (Note that although the authors call these “Archaean microfossils,” that does not mean they are members of the biological group Arachaea—only that they occurred during the eon that geologists call the “Archaean,” which lasted from 3.80-2.5 bya.)  But never mind for the nonce—they’re still the oldest indisputable forms of life known on Earth.

I am very curious, though, why this paper was published in Nature Geoscience rather than a higher profile journal like Science or Nature itself.  Both of those journals publish findings that are far less significant than this one, and this finding certainly deserves a berth in the very highest-profile journals. It is, after all, a report of the oldest known life on Earth.  I suspect there are some editorial dynamics here that we don’t know about.


Wacey, D.,M. R. Kilburn, M. Saudners, J. Cliff, and M. D. Brasier.  2011.  Microfossils of sulphur-metabolizing cells in 3.4-billion-year-old rocks of Western Australia.  Nature Geoscience online: doi:10.1038/ngeo1238

54 thoughts on “Newly found: the world’s oldest fossils!

    1. God practiced making bacteria first, then launched right into humans a couple of days later. What’s your problem with that?

        1. He got sidetracked making more, naturally. Look at how many different types of bacteria there are, and how varied their biochemistries. Even more-so than the proverbial beetles.

          Bacteria are his true chosen clade. Humans were just an afterthought. Tacked on to the end of the sixth day, because he had nothing better left to do and got bored.

          1. But, but, but, we’re not special? Oh the pain, oh the agony. You meanie, dad’s gonna get you when he gets home.

          2. I think he ran out of stuff to do, and said to himself, “I wonder what would happen if I made a really big bacterium?” Of course that’s rhetorical, since he’s omniscient and all.

  1. While the popular press has been all atwitter over the implications this has for finding life on Mars, what really gets me excited…is that these might actually be the bones of our ancestors!



  2. Is it correct to call the cells “prokaryotes” rather than “prokaryote-like”?
    We don’t know of these fossil cells are genetically related to todays living prokaryotic bacteria.

    1. Unless one posits multiple instances of abiogenesis, these fossils are both related to modern prokaryotic bacteria and to humans — and, indeed, to everything else alive today.

      Whether or not modern prokaryotic bacteria are descended from these fossils, on the other hand….



      1. Depending on how you define life (for example the occurance of a self replicating biochemical molecule with the ability to ‘evolve’), I think multiple instances of abiogenesis is the most plausible scenario.

        1. Yes, but we as of yet have no reason to suspect that any lineage but ours was extant 3.4 BYA. There are others who can address this with more authority, but I’d personally be rather surprised if these fossils turned out to be completely unrelated to modern life.



          1. Well, I think LUCA dates back to before 3.4 bya, by molecular clock, but calibration is an issue, and there’s nothing to say that other lineages could not have persisted well past that date, only to die off later.

        2. Well, it may have happened more than once on Earth, but I think the overwhelming consensus is that all living organisms are descended from only a single ancestor that formed by abiogenesis. Why else would the genetic code be the same in virtually all species, and that every creature uses L rather than R amino acids?

          1. I agree that the current evidence suggests all current life comes from a single ancestor who had L amino acids and the same genetic code.
            This common ancestor,
            however, probably lived a long
            time after its ancestors
            abiogenesis. We don’t know
            whether the fossil bacteria in the current story had the same genetic code or protein chirality as living organisms.
            The simple biochemistry of cell membranes means that it is likely that a bacteria-like shape is to be expected in early organisms, whether they have the same genetic code or not.

            1. How do you propose that we test this?
              Or are all traces of such chymical admixtures likely to have been obliterated by now?
              If it is not testable, then the proposition remains in the realms of the imaginary.
              I understand that Prof Paul Davies is interested in such investigations of alternate molecular geneses.

            2. While it might be plausible that primitive replicators arose more than once, it seems rather less plausible that they could have remained isolated from each other long enough to independently evolve to the “prokaryote-like” stage.

              1. I think it’s also worth remembering that the first successful replicator would presumably have eaten the competition for lunch and spread like wildfire as a result. There would have been a very small window for multiple progenitors to exist simultaneously.

                With that thought in mind, and considering how readily young planets swap rocks, I wouldn’t at all be surprised if any life we may or may not find elsewhere in the Solar System (Mars, Europa, wherever) also shares a common ancestor with us.



              2. Perhaps if the different abiogenesis events occurred in isolated microenvironments. If, for example, more than one of the current abiogenesis hypotheses turn out to be true, we might have gotten several independent starts to life, whose separate lineages would not have had a chance to intermingle and compete until well after they evolved the capacity to leave their originating environments. If, for example, abiogenesis occurred at both deep ocean alkaline vents and in pockets of polar sea ice (two popular hypotheses I am familiar with). In both those scenarios, early life would not have been able to leave that cradle environment before evolving cell membranes and the ability to maintain them independently. In which case, they would not be able to meet and compete until they both achieved a prokaryotic grade of organization.

                Although the more likely scenario would be that one reaches this grade before the other, allowing it to invade the other’s native habitat, and eat it out of existence.

    2. Prokaryote is not a term of evolutionary relatedness. Bacteria and archaea are prokaryotes but likely not monophyletic. Prokaryotic is simply descriptive, cells lacking a nucleus, so the term fits.

      1. Bacteria and archaea share a common ancestor. The question is whether tje term “prokaryote” is applicable to an organism that is not related to current life. For instance if we found a bacteria-like organism living under the ice of Europa or Enceladus, would it be correct to call them “prokaryotes” rather than “prokaryote-like”?

      2. Bacteria and archaea are monophyletic, as far as we know. The clade that includes them both is “Life”. (Their division is the earliest branch point known among all extant organisms)

  3. The piccys don’t look much like the Jack Hills.

    Did they give the geographical location more precisely than Western Australia?

    (I have a special interest, in that one of my elder sisters is managing a sheep-station[1] that encompasses both the Jack Hills[2] as well as the proposed site for the Square Kilometre Array Telescope.)

    I understand that the topic is fossils, not rocks, but am vitally interested in their location, nonetheless!
    Given that the exposed surface of WA is the most expansive virgin geological topology that we have uncovered to date, this find is hardly surprising.
    Who knows what lieth under the ice of Antarctica?!

    [1] A single “farm” that is probably the size of New Jersey, if not larger.
    [2] The site of the (previously?) oldest mineral (zircon) found unchanged on Earth.

      1. I thank thee, Prof. Coyne for your reply.
        Your info, combined with the answer below, have solved my geographical coynunudrum.
        What should I be looking for on my next visit?
        Or: can you refer me to one who may be able to answer such an enquiry, should you feel less than comfortable proferring advice that might lie outside of your field of expertise?

      1. Ta kindly!
        It is not quite on my sister’s station/farm/ranch, but close enough for me to visit the location on my next trip.
        Not in the summer months, if I can help it!
        She tells me that it gets “a bit warm” there.

        1. Rather you than me ! Not even for a free, lifetime supply of slabs (slabbies?) of the amber nectar

          This is my idea of Hell: Marble Bar

          has an arid climate with very hot summers and mild to warm winters. The town set a world record of most consecutive days of maximum temperatures of 37.8 degrees Celsius (100 degrees Fahrenheit) or more, during a period of 160 such days from 31 October 1923 to 7 April 1924. During December and January, temperatures in excess of 45 degrees Celsius (113 degrees Fahrenheit) are common, and the average maximum temperature exceeds normal human body temperature for 6 months each year. Rainfall mostly occurs in the summer months

          This is an interesting read on the Pilbara Craton

          1. Well, for what it’s worth, we’re expecting it to finally cool down to 107 this weekend here in Arizona, and the overnight low might make it below 90 by then, too….



              1. Well, I never turn on the radio (there’s nothing left worth listening to), and the heat keeps the door-to-door Jesus salescritters at bay, so it’s not all that bad….


              2. Except that 2 of the 3 times I’ve been in Phoenix, all the locals were complaining about the rare humidity. As far as I’m concerned, it’s usually humid there.

              3. This time of year, we usually alternate between hundred-teens with humidity in the teens and upper hundreds with humidity in the 40-50% range. The latter usually translates to a 30% chance of an evening thunderstorm…but 70% of that 30% of the time, you just get blowing dust and / or mud. What’s really bad is when you get hundred-teens with 40% humidity, or hundred-twenties with any kind of humidity; fortunately, both are rare.

                As far as seasons go…well, we’re in the tail end (I hope!) of Hell right now. In a few weeks, summer will hopefully start and last until October. Then we’ll have a very pleasant fall, a day or two of winter, and a lovely spring before summer starts again in March. Hell (hopefully) won’t be back again until late June, and both summers are usually pretty nice, so it’s not all that much to complain about. I mean, it’s not like we spend eight months of the year shoveling snow out of driveways….



    1. Was there ever really any much dispute over the at least the promising potential/possibility for pyrite to serve as an important inorganic catalyst for abiogenesis, seeing as iron and sulfur were, after all, bog common in most of the likely locales hypothesized to be the site of abiogenesis?

  4. “And perhaps the earliest cellular life metabolized sulfur rather than oxygen.”

    Whether or not sulfur was the first metabolic fuel, it’s certain that oxygen was not. Oxygen is too reactive to remain free without a constantly replenished supply, which was available only after photosynthesis evolved.

    1. Not entirely true. A small amount of molecular oxygen would have been created by UV splitting water right from the very beginning. (No ozone to block UV prior to the buildup of molecular oxygen in the atmosphere) This oxygen would have reacted almost as soon as it formed with reduced chemicals dissolved in the water, but the very top layer of the surface of the oceans would have had a very low, but non-zero, constant level of oxygen because of that.

      1. Which might drive some microbes to develop a defense against it, but is it enough to develop a metabolism based on it? I’d guess no, but I’m fine with being wrong about that.

  5. demonstrating that prokaryotic organisms had already evolved only a billion years after Earth had formed.

    And only about 400 million years after the end of the Hadean, give or take. What is the current thinking about life in the Hadean? Was it even possible?

    If conditions for life were only possible after the end of the Hadean, that means life formed pretty damned quickly once it had the chance. Like, pretty much right away. So perhaps life isn’t so unlikely an occurence? I’d be interested in hearing opinions.

    1. I like Panspermia & good old Fred Hoyle. Even today after all this time hyperbolic comets pass ONCE through the inner Solar System before being thrown out into interstellar space. I reckon that solar systems are constantly exchanging material between themselves & the dust clouds from which they form. Maybe not ‘spores from space’ so much as amino acids. I would love us to send up some huge sticky nets & check out what’s up there ~ maybe interstellar space is one big ocean. Larry Niven & all that…

    2. I don’t think abiogenesis during the Hadean (at least the late Hadean, when at least some standing pools of liquid water existed) would have been a problem. The biggest problem with life in the Hadean was more survival through it, what with the various Heavy Bombardments going on. The idea being that there could have been multiple abiogenesis events, but only the last one, after the bombardments were over, managed to “take”.

      But if there was some refuge where early life could colonize and be sheltered from bombardments producing whole planet vaporization events, then it’s quite possible for life to have emerged in the Hadean, and then go on to spread during the Archaean.

      1. FWIW this late, later models predicts that prokaryotes proliferate and spread, in the crust, faster than any reasonable HB could sterilize it. See Abramov et al (ref in the follow up post, if it clears the acceptance queue).

  6. I’m surprised that no one has noted that we Australians not only invented sex (well, actually internal fertilization), but also invented life itself …

  7. Space in Nature is in pretty high demand; perhaps the authors wanted more room to present their evidence than the premier magazine could give them.

  8. As for those folks talking about an archaen/ bacterial LUCA…

    …it really doesn’t work that way at that level. At the level of single cells, there is non-reproductive genome exchange happening through plasmids and other mechanisms that make relationships look more like a tangled, circular creosote than anything so arboreal as to contain something that could be called a LUCA.

    1. I think several related methods such as super-matrix methods, that simultaneously keeps track of genome state changes during vertical and horizontal descent, shows that there is a reasonable phylogenetic signal. And it is dominated by vertical descent.

      Besides, the prediction of a genetic machinery LUCA is one of the best tested in all of science (1:10^-2040 risk of multiple ancestors). Whether that ancestor was a tangled or plain vanilla LUCA seems more like a philosophic question. It was a phylogenetic signal ancestor.

Leave a Reply to ThannyCancel reply